CN107436930A - Information recommends method and device - Google Patents
Information recommends method and device Download PDFInfo
- Publication number
- CN107436930A CN107436930A CN201710580717.1A CN201710580717A CN107436930A CN 107436930 A CN107436930 A CN 107436930A CN 201710580717 A CN201710580717 A CN 201710580717A CN 107436930 A CN107436930 A CN 107436930A
- Authority
- CN
- China
- Prior art keywords
- user
- attribute
- information
- label
- tally set
- 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.)
- Withdrawn
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of information to recommend method, including step：The user behaviors log of real-time collecting user, the information pointed to by label engine to the user behaviors log enters row label extraction, so as to generate the first tally set；The user includes user property, custom attribute and interest attribute；The interest attribute of the user is obtained based on first tally set, and the custom attribute of the user is obtained based on the user behaviors log；Feature extraction is carried out according to the user property of the user, interest attribute and custom attribute, the classification of user described in Dynamic Recognition, corresponding information is recommended to the user according to the classification of the user, it can realize and information recommendation is carried out based on user Web daily records, and can be measured according to user interest and complete personalized recommendation.
Description
Technical field
The present invention relates to computer realm, more particularly to a kind of information to recommend method and device.
Background technology
With the fast development of internet, there are a large amount of webpages to update or issue on the internet daily.For vast use
Want that the information for finding oneself satisfaction has been more and more difficult in substantial amounts of information for family, so as to result in " information is excessive "
With the contradictory phenomena of " information is hungry ".To solve this problem, it is proposed that individual info service, this is a kind of intelligent information clothes
Business mode.Information requirement that can be according to user and customizing mode, relevant information is actively searched, and pushed away using on-line intelligence
Service or push technology are recommended, accurately by the information transmission needed for user to corresponding user.In Personalized Service Technology,
Using being more successfully collaborative filtering method.This method refers to demand of the user according to itself, by being closed with other users
Make, form certain cooperation rule, or the interest of unique user, Ran Hougen are predicted using the tendentiousness of multiple information users
Information is evaluated according to the user that same interest is liked, so as to obtain recommendation results.It is big due to have recorded in Web daily records
The user behavior information of amount, important data can be provided for personalized service using Web daily records and are supported.It is in addition, emerging in user
Interesting measurement aspect, the method that the access module that user is extracted from access log file that presently, there are is recommended, is not examined
Consider the time response of user to access pages.
The content of the invention
The purpose of the embodiment of the present invention is to provide a kind of information and recommends method, can realize and be provided based on user's Web daily records
News are recommended, and can be measured according to user interest and be completed personalized recommendation.
To achieve the above object, the embodiments of the invention provide a kind of information to recommend method, including step：
The user behaviors log of real-time collecting user, the information pointed to by label engine to the user behaviors log are entered row label and taken out
Take, so as to generate the first tally set；The user includes user property, custom attribute and interest attribute；
The interest attribute of the user is obtained based on first tally set, and the use is obtained based on the user behaviors log
The custom attribute at family；
Feature extraction is carried out according to the user property of the user, interest attribute and custom attribute, used described in Dynamic Recognition
The classification at family, corresponding information is recommended to the user according to the classification of the user.
Compared with prior art, information disclosed by the invention recommends user behaviors log of the method by real-time collecting user, leads to
Cross the information that label engine points to the user behaviors log and enter row label extraction, so as to generate the first tally set, be then based on institute
The interest attribute that the first tally set obtains the user is stated, and the custom attribute of the user is obtained based on the user behaviors log,
Feature extraction, the class of user described in Dynamic Recognition are carried out further according to the user property of the user, interest attribute and custom attribute
Not, corresponding information is recommended to the user according to the classification of the user, can realizes that carrying out information based on user Web daily records pushes away
Recommend, and can be measured according to user interest and complete personalized recommendation.
As the improvement of such scheme, the information pointed to by label engine to the user behaviors log enters row label extraction,
It is specially so as to generate the first tally set：
The second tally set is collected, the granularity of each label in second tally set is identified by document subject matter generation model
Attribute；
Feature of the label in second tally set in a large amount of articles is identified, according to the feature to the behavior
The information that daily record is pointed to enters row label extraction, so as to generate the first tally set；
According to the granularity attribute of each label in second tally set, each label in first tally set is obtained
Granularity attribute.
As the improvement of such scheme, in addition to step：
The granularity attribute of label in the first tally set, recommend the information of different grain size attribute to the user.
As the improvement of such scheme, the user behaviors log of real-time collecting user is specially：
Carried out data transmission by the message-oriented middleware of high capacity with the user behaviors log of real-time collecting user.
As the improvement of such scheme, the custom attribute that the user is obtained based on the user behaviors log is specially：
Time series analysis is carried out to the timestamp of the user behaviors log, so as to obtain the custom attribute of user.
The embodiment of the present invention additionally provides a kind of information recommendation apparatus, including：
Collection module, for the user behaviors log of real-time collecting user, the user behaviors log is pointed to by label engine
Information enters row label extraction, so as to generate the first tally set；The user includes user property, custom attribute and interest attribute；
Attribute acquisition module, for obtaining the interest attribute of the user based on first tally set, and based on described
User behaviors log obtains the custom attribute of the user；
First recommending module, carry out feature for the user property according to the user, interest attribute and custom attribute and carry
Take, the classification of Dynamic Recognition user, corresponding information is recommended to the user according to the classification of the user.
