[go: up one dir, main page]

CN103631803A - Method, device and server for advertisement orientation based on input behaviors - Google Patents

Method, device and server for advertisement orientation based on input behaviors Download PDF

Info

Publication number
CN103631803A
CN103631803A CN201210304052.9A CN201210304052A CN103631803A CN 103631803 A CN103631803 A CN 103631803A CN 201210304052 A CN201210304052 A CN 201210304052A CN 103631803 A CN103631803 A CN 103631803A
Authority
CN
China
Prior art keywords
user
advertisement
keyword
point
interest
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN201210304052.9A
Other languages
Chinese (zh)
Inventor
何径舟
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Baidu International Technology Shenzhen Co Ltd
Original Assignee
Baidu International Technology Shenzhen Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Baidu International Technology Shenzhen Co Ltd filed Critical Baidu International Technology Shenzhen Co Ltd
Priority to CN201210304052.9A priority Critical patent/CN103631803A/en
Publication of CN103631803A publication Critical patent/CN103631803A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0241—Advertisements
    • G06Q30/0251—Targeted advertisements
    • G06Q30/0255—Targeted advertisements based on user history
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/30—Semantic analysis

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Strategic Management (AREA)
  • Finance (AREA)
  • Development Economics (AREA)
  • Accounting & Taxation (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • General Health & Medical Sciences (AREA)
  • General Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Computational Linguistics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Game Theory and Decision Science (AREA)
  • Artificial Intelligence (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • General Business, Economics & Management (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Information Transfer Between Computers (AREA)

Abstract

The invention discloses a method, device and server for advertisement orientation based on input behaviors. The method comprises the following steps: at least one of user key words and user interest points is acquired from acquired user input historical text information; at least one of the user key words and the user interest points is sent to the server through a network; an advertisement sent by the server is received and displayed at a preset position, wherein the advertisement is acquired in the mode that at least one of the user key words and the user interest points is matched by the server. By means of the method, device and server for advertisement orientation based on the input behaviors, advertisements with more accurate positioning can be sent to users quickly.

Description

Based on input behavior carry out advertisement fixing to method, device and server
Technical field
The present invention relates to input method field, particularly relate to a kind of based on input behavior carry out advertisement fixing to method, device and server.
Background technology
Input method application is according to certain coding rule, to carry out the program of input characters, and user generally need to apply word input by specific input method when using computing machine.
While utilizing input method application to carry out word input, generally can be historical in backstage generation input.In prior art, for the application of user's input history, be mainly the aspects such as language model performance of collecting language material, optimizing input method application or other Related products.
Present inventor finds in R&D process, and prior art is used the application of input history of input method insufficient to user, not have further the fully input of digging user historical, wastes resource.
Summary of the invention
The technical matters that the present invention mainly solves be to provide a kind of based on input behavior carry out advertisement fixing to method, device and server, can to user, send location advertisement more accurately rapidly.
For solving the problems of the technologies described above, the technical scheme that the present invention adopts is: provide a kind of based on input behavior carry out advertisement fixing to method, comprising: from the user who obtains, input and history text information, obtain at least one in user's keyword, user interest point; By network, to server, send at least one in described user's keyword, user interest point; Receive the advertisement that described server sends, and in precalculated position, represent described advertisement, described advertisement in server match user keyword, user interest point at least one and obtain.
Wherein, the described user from acquisition inputs the step that obtains user's keyword history text information, comprising: the user of acquisition is inputted to history text information and mate with the keyword set prestoring, obtain described user's keyword.
Wherein, the described user by acquisition inputs history text information mates with the keyword set prestoring, after obtaining the step of user's keyword, also comprise: according to described user's keyword, user, input the frequency and/or the time that in history text information, occur and add up, obtain the degree of confidence of described user's keyword, wherein, user, to input the frequency and/or the time that in history text information, occur be positively related relation for the degree of confidence of described user's keyword and described user's keyword; According to the degree of confidence of described user's keyword, user's keyword is sorted, obtain user's lists of keywords, described user's lists of keywords comprises each user's keyword and degree of confidence thereof.
Wherein, the described user from acquisition inputs the step that obtains user interest point history text information, comprise: adopt topic model, the user of described acquisition is inputted to history text information integral body and carry out theme statistics, obtain described user interest point, wherein, each theme is a user interest point.
Wherein, described employing topic model, the user of described acquisition is inputted to the step that history text information integral body is carried out theme statistics, comprise: adopt probability latent semantic analysis model (PLSA) or the Di Li Cray assignment-type model (LDA) of hiding, the user of described acquisition is inputted to the statistics that history text information integral body is carried out the degree of confidence of theme statistics and each theme, obtain user interest point list, described user interest point list comprises each user interest point and degree of confidence thereof.
Wherein, the described user from acquisition inputs the step that obtains user interest point history text information, comprise: according to the theme distributed intelligence of each entry prestoring, the user who adds up described acquisition input history text information the theme distributed intelligence of corresponding entry, obtain user interest point list, described user interest point list comprises each user interest point and weight thereof, and wherein, each theme is a user interest point.
Wherein, describedly by network, to server, send at least one the step in described user's keyword, user interest point, comprising: by network, to server, send at least one in user's lists of keywords, user interest point list.
Wherein, the described step that represents described advertisement in precalculated position, comprise: the precalculated position outside the Zhong Huo interface, interface of input method application represents described advertisement, described precalculated position is candidate frame, status bar or applies the candidate's who is associated expansion explanation frame with input method; Or the precalculated position of the interface phase association of applying third party represents described advertisement; Or the showing advertisement position providing in third party's application represents described advertisement.
Wherein, the advertisement that the described server of described reception sends, and in precalculated position, represent the step of described advertisement, comprising: receive the advertisement that described server sends, and in the precalculated position of predetermined scene, represent the advertisement of mating with described predetermined scene.
Wherein, before described at least one the step obtaining from the user who obtains inputs history text information in user's keyword, user interest point, comprising: the input historical data to user is carried out pre-service, obtain described user and input history text information.
Wherein, the described input historical data to user is carried out pre-service, obtain the step that described user inputs history text information, comprise: the form of inputting historical data according to user, input historical data to user is screened, is spliced, punctuate and impurity elimination, obtains described user and inputs history text information.
For solving the problems of the technologies described above, another technical solution used in the present invention is: provide a kind of based on input behavior carry out advertisement fixing to method, comprising: by network, receives user's keyword of input method application transmission, at least one in user interest point; Obtain and the user's keyword having received, at least one advertisement prestoring matching in user interest point; By network, to described input method application, send the advertisement that coupling obtains.
Wherein, described acquisition and the user's keyword having received, the step of at least one advertisement prestoring matching in user interest point, comprise: from the advertisement keyword index prestoring, search the advertisement keyword matching with the described user's keyword having received, according to the described advertisement keyword finding, from advertisement base, search and obtain described advertisement, or, by received each described in user interest point carry out respectively the calculating of point of interest similarity with a plurality of advertisement points of interest of prestoring, the matching score that obtains the point of interest similarity between the described user interest point having received and the advertisement point of interest prestoring, the corresponding advertisement of advertisement point of interest of the matching score first predetermined condition of described point of interest similarity is the advertisement that coupling obtains, or, from the advertisement keyword index prestoring, search the advertisement keyword matching with the described user's keyword having received, according to the described advertisement keyword finding, from advertisement base, search and obtain described advertisement, described advertisement is the advertisement of the coupling obtaining according to described user's keyword, by received each described in user interest point carry out respectively the calculating of point of interest similarity with a plurality of advertisement points of interest of prestoring, obtain the matching score of the point of interest similarity between the user interest point having received and the advertisement point of interest prestoring, obtain the corresponding advertisement of advertisement point of interest of the matching score first predetermined condition of point of interest similarity, described advertisement is the advertisement of the coupling obtaining according to described user interest point, in the advertisement of the advertisement of the described coupling obtaining according to user's keyword and the coupling that obtains according to user interest point, identical advertisement is the advertisement that coupling obtains.
Wherein, describedly by network, receive user's keyword that input method application sends, at least one step in user interest point, comprising: by network, receives user's lists of keywords of input method application transmission, at least one in user interest point list.
Wherein, described acquisition and the user's keyword having received, the step of at least one advertisement prestoring matching in user interest point, comprising: obtain and the user's lists of keywords having received, at least one advertisement prestoring matching in user interest point list.
Wherein, the step of the advertisement prestoring that described acquisition and the user's lists of keywords having received match, comprising: according to the user's lists of keywords receiving, according to degree of confidence order from high to low, determine user's keyword of degree of confidence second predetermined condition; From the advertisement keyword index prestoring, search the advertisement keyword that the user's keyword with degree of confidence second predetermined condition matches, according to the described advertisement keyword finding, from advertisement base, search and obtain described advertisement.
Wherein, the step of the advertisement prestoring that described acquisition and the user interest point list having received match, comprising: according to the user interest point list receiving, according to degree of confidence order from high to low, determine that degree of confidence meets the user interest point of the 3rd predetermined condition; The user interest point that described degree of confidence is met to the 3rd predetermined condition carries out respectively the calculating of point of interest similarity with a plurality of advertisement points of interest that prestore, the matching score of the point of interest similarity between the advertisement point of interest that obtains the user interest point of satisfied the 3rd predetermined condition of degree of confidence and prestore, the corresponding advertisement of advertisement point of interest of the matching score first predetermined condition of described point of interest similarity is the advertisement that coupling obtains.
Wherein, the step of the advertisement prestoring that described acquisition and the user's lists of keywords having received and user interest point list match, comprise: according to the user's lists of keywords and the user interest point list that receive, according to degree of confidence order from high to low, determine that degree of confidence meets user's keyword and the user interest point of the 4th predetermined condition, from the advertisement keyword index prestoring, search with degree of confidence and meet the advertisement keyword that user's keyword of the 4th predetermined condition matches, according to the described advertisement keyword finding, from advertisement base, search and obtain described advertisement, described advertisement is the advertisement of the coupling obtaining according to described user's keyword, the user interest point that described degree of confidence is met to the 4th predetermined condition carries out respectively the calculating of point of interest similarity with a plurality of advertisement points of interest that prestore, the matching score of the point of interest similarity between the advertisement point of interest that obtains the user interest point of satisfied the 4th predetermined condition of degree of confidence and prestore, obtain the corresponding advertisement of advertisement point of interest of the matching score first predetermined condition of point of interest similarity, described advertisement is the advertisement of the coupling obtaining according to described user interest point, in the advertisement of the advertisement of the described coupling obtaining according to user's keyword and the coupling that obtains according to user interest point, identical advertisement is the advertisement that coupling obtains.
For solving the problems of the technologies described above, another technical scheme that the present invention adopts is: provide a kind of based on input behavior carry out advertisement fixing to method, comprising: by network, to server, send the user who obtains and input history text information; Receive the advertisement that described server sends, and in precalculated position, represent described advertisement, described advertisement is inputted history text information by server according to user and is mated and obtain.
For solving the problems of the technologies described above, another technical scheme that the present invention adopts is: provide a kind of based on input behavior carry out advertisement fixing to method, comprising: the user who receives input method application transmission by network inputs history text information; From the user who has received, input and history text information, obtain at least one in user's keyword, user interest point; At least one advertisement prestoring matching in acquisition and described user's keyword, user interest point; By network, to described input method application, send the advertisement that coupling obtains.
For solving the problems of the technologies described above, another technical scheme that the present invention adopts is: provide a kind of based on input behavior carry out advertisement fixing to device, described device comprises: first obtains module, inputs history text information obtain at least one user's keyword, user interest point for the user from obtaining; Sending module, for sending at least one of described user's keyword, user interest point to server by network; Receiver module, the advertisement sending for receiving described server, and represent described advertisement in precalculated position, described advertisement in server match user keyword, user interest point at least one and obtain.
Wherein, described first obtains module comprises: keyword obtains unit, and described keyword obtains unit and mates with the keyword set prestoring for the user of acquisition is inputted to history text information, obtains described user's keyword.
Wherein, described first obtains module also comprises: statistic unit, for inputting user the frequency and/or the time that history text information occurs according to described user's keyword, add up, obtain the degree of confidence of described user's keyword, wherein, user, to input the frequency and/or the time that in history text information, occur be positively related relation for the degree of confidence of described user's keyword and described user's keyword; Lists of keywords obtains unit, for according to the degree of confidence of described user's keyword, user's keyword is sorted, and obtains user's lists of keywords, and described user's lists of keywords comprises each user's keyword and degree of confidence thereof.
Wherein, described first obtains module comprises: the first point of interest obtains unit, described the first point of interest obtains unit and is used for adopting topic model, the user of described acquisition is inputted to history text information integral body and carry out theme statistics, obtain described user interest point, wherein, each theme is a user interest point.
Wherein, described point of interest obtains unit specifically for adopting probability latent semantic analysis model (PLSA) or the Di Li Cray assignment-type model (LDA) of hiding, the user of described acquisition is inputted to the statistics that history text information integral body is carried out the degree of confidence of theme statistics and each theme, obtain user interest point list, described user interest point list comprises each user interest point and degree of confidence thereof.
Wherein, described first obtains module comprises: the second point of interest obtains unit, described the second point of interest obtains unit and is used for according to the theme distributed intelligence of each entry prestoring, the user who adds up described acquisition input history text information the theme distributed intelligence of corresponding entry, obtain user interest point list, described user interest point list comprises each user interest point and weight thereof, and wherein, each theme is a user interest point.
Wherein, described sending module is specifically for sending at least one in user's lists of keywords, user interest point list by network to server.
Wherein, described receiver module comprises: represent unit, the described unit that represents comprises: first represents subelement, for the precalculated position outside the Zhong Huo interface, interface in input method application, represent described advertisement, described precalculated position is candidate frame, status bar or applies the candidate's who is associated expansion explanation frame with input method; Or second represents subelement, for the precalculated position of the interface phase association in third party application, represent described advertisement; Or the 3rd represents subelement, for the showing advertisement position providing in third party's application, represent described advertisement.
Wherein, the advertisement that described receiver module sends specifically for receiving described server, and in the precalculated position of predetermined scene, represent the advertisement of mating with described predetermined scene.
Wherein, described device also comprises: second obtains module, and described second obtains module for user's input historical data is carried out to pre-service, obtains described user and inputs history text information.
Wherein, described second obtains module specifically for input the form of historical data according to user, to user's input historical data screen, splice, punctuate and impurity elimination, obtain described user and input history text information.
For solving the problems of the technologies described above, another technical scheme that the present invention adopts is: a kind of server is provided, and described server comprises: receiver module, for receive user's keyword of input method application transmission, at least one of user interest point by network; Obtain module, for the user's keyword that obtains and received, at least one advertisement prestoring matching of user interest point; Sending module, for sending to described input method application the advertisement that coupling obtains by network.
Wherein, described acquisition module comprises: the first matching unit, for searching from the advertisement keyword index prestoring the advertisement keyword matching with the described user's keyword having received, according to the described advertisement keyword finding, from advertisement base, search and obtain described advertisement, or, the second matching unit, for by received each described in user interest point carry out respectively the calculating of point of interest similarity with a plurality of advertisement points of interest of prestoring, the matching score that obtains the point of interest similarity between the user interest point having received and the advertisement point of interest prestoring, the corresponding advertisement of advertisement point of interest of the matching score first predetermined condition of described point of interest similarity is the advertisement that coupling obtains, or, the 3rd matching unit, for searching from the advertisement keyword index prestoring the advertisement keyword matching with the described user's keyword having received, according to the described advertisement keyword finding, from advertisement base, search and obtain described advertisement, described advertisement is the advertisement of the coupling obtaining according to described user's keyword, by received each described in user interest point carry out respectively the calculating of point of interest similarity with a plurality of advertisement points of interest of prestoring, obtain the matching score of the point of interest similarity between the user interest point having received and the advertisement point of interest prestoring, obtain the corresponding advertisement of advertisement point of interest of the matching score first predetermined condition of point of interest similarity, described advertisement is the advertisement of the coupling obtaining according to described user interest point, in the advertisement of the advertisement of the described coupling obtaining according to user's keyword and the coupling that obtains according to user interest point, identical advertisement is the advertisement that coupling obtains.
For solving the problems of the technologies described above, another technical scheme that the present invention adopts is: provide a kind of based on input behavior carry out advertisement fixing to device, described device comprises: sending module, for sending to server the user who obtains by network, input history text information; Receiver module, the advertisement sending for receiving described server, and represent described advertisement in precalculated position, described advertisement is inputted history text information by server according to user and is mated and obtain.
For solving the problems of the technologies described above, another technical scheme that the present invention adopts is: a kind of server is provided, and described server comprises: receiver module, for receive the user of input method application transmission by network, input history text information; First obtains module, inputs history text information obtain at least one user's keyword, user interest point for the user from having received; Second obtains module, for obtaining at least one advertisement prestoring matching with described user's keyword, user interest point; Sending module, for sending to described input method application the advertisement that coupling obtains by network.
The invention has the beneficial effects as follows: the situation that is different from prior art, the present invention inputs at least one history text information acquisition user keyword, user interest point from the user who obtains, utilize described user's keyword or user interest point to carry out match advertisements, and at precalculated position showing advertisements.By the way, can input user's keyword and/or the user interest point excavating in history text information according to user, to user, push rapidly location advertisement more accurately, thereby further make full use of user's the historical resource of input.
Accompanying drawing explanation
Fig. 1 be the present invention is based on input behavior carry out advertisement fixing to the process flow diagram of method one embodiment;
Fig. 2 be the present invention is based on input behavior carry out advertisement fixing to the process flow diagram of another embodiment of method;
Fig. 3 be the present invention is based on input behavior carry out advertisement fixing to the process flow diagram of the another embodiment of method;
Fig. 4 be the present invention is based on input behavior carry out advertisement fixing to the process flow diagram of the another embodiment of method;
Fig. 5 be the present invention is based on input behavior carry out advertisement fixing to the process flow diagram of the another embodiment of method;
Fig. 6 be the present invention is based on input behavior carry out advertisement fixing to the structural representation of device one embodiment;
Fig. 7 be the present invention is based on input behavior carry out advertisement fixing to the structural representation of another embodiment of device;
Fig. 8 be the present invention is based on input behavior carry out advertisement fixing to the structural representation of the another embodiment of device;
Fig. 9 be the present invention is based on input behavior carry out advertisement fixing to the structural representation of the another embodiment of device;
Figure 10 be the present invention is based on input behavior carry out advertisement fixing to the structural representation of the another embodiment of device;
Figure 11 be the present invention is based on input behavior carry out advertisement fixing to the structural representation of the another embodiment of device;
Figure 12 is the structural representation of server one embodiment of the present invention;
Figure 13 is the structural representation of another embodiment of server of the present invention;
Figure 14 is the structural representation of the another embodiment of server of the present invention;
Figure 15 is the structural representation of the another embodiment of server of the present invention;
Figure 16 is the structural representation of the another embodiment of server of the present invention.
Embodiment
Below in conjunction with drawings and embodiments, the present invention is described in detail.
Consult Fig. 1, Fig. 1 be the present invention is based on input behavior carry out advertisement fixing to the process flow diagram of method one embodiment, present embodiment is the process flow diagram of input method application end, comprising:
Step S11: input and obtain at least one in user's keyword, user interest point history text information from the user who obtains;
When user uses input method application, user's input storage of history data P is in this locality, and these input historical datas are extremely useful resources, can fully excavate and utilize these users' input historical data resource.
First, from user's input historical data resource, obtain user and input history text information, then, from the user who obtains, input history text information and obtain user's keyword or user interest point or user's keyword and user interest point, be beneficial to follow-up coupling.
Step S12: send at least one in described user's keyword, user interest point to server by network;
If step S11 only obtains user's keyword, step S12 sends user's keyword to server; If step S11 only obtains user interest point, step S12 sends user interest point to server; If step S11 obtains user's keyword and user interest point, step S12 sends user's keyword and user interest point to server.
Step S13: receive the advertisement that described server sends, and in precalculated position, represent described advertisement, described advertisement in server match user keyword, user interest point at least one and obtain.
At step S12, to server, send after at least one in described user's keyword, user interest point, server mates according to the described user's keyword receiving, at least one in user interest point, obtain the advertisement of coupling, and the advertisement of described coupling is sent to input method application end, input method application end receives the advertisement that described server sends, and in precalculated position, represents described advertisement.
The present invention inputs at least one history text information acquisition user keyword, user interest point from the user who obtains, and utilizes described user's keyword or user interest point to carry out match advertisements, and at precalculated position showing advertisements.By the way, can input user's keyword and/or the user interest point excavating in history text information according to user, to user, push rapidly location advertisement more accurately, thereby further make full use of user's the historical resource of input.
Consult Fig. 2, Fig. 2 be the present invention is based on input behavior carry out advertisement fixing to the process flow diagram of another embodiment of method, present embodiment is the process flow diagram of input method application end, comprising:
Step S101: the input historical data to user is carried out pre-service, obtains user and inputs history text information.
When user uses input method application, user's input storage of history data P is in this locality, and these input historical datas are extremely useful resources, can fully excavate and utilize these users' input historical data resource.Certainly, original user's input historical data is conventionally more scattered, and content is also very assorted, therefore, need to carry out pre-service to user's input historical data, to obtain relatively high-quality user, inputs history text information.
Step S101 specifically comprises: according to user, input the form of historical data, to user's input historical data screen, splice, punctuate and impurity elimination, obtain user and input history text information.
Pretreated operation includes but not limited to: screening, splicing, punctuate and impurity elimination, according to user, input the form of historical data, for example has plenty of complete sentence, has plenty of incomplete sentence, some a words have occurred many times, the just word having, or a word, or the words without any meaning is pieced together, to these users' input historical data screen, splice, punctuate and impurity elimination etc.
Step S102: input and obtain at least one in user's keyword, user interest point history text information from the user who obtains.
Can input history text information and obtain user's keyword or user interest point or user's keyword and user interest point from the user who obtains, be beneficial to follow-up coupling.
When the user from obtaining inputs while obtaining user's keyword history text information, comprising:
102a1: the user of acquisition is inputted to history text information and mate with the keyword set prestoring, obtain user's keyword.
Keyword set comprises a series of entries with commercial value, for example: automobile, fresh flower, mobile phone, camera etc.
For example, user inputs history text information and is: iphone4s, and the keyword mating in keyword set is so: mobile phone; And for example, user inputs history text information and is: benz, BMW etc., and the keyword mating in keyword set is so: automobile; User inputs history text information: rose, carnation, lily etc., the keyword mating in keyword set is so: fresh flower.
The user's keyword obtaining, can reflect focus and the demand point of user's interest, hobby or certain period.
Further, after 102a1, also comprise:
102a2: input user the frequency and/or the time that occur in history text information according to described user's keyword and add up, obtain the degree of confidence of described user's keyword, wherein, user, to input the frequency and/or the time that in history text information, occur be positively related relation for the degree of confidence of described user's keyword and described user's keyword.
It is positively related relation that the degree of confidence of user's keyword and user's keyword input user the frequency and/or the time that in history text information, occur, for example, it is higher that user's keyword inputs user the frequency occurring in history text information, and the degree of confidence of user's keyword is higher; It is nearer that user's keyword inputs user the time occurring in history text information, and the degree of confidence of user's keyword is higher.
102a3: according to the degree of confidence of described user's keyword, user's keyword is sorted, obtain user's lists of keywords, described user's lists of keywords comprises each user's keyword and degree of confidence thereof.
According to the degree of confidence of user's keyword, after user's keyword is sorted, obtain the user's lists of keywords after sequence, this user's lists of keywords also comprises each user's keyword and degree of confidence thereof.
In actual applications, input method application can select user's keyword that degree of confidence is the highest to send to server; Also user's lists of keywords can be sent to server, by server, according to the degree of confidence of user's keyword, select user's keyword; Or the user's keyword only 102a1 being obtained sends to server.That is to say, after 102a1 is finished, can select to carry out 102a2 and 102a3.
When the user from obtaining inputs while obtaining user interest point history text information, comprising:
102b: adopt topic model, the user of described acquisition is inputted to history text information integral body and carry out theme statistics, obtain described user interest point, wherein, each theme is a user interest point.
Topic model refers to the angle from statistical model, with a specific word frequency distribution, portrays theme, it has been generally acknowledged that one piece of article, one section of word, a sentence generate from a probability model.
User interest point can be concrete theme, such as " business ", " pet ", " amusement " so concrete theme, can be only also the non-Sexual Themes that refers in particular to after cluster.Owing to automatically generating by topic model, the non-Sexual Themes that refers in particular to probably cannot Direct function description, is all to distinguish by sign (ID).During actual use, often for example, by refering in particular to main support word contained in Sexual Themes and described non-:
Theme 567: bean vermicelli, whenever and wherever possible, microblogging, Sina, fresh,
Theme 83: Sohu, record, adjust, microblogging, solo.
Wherein, 102b specifically comprises: adopt probability latent semantic analysis model (PLSA) or the Di Li Cray assignment-type model (LDA) of hiding, the user of described acquisition is inputted to the statistics that history text information integral body is carried out the degree of confidence of theme statistics and each theme, obtain user interest point list, described user interest point list comprises each user interest point and degree of confidence thereof.
Probability latent semantic analysis model (PLSA) is the classical statistical models that the data analysing method based on double mode and co-occurrence extends.PLSA is applied to machine learning or other association area of information retrieval, filtration, natural language processing, text.The difference of probability latent semantic analysis and standard latent semantic analysis is, standard latent semantic analysis is with the form performance of the svd of co-occurrence (being the matrix of co-occurrence), and the hybrid matrix that probability latent semantic analysis is lowest common multiple (LCM) matrix based on deriving from decomposes, the mixing based on multinomial distribution and condition distribution carrys out the probability of modeling co-occurrence.
The Di Li Cray assignment-type model (LDA) of hiding is a kind of non-supervisory machine learning techniques, can be used for identifying subject information hiding in extensive document sets or corpus.It has adopted the method for word bag, and this method is considered as a word frequency vector by each piece of document, thereby text message is converted into the numerical information that is easy to modeling.Each piece of document represented the probability distribution that some themes form, and each theme has represented the probability distribution that a lot of words form.
In actual applications, input method application can select the user interest point that degree of confidence is the highest to send to server; Also user interest point list can be sent to server, by server, according to the degree of confidence of user interest point, select user interest point; Or the user interest point only 102b being obtained sends to server.
When the user from obtaining inputs while obtaining user interest point history text information, also comprise:
102c: according to the theme distributed intelligence of each entry prestoring, the user who adds up described acquisition input history text information the theme distributed intelligence of corresponding entry, obtain user interest point list, described user interest point list comprises each user interest point and weight thereof, wherein, each theme is a user interest point.
For example: in the dictionary of an input method, the theme distributed intelligence of the entry prestoring is as table 1:
Figure BDA00002048818900131
Figure BDA00002048818900141
The theme distributed intelligence of the entry that table 1 prestores
The form of dictionary is " entry: theme distributes and weight ", if entry does not contain interested subject information, " theme distributes and weight " is 0.
If user's input history text information only comprises two sentences:
" I often remove Baidu's barbecue ", and
The information of “Wo Baidu apple ",
The entry that comprises subject information in this user's input history text information, comprising:
2 Baidu, 1 barbecue, 1 apple;
According to the theme distributed intelligence of entry, COMPREHENSIVE CALCULATING obtains theme distribution and weight is respectively:
Science and technology: 0.8*2+0.6=2.2,
Cuisines: 0.2*2+1+0.3=1.7,
Video display: 0.1.
This information can upload onto the server, and for server, carries out user interest point coupling and advertisement pushing.
Wherein, the theme distributed intelligence of entry, when setting up dictionary, can adopt the topic model statistics such as PLSA or LDA to obtain; But in dictionary, only store final numerical value, avoided like this carrying out complicated field statistics on subscriber set.
In addition, the method that the theme of the concrete numerical value of weight and counting user input history text information distributes, fairly simple in example, adopt direct average weighted method, in actual applications, and in order to promote effect, method that can design comparison complexity.
102a, 102b and 102c can select an execution, also can carry out 102a and 102b, or carry out 102a and 102c.
Step S103: send at least one in described user's keyword, user interest point to server by network;
If step S102 only obtains user's keyword, step S103 sends user's keyword to server; If step S102 only obtains user interest point, step S103 sends user interest point to server; If step S102 obtains user's keyword and user interest point, step S103 sends user's keyword and user interest point to server.
Wherein, step S103 specifically comprises: by network, to server, send at least one in user's lists of keywords, user interest point list.
Step S104: receive the advertisement that described server sends, and in precalculated position, represent described advertisement, described advertisement in server match user keyword, user interest point at least one and obtain.
Wherein, in precalculated position, represent the step of described advertisement, comprising:
Precalculated position outside the Zhong Huo interface, interface of input method application represents described advertisement, and described precalculated position is candidate frame, status bar or applies the candidate's who is associated expansion explanation frame with input method;
Or the precalculated position of the interface phase association of applying third party represents described advertisement;
Or the showing advertisement position providing in third party's application represents described advertisement.
Wherein, step S104 specifically comprises: receive the advertisement that described server sends, and in the precalculated position of predetermined scene, represent the advertisement of mating with described predetermined scene.
User's input history can record input and call scene, and such as being that input when the browsing page is historical, or the input when writing mail is historical, or the input history when playing games etc.So, during browsing page, input history is all cosmetics, fashion brand etc., can in browser, represent the advertisement with this respect of scene matching; While writing mail, represent the commercial product advertisement with scene matching; While playing games, user's input history is all game term, represents with other game advertisement of scene matching etc.In this way, these scene informations are collected, contribute to represent more accurate advertisement.
In a word, the present invention inputs at least one history text information acquisition user keyword, user interest point from the user who obtains, and utilizes described user's keyword or user interest point to carry out match advertisements, and at precalculated position showing advertisements.By the way, can input user's keyword and/or the user interest point excavating in history text information according to user, to user, push rapidly location advertisement more accurately, thereby further make full use of user's the historical resource of input.
In addition, the present invention can fully excavate and utilize user to input the commercial value of history text information, targetedly to information such as user's advertisement, news, blogs, not only can meet consumers' demand better, and can set up according to this good upstream industry chain, and then form independently business model.
Although user behavior be collected and be utilized to the consumer products based on web service such as search engine, microblogging also can, the industry propelling movement of doing business of going forward side by side,, the business propelling movement of inputting history text information based on user has two advantages:
(1) popularity, user uses web service to have purpose, and its behavior is affected by this also may, and therefore, user uses the behavior of web service to differ to reflect surely all interest characteristicses of user.Input history text information is applied to user's various use scenes, can collect more widely user's point of interest and carry out more effectively comprehensively pushing;
(2) privacy, user does not generally wish that its behavior is collected and monitors, and input history text information can be stored in subscriber's local, carries out local modeling and extraction, what final upload server mated is only some keywords and point of interest, can not reveal user's out of Memory.Certainly, the in the situation that of subscriber authorisation, input history text information also can upload onto the server and carry out more accurate and detailed analysis.
Consult Fig. 3, Fig. 3 be the present invention is based on input behavior carry out advertisement fixing to the process flow diagram of the another embodiment of method, present embodiment is the process flow diagram of server end, comprising:
Step S201: receive user's keyword of input method application transmission, at least one in user interest point by network;
If input method application sends user's keyword, step S201 receives user's keyword; If input method application sends user interest point, step S201 receives user interest point; If input method application sends user's keyword and user interest point, step S201 receives user's keyword and user interest point.
Wherein, step S201 specifically comprises: by network, receive user's lists of keywords of input method application transmission, at least one in user interest point list.
If input method application sends user's lists of keywords, step S201 receives user's lists of keywords; If input method application sends user interest point list, step S201 receives user interest point list; If input method application sends user's lists of keywords and user interest point list, step S201 receives user's lists of keywords and user interest point list.
Step S202: obtain and the user's keyword having received, at least one advertisement prestoring matching in user interest point;
Wherein, step S202 specifically comprises:
202a: search the advertisement keyword matching with the user's keyword having received from the advertisement keyword index prestoring, according to the advertisement keyword finding, search acquisition advertisement from advertisement base;
In server, preserve in advance advertisement keyword index, the user's keyword having received is mated with the advertisement keyword in the advertisement keyword index prestoring, according to the advertisement keyword in the advertisement keyword index of coupling, from advertisement base, search, can obtain the mate advertisement obtaining.
202b: by received each described in user interest point carry out respectively the calculating of point of interest similarity with a plurality of advertisement points of interest of prestoring, the matching score that obtains the point of interest similarity between the user interest point having received and the advertisement point of interest prestoring, the corresponding advertisement of advertisement point of interest of the matching score first predetermined condition of described point of interest similarity is the advertisement that coupling obtains;
For example, the first predetermined condition: the matching score of point of interest similarity is the highest, now, the corresponding advertisement of advertisement point of interest that the matching score of point of interest similarity is the highest is the advertisement that coupling obtains.
Ad content also can adopt the mode of the topic models such as PLSA or LDA to obtain advertisement point of interest and degree of confidence, the advertisement point of interest of acquisition is kept in server, so that carry out the calculating of point of interest similarity with the user interest point receiving.
202c: search the advertisement keyword matching with the user's keyword having received from the advertisement keyword index prestoring, according to the advertisement keyword finding, from advertisement base, search acquisition advertisement, described advertisement is the advertisement of the coupling obtaining according to described user's keyword, by received each described in user interest point carry out respectively the calculating of point of interest similarity with a plurality of advertisement points of interest of prestoring, obtain the matching score of the point of interest similarity between the user interest point having received and the advertisement point of interest prestoring, obtain the corresponding advertisement of advertisement point of interest of the matching score first predetermined condition of point of interest similarity, described advertisement is the advertisement of the coupling obtaining according to described user interest point, in the advertisement of the advertisement of the described coupling obtaining according to user's keyword and the coupling that obtains according to user interest point, identical advertisement is the advertisement that coupling obtains.
202c combines user's keyword and user interest point to mate.Certainly, 202c obtains advertisement and obtains advertisement according to user's keyword coupling according to user interest point coupling does not have sequencing, and in the advertisement of the advertisement of the coupling just finally obtaining according to user's keyword and the coupling that obtains according to user interest point, identical advertisement is the advertisement that coupling obtains.
In actual applications, while there is no identical advertisement in the advertisement of the advertisement of the coupling obtaining according to user's keyword and the coupling that obtains according to user interest point, both mate respectively the advertisement that obtained advertisement can think that final coupling obtains.
Wherein, 202a, 202b and 202c select an execution.
When step S201 specifically comprises user's lists of keywords of receiving input method application by network and sending, at least one in user interest point list, step S202 specifically comprises: obtain and the user's lists of keywords having received, at least one advertisement prestoring matching in user interest point list.Now, step S202 comprises following particular content:
When obtaining the advertisement prestoring matching with the user's lists of keywords having received, comprising:
202d1: according to the user's lists of keywords receiving, according to degree of confidence order from high to low, determine user's keyword of degree of confidence second predetermined condition;
For example, the second predetermined condition: according to degree of confidence order from high to low, select the highest user's keyword of degree of confidence of user's keyword.
202d2: search the advertisement keyword that the user's keyword with degree of confidence second predetermined condition matches from the advertisement keyword index prestoring, according to the advertisement keyword finding, search acquisition advertisement from advertisement base.
When obtaining the advertisement prestoring matching with the user interest point list having received, comprising:
202e1: according to the user interest point list receiving, according to degree of confidence order from high to low, determine that degree of confidence meets the user interest point of the 3rd predetermined condition;
For example, the 3rd predetermined condition: according to degree of confidence order from high to low, select the highest user interest point of degree of confidence of user interest point.
202e2: the user interest point that described degree of confidence is met to the 3rd predetermined condition carries out respectively the calculating of point of interest similarity with a plurality of advertisement points of interest that prestore, the matching score that obtains degree of confidence and be the point of interest similarity between primary user interest point and the advertisement point of interest prestoring, the corresponding advertisement of advertisement point of interest of the matching score first predetermined condition of described point of interest similarity is mates the advertisement obtaining.
When obtaining the advertisement prestoring matching with the user's lists of keywords having received and user interest point list, comprising:
202f1: according to the user's lists of keywords and the user interest point list that receive, according to degree of confidence order from high to low, determine that degree of confidence meets user's keyword and the user interest point of the 4th predetermined condition;
For example, the 4th predetermined condition: according to degree of confidence order from high to low, the degree of confidence of selecting user's keyword and user interest point is the highest user's keyword and user interest point all.
202f2: search from the advertisement keyword index prestoring with degree of confidence and meet the advertisement keyword that user's keyword of the 4th predetermined condition matches, according to the advertisement keyword finding, from advertisement base, search acquisition advertisement, described advertisement is the advertisement of the coupling obtaining according to described user's keyword, the user interest point that described degree of confidence is met to the 4th predetermined condition carries out respectively the calculating of point of interest similarity with a plurality of advertisement points of interest that prestore, the matching score of the point of interest similarity between the advertisement point of interest that obtains the user interest point of satisfied the 4th predetermined condition of degree of confidence and prestore, obtain the corresponding advertisement of advertisement point of interest of the matching score first predetermined condition of point of interest similarity, described advertisement is the advertisement of the coupling obtaining according to described user interest point, in the advertisement of the advertisement of the described coupling obtaining according to user's keyword and the coupling that obtains according to user interest point, identical advertisement is the advertisement that coupling obtains.
Certainly, 202f2 carries out match advertisements and carries out match advertisements according to user's keyword of satisfied the 4th predetermined condition of degree of confidence there is no sequencing according to the user interest point of satisfied the 4th predetermined condition of degree of confidence, just finally according to the advertisement of the coupling of user's keyword acquisition with according to identical advertisement in the advertisement of the coupling of user interest point acquisition, is and mates the advertisement obtaining.
In actual applications, while there is no identical advertisement in the advertisement of the advertisement of the coupling obtaining according to user's keyword and the coupling that obtains according to user interest point, both mate respectively the advertisement that obtained advertisement can think that final coupling obtains.
Step S203: send to described input method application the advertisement that coupling obtains by network.
The present invention receives at least one user's keyword, user interest point by network from input method application; After the coupling of server, to input method application, send advertisement, and at precalculated position showing advertisements.By the way, can input user's keyword and/or the user interest point excavating in history text information according to user, to user, push rapidly location advertisement more accurately, thereby further make full use of user's the historical resource of input.
Consult Fig. 4, Fig. 4 be the present invention is based on input behavior carry out advertisement fixing to the process flow diagram of the another embodiment of method, present embodiment is the process flow diagram of input method application end, comprising:
Step S301: send to server the user who obtains by network and input history text information;
Step S302: the advertisement that reception server sends, and represent described advertisement in precalculated position, advertisement is inputted history text information by server according to user and mated and obtain.
The embodiment of present embodiment and Fig. 1 is basic identical, and the detailed description of step S301 and step S302, refers to Fig. 1, at this, no longer goes to live in the household of one's in-laws on getting married and chats; The embodiment difference of present embodiment and Fig. 1 is: present embodiment is inputted history text information by the user of acquisition and directly to server, sent, and by server, obtains at least one in user's keyword, user interest point.
Consult Fig. 5, Fig. 5 be the present invention is based on input behavior carry out advertisement fixing to the process flow diagram of the another embodiment of method, present embodiment is the process flow diagram of server end, comprising:
Step S401: the user who receives input method application transmission by network inputs history text information;
Step S402: input and obtain at least one in user's keyword, user interest point history text information from the user who has received;
Step S403: at least one advertisement prestoring matching in acquisition and user's keyword, user interest point;
Step S404: send to input method application the advertisement that coupling obtains by network.
The embodiment of present embodiment and Fig. 3 is basic identical, and the detailed description of step S401 and step S402, refers to Fig. 2, and the detailed description of step S403 and step S404, refers to Fig. 3, at this, no longer goes to live in the household of one's in-laws on getting married and chats; The embodiment difference of present embodiment and Fig. 3 is: present embodiment server receives user and inputs history text information, by server, obtains at least one in user's keyword, user interest point.In actual applications, obtain user's mandate, user can be inputted to history text information and send to server.
Consult Fig. 6 to Figure 10, Fig. 6 to Figure 10 be the present invention is based on input behavior carry out advertisement fixing to the structural representation of five embodiments of device, these five devices that embodiment is input method application end, described device comprises: second obtains module 101, first obtains module 102, sending module 103 and receiver module 104.
Second obtains module 101 carries out pre-service for the input historical data to user, obtains user and inputs history text information.
Second obtains module 101 specifically for input the form of historical data according to user, to user's input historical data screen, splice, punctuate and impurity elimination, obtain user and input history text information.
The first acquisition module 102 is inputted history text information for the user from obtaining and is obtained at least one user's keyword, user interest point;
First obtains module 102 comprises: keyword obtains unit 1021, statistic unit 1022, lists of keywords acquisition unit 1023, the first point of interest obtains unit 1024 or the second point of interest obtains unit 1025.Wherein, the first point of interest obtains unit 1024 or a selection is selected in the second point of interest acquisition unit 1025, and first other unit that obtains module 102, except Fig. 6 to Figure 10, can increase and decrease according to actual needs, at this, no longer goes to live in the household of one's in-laws on getting married and chats.
Keyword obtains unit 1021 and mates with the keyword set prestoring for the user of acquisition being inputted to history text information, obtains user's keyword.
Statistic unit 1022 is added up for inputting user the frequency and/or the time that history text information occurs according to described user's keyword, obtain the degree of confidence of described user's keyword, wherein, user, to input the frequency and/or the time that in history text information, occur be positively related relation for the degree of confidence of described user's keyword and described user's keyword;
Lists of keywords obtains unit 1023 for according to the degree of confidence of described user's keyword, and user's keyword is sorted, and obtains user's lists of keywords, and described user's lists of keywords comprises each user's keyword and degree of confidence thereof.
The first point of interest obtains unit 1024 for adopting topic model, the user of described acquisition is inputted to history text information integral body and carry out theme statistics, obtains described user interest point, and wherein, each theme is a user interest point.
The first point of interest obtains unit 1024 specifically for adopting probability latent semantic analysis model (PLSA) or the Di Li Cray assignment-type model (LDA) of hiding, the user of described acquisition is inputted to the statistics that history text information integral body is carried out the degree of confidence of theme statistics and each theme, obtain user interest point list, described user interest point list comprises each user interest point and degree of confidence thereof.
The theme distributed intelligence of each entry that the second point of interest acquisition unit 1025 prestores for basis, the user who adds up described acquisition input history text information the theme distributed intelligence of corresponding entry, obtain user interest point list, described user interest point list comprises each user interest point and weight thereof, wherein, each theme is a user interest point.
Sending module 103 is for sending at least one of described user's keyword, user interest point to server by network;
Sending module 103 is specifically for sending at least one in user's lists of keywords, user interest point list by network to server.
The advertisement that receiver module 104 sends for receiving described server, and represent described advertisement in precalculated position, described advertisement in server match user keyword, user interest point at least one and obtain.
Receiver module 104 comprises and represents unit, described in represent unit and comprise: first represents subelement, second represents subelement and the 3rd and represents subelement.
First represents subelement represents described advertisement for the precalculated position outside the Zhong Huo interface, interface in input method application, and described precalculated position is candidate frame, status bar or applies the candidate's who is associated expansion explanation frame with input method;
Or second represents subelement represents described advertisement for the precalculated position of the interface phase association in third party application;
Or the 3rd represents subelement represents described advertisement for the showing advertisement position providing in third party's application.
Wherein, the advertisement that receiver module 104 sends specifically for receiving described server, and in the precalculated position of predetermined scene, represent the advertisement of mating with described predetermined scene.
The present invention inputs at least one history text information acquisition user keyword, user interest point from the user who obtains, and utilizes described user's keyword or user interest point to carry out match advertisements, and at precalculated position showing advertisements.By the way, can input user's keyword and/or the user interest point excavating in history text information according to user, to user, push rapidly location advertisement more accurately, thereby further make full use of user's the historical resource of input.
Consult Figure 11, Figure 11 be the present invention is based on input behavior carry out advertisement fixing to the structural representation of the another embodiment of device, this implementation method is the device of input method application end, described device comprises: sending module 201 and receiver module 202.
Sending module 201 is inputted history text information for sending to server the user who obtains by network;
The advertisement that receiver module 202 sends for receiving described server, and represent described advertisement in precalculated position, described advertisement is inputted history text information by server according to user and is mated and obtain.
Consult Figure 12 to Figure 14, Figure 12 to Figure 14 is the structural representation of three embodiments of server of the present invention, and server comprises: receiver module 301, acquisition module 302 and sending module 303.
Receiver module 301 is for receiving user's keyword of input method application transmission, at least one of user interest point by network;
Receiver module 301 is specifically for receiving user's lists of keywords of input method application transmission, at least one in user interest point list by network.
Obtain module 302 for the user's keyword that obtains and received, at least one advertisement prestoring matching of user interest point;
Obtaining module 302 comprises: the first matching unit 3021 or the second matching unit 3022 or the 3rd matching unit 3023.
The first matching unit 3021 is searched the advertisement keyword matching with the user's keyword having received from the advertisement keyword index prestoring, and according to the advertisement keyword finding, searches acquisition advertisement from advertisement base;
Or, the second matching unit 3022 for by received each described in user interest point carry out respectively the calculating of point of interest similarity with a plurality of advertisement points of interest of prestoring, the matching score that obtains the point of interest similarity between the user interest point having received and the advertisement point of interest prestoring, the corresponding advertisement of advertisement point of interest of the matching score first predetermined condition of described point of interest similarity is the advertisement that coupling obtains;
Or, the 3rd matching unit 3023 is searched the advertisement keyword matching with the user's keyword having received for the advertisement keyword index from prestoring, according to the advertisement keyword finding, from advertisement base, search acquisition advertisement, described advertisement is the advertisement of the coupling obtaining according to described user's keyword, by received each described in user interest point carry out respectively the calculating of point of interest similarity with a plurality of advertisement points of interest of prestoring, obtain the matching score of the point of interest similarity between the user interest point having received and the advertisement point of interest prestoring, obtain the corresponding advertisement of advertisement point of interest of the matching score first predetermined condition of point of interest similarity, described advertisement is the advertisement of the coupling obtaining according to described user interest point, in the advertisement of the advertisement of the described coupling obtaining according to user's keyword and the coupling that obtains according to user interest point, identical advertisement is the advertisement that coupling obtains.
Sending module 303 is for sending to described input method application the advertisement that coupling obtains by network.
The present invention receives at least one user's keyword, user interest point by network from input method application; After the coupling of server, to input method application, send advertisement, and at precalculated position showing advertisements.By the way, can input user's keyword and/or the user interest point excavating in history text information according to user, to user, push rapidly location advertisement more accurately, thereby further make full use of user's the historical resource of input.
Consult Figure 15, Figure 15 is the structural representation of the another embodiment of server of the present invention, and server comprises: receiver module 401, acquisition module 402 and sending module 403.
Receiver module 401 is for receiving user's lists of keywords of input method application transmission, at least one of user interest point list by network;
Obtain module 402 for the user's lists of keywords that obtains and received, at least one advertisement prestoring matching of user interest point list;
Obtaining module 402 also comprises: the first determining unit, first obtains unit, or the second determining unit, second obtains unit, or the 3rd determining unit and the 3rd obtains unit.
The first determining unit, for according to the user's lists of keywords receiving, according to degree of confidence order from high to low, is determined user's keyword of degree of confidence second predetermined condition;
First obtains the advertisement keyword that unit matches for user's keyword of searching from the advertisement keyword index prestoring with degree of confidence second predetermined condition, according to the advertisement keyword finding, searches acquisition advertisement from advertisement base.
The second determining unit, for according to the user interest point list receiving, according to degree of confidence order from high to low, determines that degree of confidence meets the user interest point of the 3rd predetermined condition;
Second obtains unit carries out respectively the calculating of point of interest similarity for described degree of confidence being met to the user interest point of the 3rd predetermined condition with a plurality of advertisement points of interest that prestore, the matching score of the point of interest similarity between the advertisement point of interest that obtains the user interest point of satisfied the 3rd predetermined condition of degree of confidence and prestore, the corresponding advertisement of advertisement point of interest of the matching score first predetermined condition of described point of interest similarity is the advertisement that coupling obtains.
The 3rd determining unit, for according to the user's lists of keywords and the user interest point list that receive, according to degree of confidence order from high to low, determines that degree of confidence meets user's keyword and the user interest point of the 4th predetermined condition;
The 3rd obtains unit meets for searching from the advertisement keyword index prestoring with degree of confidence the advertisement keyword that user's keyword of the 4th predetermined condition matches, according to the advertisement keyword finding, from advertisement base, search acquisition advertisement, described advertisement is the advertisement of the coupling obtaining according to described user's keyword, the user interest point that described degree of confidence is met to the 4th predetermined condition carries out respectively the calculating of point of interest similarity with a plurality of advertisement points of interest that prestore, the matching score of the point of interest similarity between the advertisement point of interest that obtains the user interest point of satisfied the 4th predetermined condition of degree of confidence and prestore, obtain the corresponding advertisement of advertisement point of interest of the matching score first predetermined condition of point of interest similarity, described advertisement is the advertisement of the coupling obtaining according to described user interest point, in the advertisement of the advertisement of the described coupling obtaining according to user's keyword and the coupling that obtains according to user interest point, identical advertisement is the advertisement that coupling obtains.
Sending module 403 is for sending to described input method application the advertisement that coupling obtains by network.
The present invention receives at least one user's keyword, user interest point by network from input method application; After the coupling of server, to input method application, send advertisement, and at precalculated position showing advertisements.By the way, can input user's keyword and/or the user interest point excavating in history text information according to user, to user, push rapidly location advertisement more accurately, thereby further make full use of user's the historical resource of input.
Consult Figure 16, Figure 16 is the structural representation of the another embodiment of server of the present invention, and described server comprises: receiver module 501, first obtains module 502, second and obtains module 503 and sending module 504.
Receiver module 501 is inputted history text information for receive the user of input method application transmission by network;
The first acquisition module 502 is inputted history text information for the user from having received and is obtained at least one user's keyword, user interest point;
Second obtains module 503 for obtaining at least one advertisement prestoring matching with described user's keyword, user interest point;
Sending module 504 is for sending to described input method application the advertisement that coupling obtains by network.
The present invention inputs at least one history text information acquisition user keyword, user interest point from the user who obtains, and utilizes described user's keyword or user interest point to carry out match advertisements, and at precalculated position showing advertisements.By the way, can input user's keyword and/or the user interest point excavating in history text information according to user, to user, push rapidly location advertisement more accurately, thereby further make full use of user's the historical resource of input.
In several embodiments provided by the present invention, should be understood that, disclosed system, apparatus and method, can realize by another way.For example, device embodiments described above is only schematic, for example, the division of described module or unit, be only that a kind of logic function is divided, during actual realization, can have other dividing mode, for example a plurality of unit or assembly can in conjunction with or can be integrated into another system, or some features can ignore, or do not carry out.Another point, shown or discussed coupling each other or direct-coupling or communication connection can be by some interfaces, indirect coupling or the communication connection of device or unit can be electrically, machinery or other form.
The described unit as separating component explanation can or can not be also physically to separate, and the parts that show as unit can be or can not be also physical locations, can be positioned at a place, or also can be distributed in a plurality of network element.Can select according to the actual needs some or all of unit wherein to realize the object of present embodiment scheme.
In addition, each functional unit in each embodiment of the present invention can be integrated in a processing unit, can be also that the independent physics of unit exists, and also can be integrated in a unit two or more unit.Above-mentioned integrated unit both can adopt the form of hardware to realize, and also can adopt the form of SFU software functional unit to realize.
If the form of SFU software functional unit of usining described integrated unit realizes and during as production marketing independently or use, can be stored in a computer read/write memory medium.Understanding based on such, the all or part of of the part that technical scheme of the present invention contributes to prior art in essence in other words or this technical scheme can embody with the form of software product, this computer software product is stored in a storage medium, comprise that some instructions are with so that a computer equipment (can be personal computer, server, or the network equipment etc.) or processor (processor) carry out all or part of step of method described in each embodiment of the application.And aforesaid storage medium comprises: various media that can be program code stored such as USB flash disk, portable hard drive, ROM (read-only memory) (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disc or CDs.
The foregoing is only embodiments of the present invention; not thereby limit the scope of the claims of the present invention; every equivalent structure or conversion of equivalent flow process that utilizes instructions of the present invention and accompanying drawing content to do; or be directly or indirectly used in other relevant technical fields, be all in like manner included in scope of patent protection of the present invention.

Claims (35)

  1. Based on input behavior carry out advertisement fixing to a method, it is characterized in that, comprising:
    From the user who obtains, input and history text information, obtain at least one in user's keyword, user interest point;
    By network, to server, send at least one in described user's keyword, user interest point;
    Receive the advertisement that described server sends, and in precalculated position, represent described advertisement, described advertisement in server match user keyword, user interest point at least one and obtain.
  2. 2. method according to claim 1, it is characterized in that, the described user from acquisition inputs the step that obtains user's keyword history text information, comprising: the user of acquisition is inputted to history text information and mate with the keyword set prestoring, obtain described user's keyword.
  3. 3. method according to claim 2, is characterized in that, the described user by acquisition inputs history text information mates with the keyword set prestoring, and after obtaining the step of user's keyword, also comprises:
    According to described user's keyword, user, inputting the frequency and/or the time that in history text information, occur adds up, obtain the degree of confidence of described user's keyword, wherein, user, to input the frequency and/or the time that in history text information, occur be positively related relation for the degree of confidence of described user's keyword and described user's keyword;
    According to the degree of confidence of described user's keyword, user's keyword is sorted, obtain user's lists of keywords, described user's lists of keywords comprises each user's keyword and degree of confidence thereof.
  4. 4. method according to claim 1, is characterized in that, the described user from acquisition inputs the step that obtains user interest point history text information, comprising:
    Adopt topic model, the user of described acquisition is inputted to history text information integral body and carry out theme statistics, obtain described user interest point, wherein, each theme is a user interest point.
  5. 5. method according to claim 4, is characterized in that, described employing topic model inputs to the user of described acquisition the step that history text information integral body is carried out theme statistics, comprising:
    Adopt probability latent semantic analysis model (PLSA) or the Di Li Cray assignment-type model (LDA) of hiding, the user of described acquisition is inputted to the statistics that history text information integral body is carried out the degree of confidence of theme statistics and each theme, obtain user interest point list, described user interest point list comprises each user interest point and degree of confidence thereof.
  6. 6. method according to claim 1, is characterized in that, the described user from acquisition inputs the step that obtains user interest point history text information, comprising:
    According to the theme distributed intelligence of each entry prestoring, the user who adds up described acquisition input history text information the theme distributed intelligence of corresponding entry, obtain user interest point list, described user interest point list comprises each user interest point and weight thereof, wherein, each theme is a user interest point.
  7. 7. according to the method described in claim 3,5 or 6, it is characterized in that, describedly by network, to server, send at least one the step in described user's keyword, user interest point, comprising: by network, to server, send at least one in user's lists of keywords, user interest point list.
  8. 8. according to the arbitrary described method of claim 1 to 6, it is characterized in that, the described step that represents described advertisement in precalculated position, comprising:
    Precalculated position outside the Zhong Huo interface, interface of input method application represents described advertisement, and described precalculated position is candidate frame, status bar or applies the candidate's who is associated expansion explanation frame with input method;
    Or the precalculated position of the interface phase association of applying third party represents described advertisement;
    Or the showing advertisement position providing in third party's application represents described advertisement.
  9. 9. according to the arbitrary described method of claim 1 to 6, it is characterized in that, the advertisement that the described server of described reception sends, and in precalculated position, represent the step of described advertisement, comprise: receive the advertisement that described server sends, and in the precalculated position of predetermined scene, represent the advertisement of mating with described predetermined scene.
  10. 10. according to the arbitrary described method of claim 1 to 6, it is characterized in that, before described at least one the step obtaining from the user who obtains inputs history text information in user's keyword, user interest point, comprise: the input historical data to user is carried out pre-service, obtain described user and input history text information.
  11. 11. methods according to claim 10, it is characterized in that, the described input historical data to user is carried out pre-service, obtain the step that described user inputs history text information, comprise: the form of inputting historical data according to user, input historical data to user is screened, is spliced, punctuate and impurity elimination, obtains described user and inputs history text information.
  12. 12. 1 kinds based on input behavior carry out advertisement fixing to method, it is characterized in that, comprising:
    By network, receive user's keyword of input method application transmission, at least one in user interest point;
    Obtain and the user's keyword having received, at least one advertisement prestoring matching in user interest point;
    By network, to described input method application, send the advertisement that coupling obtains.
  13. 13. methods according to claim 12, is characterized in that, described acquisition and the user's keyword having received, the step of at least one advertisement prestoring matching in user interest point, comprising:
    From the advertisement keyword index prestoring, search the advertisement keyword matching with the described user's keyword having received, according to the described advertisement keyword finding, from advertisement base, search and obtain described advertisement;
    Or, by received each described in user interest point carry out respectively the calculating of point of interest similarity with a plurality of advertisement points of interest of prestoring, the matching score that obtains the point of interest similarity between the described user interest point having received and the advertisement point of interest prestoring, the corresponding advertisement of advertisement point of interest of the matching score first predetermined condition of described point of interest similarity is the advertisement that coupling obtains;
    Or, from the advertisement keyword index prestoring, search the advertisement keyword matching with the described user's keyword having received, according to the described advertisement keyword finding, from advertisement base, search and obtain described advertisement, described advertisement is the advertisement of the coupling obtaining according to described user's keyword, by received each described in user interest point carry out respectively the calculating of point of interest similarity with a plurality of advertisement points of interest of prestoring, obtain the matching score of the point of interest similarity between the user interest point having received and the advertisement point of interest prestoring, obtain the corresponding advertisement of advertisement point of interest of the matching score first predetermined condition of point of interest similarity, described advertisement is the advertisement of the coupling obtaining according to described user interest point, in the advertisement of the advertisement of the described coupling obtaining according to user's keyword and the coupling that obtains according to user interest point, identical advertisement is the advertisement that coupling obtains.
  14. 14. methods according to claim 12, it is characterized in that, describedly by network, receive user's keyword that input method application sends, at least one step in user interest point, comprising: by network, receives user's lists of keywords of input method application transmission, at least one in user interest point list.
  15. 15. methods according to claim 14, it is characterized in that, described acquisition and the user's keyword having received, the step of at least one advertisement prestoring matching in user interest point, comprising: obtain and the user's lists of keywords having received, at least one advertisement prestoring matching in user interest point list.
  16. 16. methods according to claim 15, is characterized in that, the step of the advertisement prestoring that described acquisition and the user's lists of keywords having received match, comprising:
    According to the user's lists of keywords receiving, according to degree of confidence order from high to low, determine user's keyword of degree of confidence second predetermined condition;
    From the advertisement keyword index prestoring, search the advertisement keyword that the user's keyword with degree of confidence second predetermined condition matches, according to the described advertisement keyword finding, from advertisement base, search and obtain described advertisement.
  17. 17. methods according to claim 15, is characterized in that, the step of the advertisement prestoring that described acquisition and the user interest point list having received match, comprising:
    According to the user interest point list receiving, according to degree of confidence order from high to low, determine that degree of confidence meets the user interest point of the 3rd predetermined condition;
    The user interest point that described degree of confidence is met to the 3rd predetermined condition carries out respectively the calculating of point of interest similarity with a plurality of advertisement points of interest that prestore, the matching score of the point of interest similarity between the advertisement point of interest that obtains the user interest point of satisfied the 3rd predetermined condition of degree of confidence and prestore, the corresponding advertisement of advertisement point of interest of the matching score first predetermined condition of described point of interest similarity is the advertisement that coupling obtains.
  18. 18. according to claim 15 to 17 arbitrary described methods, it is characterized in that, the step of the advertisement prestoring that described acquisition and the user's lists of keywords having received and user interest point list match, comprising:
    According to the user's lists of keywords and the user interest point list that receive, according to degree of confidence order from high to low, determine that degree of confidence meets user's keyword and the user interest point of the 4th predetermined condition;
    From the advertisement keyword index prestoring, search with degree of confidence and meet the advertisement keyword that user's keyword of the 4th predetermined condition matches, according to the described advertisement keyword finding, from advertisement base, search and obtain described advertisement, described advertisement is the advertisement of the coupling obtaining according to described user's keyword, the user interest point that described degree of confidence is met to the 4th predetermined condition carries out respectively the calculating of point of interest similarity with a plurality of advertisement points of interest that prestore, the matching score of the point of interest similarity between the advertisement point of interest that obtains the user interest point of satisfied the 4th predetermined condition of degree of confidence and prestore, obtain the corresponding advertisement of advertisement point of interest of the matching score first predetermined condition of point of interest similarity, described advertisement is the advertisement of the coupling obtaining according to described user interest point, in the advertisement of the advertisement of the described coupling obtaining according to user's keyword and the coupling that obtains according to user interest point, identical advertisement is the advertisement that coupling obtains.
  19. 19. 1 kinds based on input behavior carry out advertisement fixing to method, it is characterized in that, comprising:
    By network, to server, send the user who obtains and input history text information;
    Receive the advertisement that described server sends, and in precalculated position, represent described advertisement, described advertisement is inputted history text information by server according to user and is mated and obtain.
  20. 20. 1 kinds based on input behavior carry out advertisement fixing to method, it is characterized in that, comprising:
    The user who receives input method application transmission by network inputs history text information;
    From the user who has received, input and history text information, obtain at least one in user's keyword, user interest point;
    At least one advertisement prestoring matching in acquisition and described user's keyword, user interest point;
    By network, to described input method application, send the advertisement that coupling obtains.
  21. 21. 1 kinds based on input behavior carry out advertisement fixing to device, it is characterized in that, described device comprises:
    First obtains module, inputs history text information obtain at least one user's keyword, user interest point for the user from obtaining;
    Sending module, for sending at least one of described user's keyword, user interest point to server by network;
    Receiver module, the advertisement sending for receiving described server, and represent described advertisement in precalculated position, described advertisement in server match user keyword, user interest point at least one and obtain.
  22. 22. devices according to claim 21, it is characterized in that, described first obtains module comprises: keyword obtains unit, and described keyword obtains unit and mates with the keyword set prestoring for the user of acquisition is inputted to history text information, obtains described user's keyword.
  23. 23. devices according to claim 22, is characterized in that, described first obtains module also comprises:
    Statistic unit, for inputting user the frequency and/or the time that history text information occurs according to described user's keyword, add up, obtain the degree of confidence of described user's keyword, wherein, user, to input the frequency and/or the time that in history text information, occur be positively related relation for the degree of confidence of described user's keyword and described user's keyword;
    Lists of keywords obtains unit, for according to the degree of confidence of described user's keyword, user's keyword is sorted, and obtains user's lists of keywords, and described user's lists of keywords comprises each user's keyword and degree of confidence thereof.
  24. 24. devices according to claim 21, it is characterized in that, described first obtains module comprises: the first point of interest obtains unit, described the first point of interest obtains unit and is used for adopting topic model, the user of described acquisition is inputted to history text information integral body and carry out theme statistics, obtain described user interest point, wherein, each theme is a user interest point.
  25. 25. devices according to claim 24, it is characterized in that, described point of interest obtains unit specifically for adopting probability latent semantic analysis model (PLSA) or the Di Li Cray assignment-type model (LDA) of hiding, the user of described acquisition is inputted to the statistics that history text information integral body is carried out the degree of confidence of theme statistics and each theme, obtain user interest point list, described user interest point list comprises each user interest point and degree of confidence thereof.
  26. 26. devices according to claim 21, it is characterized in that, described first obtains module comprises: the second point of interest obtains unit, described the second point of interest obtains unit and is used for according to the theme distributed intelligence of each entry prestoring, the user who adds up described acquisition input history text information the theme distributed intelligence of corresponding entry, obtain user interest point list, described user interest point list comprises each user interest point and weight thereof, wherein, each theme is a user interest point.
  27. 27. according to the device described in claim 23,25 or 26, it is characterized in that, described sending module is specifically for sending at least one in user's lists of keywords, user interest point list by network to server.
  28. 28. according to the arbitrary described device of claim 21 to 26, it is characterized in that, described receiver module comprises: represent unit, described in represent unit and comprise:
    First represents subelement, for the precalculated position outside the Zhong Huo interface, interface in input method application, represents described advertisement, and described precalculated position is candidate frame, status bar or applies the candidate's who is associated expansion explanation frame with input method;
    Or second represents subelement, for the precalculated position of the interface phase association in third party application, represent described advertisement;
    Or the 3rd represents subelement, for the showing advertisement position providing in third party's application, represent described advertisement.
  29. 29. according to the arbitrary described device of claim 21 to 26, it is characterized in that, and the advertisement that described receiver module sends specifically for receiving described server, and in the precalculated position of predetermined scene, represent the advertisement of mating with described predetermined scene.
  30. 30. according to the arbitrary described device of claim 21 to 26, it is characterized in that, described device also comprises: second obtains module, and described second obtains module for user's input historical data is carried out to pre-service, obtains described user and inputs history text information.
  31. 31. devices according to claim 30, it is characterized in that, described second obtains module specifically for input the form of historical data according to user, to user's input historical data screen, splice, punctuate and impurity elimination, obtain described user and input history text information.
  32. 32. 1 kinds of servers, is characterized in that, described server comprises:
    Receiver module, for receiving user's keyword of input method application transmission, at least one of user interest point by network;
    Obtain module, for the user's keyword that obtains and received, at least one advertisement prestoring matching of user interest point;
    Sending module, for sending to described input method application the advertisement that coupling obtains by network.
  33. 33. servers according to claim 32, is characterized in that, described acquisition module comprises:
    The first matching unit, searches the advertisement keyword matching with the described user's keyword having received for the advertisement keyword index from prestoring, according to the described advertisement keyword finding, search and obtain described advertisement from advertisement base;
    Or, the second matching unit, for by received each described in user interest point carry out respectively the calculating of point of interest similarity with a plurality of advertisement points of interest of prestoring, the matching score that obtains the point of interest similarity between the user interest point having received and the advertisement point of interest prestoring, the corresponding advertisement of advertisement point of interest of the matching score first predetermined condition of described point of interest similarity is the advertisement that coupling obtains;
    Or, the 3rd matching unit, for searching from the advertisement keyword index prestoring the advertisement keyword matching with the described user's keyword having received, according to the described advertisement keyword finding, from advertisement base, search and obtain described advertisement, described advertisement is the advertisement of the coupling obtaining according to described user's keyword, by received each described in user interest point carry out respectively the calculating of point of interest similarity with a plurality of advertisement points of interest of prestoring, obtain the matching score of the point of interest similarity between the user interest point having received and the advertisement point of interest prestoring, obtain the corresponding advertisement of advertisement point of interest of the matching score first predetermined condition of point of interest similarity, described advertisement is the advertisement of the coupling obtaining according to described user interest point, in the advertisement of the advertisement of the described coupling obtaining according to user's keyword and the coupling that obtains according to user interest point, identical advertisement is the advertisement that coupling obtains.
  34. 34. 1 kinds based on input behavior carry out advertisement fixing to device, it is characterized in that, described device comprises:
    Sending module, inputs history text information for sending to server the user who obtains by network;
    Receiver module, the advertisement sending for receiving described server, and represent described advertisement in precalculated position, described advertisement is inputted history text information by server according to user and is mated and obtain.
  35. 35. 1 kinds of servers, is characterized in that, described server comprises:
    Receiver module, inputs history text information for receive the user of input method application transmission by network;
    First obtains module, inputs history text information obtain at least one user's keyword, user interest point for the user from having received;
    Second obtains module, for obtaining at least one advertisement prestoring matching with described user's keyword, user interest point;
    Sending module, for sending to described input method application the advertisement that coupling obtains by network.
CN201210304052.9A 2012-08-23 2012-08-23 Method, device and server for advertisement orientation based on input behaviors Pending CN103631803A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201210304052.9A CN103631803A (en) 2012-08-23 2012-08-23 Method, device and server for advertisement orientation based on input behaviors

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201210304052.9A CN103631803A (en) 2012-08-23 2012-08-23 Method, device and server for advertisement orientation based on input behaviors

Publications (1)

Publication Number Publication Date
CN103631803A true CN103631803A (en) 2014-03-12

Family

ID=50212873

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201210304052.9A Pending CN103631803A (en) 2012-08-23 2012-08-23 Method, device and server for advertisement orientation based on input behaviors

Country Status (1)

Country Link
CN (1) CN103631803A (en)

Cited By (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104182471A (en) * 2014-07-29 2014-12-03 小米科技有限责任公司 Method, device and terminal for determining interest points of user
CN104820695A (en) * 2015-04-29 2015-08-05 百度在线网络技术（北京）有限公司 Method and device for resource acquisition
WO2015131528A1 (en) * 2014-03-07 2015-09-11 北京奇虎科技有限公司 Method and apparatus for determining topic distribution of given text
CN106133776A (en) * 2014-03-28 2016-11-16 微软技术许可有限责任公司 Based on the managed mutual structure user with user, profile is created and the trust of recommendation
WO2017024504A1 (en) * 2015-08-11 2017-02-16 常平 Method for prompting information and system for pushing advertisement when inserting advertisement according to iris
WO2017024505A1 (en) * 2015-08-11 2017-02-16 常平 Iris-based advertisement push method and advertisement push system
CN106528781A (en) * 2016-10-31 2017-03-22 努比亚技术有限公司 Link pushing device and method
CN106708802A (en) * 2016-12-20 2017-05-24 西南石油大学 Information recommendation method and system
TWI587179B (en) * 2016-05-20 2017-06-11 zheng-hong Lin Associated ad input method
WO2017190610A1 (en) * 2016-05-05 2017-11-09 腾讯科技（深圳）有限公司 Target user orientation method and device, and computer storage medium
WO2018027362A1 (en) * 2016-08-06 2018-02-15 胡明祥 Moments information-based advertisement pushing method and pushing system
CN107924527A (en) * 2015-08-20 2018-04-17 三星电子株式会社 Show ads in the app launcher
CN108108998A (en) * 2017-12-14 2018-06-01 百度在线网络技术（北京）有限公司 Showing advertisement method and apparatus, server, storage medium
CN108537596A (en) * 2018-04-16 2018-09-14 北京车慧互动广告有限公司 Method, device, system and memory for car model recommendation in search box
CN110069698A (en) * 2017-11-01 2019-07-30 北京京东尚科信息技术有限公司 Information-pushing method and device
CN113407040A (en) * 2021-05-13 2021-09-17 北京搜狗科技发展有限公司 Information processing method, device and medium
CN114936885A (en) * 2022-07-21 2022-08-23 成都薯片科技有限公司 Advertisement information matching pushing method, device, system, equipment and storage medium

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060212897A1 (en) * 2005-03-18 2006-09-21 Microsoft Corporation System and method for utilizing the content of audio/video files to select advertising content for display
CN1877582A (en) * 2006-07-12 2006-12-13 百度在线网络技术（北京）有限公司 Advertisement information retrieval system and method therefor
CN1971568A (en) * 2006-12-12 2007-05-30 北京搜狗科技发展有限公司 Method and system for issuing relative information of key characters of internet
CN101079038A (en) * 2006-06-28 2007-11-28 腾讯科技（深圳）有限公司 System and method for implementing key word advertisement
CN101216825A (en) * 2007-12-29 2008-07-09 朱廷劭 Indexing key words extraction/ prediction method, on-line advertisement recommendation method and device
CN101420313A (en) * 2007-10-22 2009-04-29 北京搜狗科技发展有限公司 Method and system for clustering customer terminal user group
CN101635761A (en) * 2008-07-23 2010-01-27 北京搜狗科技发展有限公司 Method and device for displaying information in input process by using mobile terminal
US20110231252A1 (en) * 2010-03-17 2011-09-22 Microsoft Corporation Voice-enabled text advertisements

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060212897A1 (en) * 2005-03-18 2006-09-21 Microsoft Corporation System and method for utilizing the content of audio/video files to select advertising content for display
CN101079038A (en) * 2006-06-28 2007-11-28 腾讯科技（深圳）有限公司 System and method for implementing key word advertisement
CN1877582A (en) * 2006-07-12 2006-12-13 百度在线网络技术（北京）有限公司 Advertisement information retrieval system and method therefor
CN1971568A (en) * 2006-12-12 2007-05-30 北京搜狗科技发展有限公司 Method and system for issuing relative information of key characters of internet
CN101420313A (en) * 2007-10-22 2009-04-29 北京搜狗科技发展有限公司 Method and system for clustering customer terminal user group
CN101216825A (en) * 2007-12-29 2008-07-09 朱廷劭 Indexing key words extraction/ prediction method, on-line advertisement recommendation method and device
CN101635761A (en) * 2008-07-23 2010-01-27 北京搜狗科技发展有限公司 Method and device for displaying information in input process by using mobile terminal
US20110231252A1 (en) * 2010-03-17 2011-09-22 Microsoft Corporation Voice-enabled text advertisements

Cited By (21)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2015131528A1 (en) * 2014-03-07 2015-09-11 北京奇虎科技有限公司 Method and apparatus for determining topic distribution of given text
CN106133776A (en) * 2014-03-28 2016-11-16 微软技术许可有限责任公司 Based on the managed mutual structure user with user, profile is created and the trust of recommendation
CN104182471A (en) * 2014-07-29 2014-12-03 小米科技有限责任公司 Method, device and terminal for determining interest points of user
CN104182471B (en) * 2014-07-29 2018-04-06 小米科技有限责任公司 The determination method, apparatus and terminal of user interest point
CN104820695A (en) * 2015-04-29 2015-08-05 百度在线网络技术（北京）有限公司 Method and device for resource acquisition
WO2017024504A1 (en) * 2015-08-11 2017-02-16 常平 Method for prompting information and system for pushing advertisement when inserting advertisement according to iris
WO2017024505A1 (en) * 2015-08-11 2017-02-16 常平 Iris-based advertisement push method and advertisement push system
CN107924527A (en) * 2015-08-20 2018-04-17 三星电子株式会社 Show ads in the app launcher
EP3407285B1 (en) * 2016-05-05 2023-06-28 Tencent Technology (Shenzhen) Company Limited Target user orientation method and device, and computer storage medium
US11004107B2 (en) 2016-05-05 2021-05-11 Tencent Technology (Shenzhen) Company Limited Target user directing method and apparatus and computer storage medium
WO2017190610A1 (en) * 2016-05-05 2017-11-09 腾讯科技（深圳）有限公司 Target user orientation method and device, and computer storage medium
TWI587179B (en) * 2016-05-20 2017-06-11 zheng-hong Lin Associated ad input method
WO2018027362A1 (en) * 2016-08-06 2018-02-15 胡明祥 Moments information-based advertisement pushing method and pushing system
CN106528781A (en) * 2016-10-31 2017-03-22 努比亚技术有限公司 Link pushing device and method
CN106708802A (en) * 2016-12-20 2017-05-24 西南石油大学 Information recommendation method and system
CN110069698A (en) * 2017-11-01 2019-07-30 北京京东尚科信息技术有限公司 Information-pushing method and device
CN110069698B (en) * 2017-11-01 2021-09-03 北京京东尚科信息技术有限公司 Information pushing method and device
CN108108998A (en) * 2017-12-14 2018-06-01 百度在线网络技术（北京）有限公司 Showing advertisement method and apparatus, server, storage medium
CN108537596A (en) * 2018-04-16 2018-09-14 北京车慧互动广告有限公司 Method, device, system and memory for car model recommendation in search box
CN113407040A (en) * 2021-05-13 2021-09-17 北京搜狗科技发展有限公司 Information processing method, device and medium
CN114936885A (en) * 2022-07-21 2022-08-23 成都薯片科技有限公司 Advertisement information matching pushing method, device, system, equipment and storage medium

Similar Documents

Publication Publication Date Title
CN103631803A (en) Method, device and server for advertisement orientation based on input behaviors
US10180979B2 (en) System and method for generating suggestions by a search engine in response to search queries
US11687968B1 (en) Serving advertisements based on partial queries
US9798820B1 (en) Classification of keywords
US9396413B2 (en) Choosing image labels
CN101984420B (en) Method and equipment for searching pictures based on word segmentation processing
US9678618B1 (en) Using an expanded view to display links related to a topic
CN102289459A (en) Automatically generating training data
WO2009108576A2 (en) Prioritizing media assets for publication
CN102193973A (en) Presenting answers
CN106062743A (en) Systems and methods for keyword suggestion
CN104102639A (en) Text classification based promotion triggering method and device
CN104572863A (en) Product recommending method and system
CN104077415A (en) Searching method and device
CN109074366A (en) Gain adjustment components for computer network routing infrastructure
CN102163228A (en) Method, apparatus and device for determining sorting result of resource candidates
CN103942257A (en) Video search method and device
CN104751354A (en) Advertisement cluster screening method
CN103425670A (en) Method, device and equipment for providing customers with content recommendation information
EP3485394B1 (en) Contextual based image search results
CN109241403A (en) Item recommendation method, device, machinery equipment and computer readable storage medium
CN112446727A (en) Advertisement triggering method, device, equipment and computer readable storage medium
US10579630B2 (en) Content creation from extracted content
CN105706081A (en) Structured informational link annotations
US20180315092A1 (en) Server For Providing Internet Content and Computer-Readable Recording Medium Including Implemented Internet Content Providing Method

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
EXSB Decision made by sipo to initiate substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20140312