Compared with prior art, the behavior that information recommendation apparatus disclosed by the invention passes through collection module real-time collecting user
Daily record, the information pointed to by label engine to the user behaviors log enter row label extraction, so as to generate the first tally set, then
The interest attribute of the user is obtained based on first tally set by attribute acquisition module, and is obtained based on the user behaviors log
The custom attribute of the user is obtained, then by the first recommending module according to the user property of the user, interest attribute and custom
Attribute carries out feature extraction, the classification of user described in Dynamic Recognition, is recommended accordingly to the user according to the classification of the user
Information, can realize based on user Web daily records carry out information recommendation, and can according to user interest measure complete personalization push away
Recommend.
As the improvement of such scheme, collection module includes：
First identification module, for collecting the second tally set, second label is identified by document subject matter generation model
Concentrate the granularity attribute of each label；
First tally set acquisition module, for identifying feature of the label in second tally set in a large amount of articles,
Enter row label extraction to the information pointed to the user behaviors log according to the feature, so as to generate the first tally set；
Second identification module, for the granularity attribute according to each label in second tally set, obtain described first
The granularity attribute of each label in tally set.
As the improvement of such scheme, in addition to:Second recommending module, the grain for the label in the first tally set
Attribute is spent, recommends the information of different grain size attribute to the user.
As the improvement of such scheme, the collection module is specifically used for carrying out data by the message-oriented middleware of high capacity
Transmission is with the user behaviors log of real-time collecting user.
As the improvement of such scheme, the attribute acquisition module is specifically used for carrying out the timestamp of the user behaviors log
Time series analysis, so as to obtain the custom attribute of user.
Brief description of the drawings
Fig. 1 is the schematic flow sheet that a kind of information recommends method in the embodiment of the present invention 1.
Fig. 2 is the schematic flow sheet that a kind of information recommends method in the embodiment of the present invention 2.
Fig. 3 is a kind of structural representation of information recommendation apparatus in the embodiment of the present invention 3.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation describes, it is clear that described embodiment is only part of the embodiment of the present invention, rather than whole embodiments.It is based on
Embodiment in the present invention, those of ordinary skill in the art are obtained every other under the premise of creative work is not made
Embodiment, belong to the scope of protection of the invention.
It is that a kind of information that the embodiment of the present invention 1 provides recommends the schematic flow sheet of method, including step referring to Fig. 1：
S1, the user behaviors log of real-time collecting user, the information pointed to by label engine to the user behaviors log enter rower
Label extract, so as to generate the first tally set；The user includes user property, custom attribute and interest attribute；
Wherein, the storage of mass users user behaviors log is carried out data transmission by the message-oriented middleware of high capacity.
S2, the interest attribute based on first tally set acquisition user, and institute is obtained based on the user behaviors log
State the custom attribute of user；
S3, feature extraction carried out according to the user property of the user, interest attribute and custom attribute, described in Dynamic Recognition
The classification of user, corresponding information is recommended to the user according to the classification of the user.
When it is implemented, the user behaviors log of first real-time collecting user, is pointed to by label engine to the user behaviors log
Information enters row label extraction, so as to generate the first tally set, is then based on the interest that first tally set obtains the user
Attribute, and based on the custom attribute of the user behaviors log acquisition user, user property, interest category further according to the user
Property and custom attribute carry out feature extraction, the classification of user described in Dynamic Recognition, according to the classification of the user to the user
Recommend corresponding information, can realize and information recommendation is carried out based on user Web daily records, and completion can be measured according to user interest
Propertyization is recommended.
In a preferred embodiment, the information pointed in step S1 by label engine to the user behaviors log enters row label
Extract, step is specifically included so as to generate the first tally set：
S11, the second tally set is collected, each label in second tally set is identified by document subject matter generation model
Granularity attribute；
Feature of the label in a large amount of articles in S12, identification second tally set, according to the feature to described
The information that user behaviors log points to enters row label extraction, so as to generate the first tally set；
S13, the granularity attribute according to each label in second tally set, obtain each mark in first tally set
The granularity attribute of label.
Wherein, granularity division is realized by LDA (topic model).
By above-mentioned steps, automatic decimation and the division of varigrained label can be realized.
It is that a kind of information that the embodiment of the present invention 2 provides recommends the schematic flow sheet of method, in embodiment 1 referring to Fig. 2
On the basis of, in addition to step：
S4, label in the first tally set granularity attribute, recommend the information of different grain size attribute to the user.
Different grain size division according to user behavior, can recommend the article of different grain size attribute, improve and recommend precision.
In a preferred embodiment, the custom attribute for obtaining the user in step S2 based on the user behaviors log is specific
For：
Time series analysis is carried out to the timestamp of the user behaviors log, so as to obtain the custom attribute of user.
It is a kind of structural representation for information recommendation apparatus that the embodiment of the present invention 3 provides referring to Fig. 3, including：
Collection module 101, for the user behaviors log of real-time collecting user, the user behaviors log is pointed to by label engine
Information enter row label extraction, so as to generate the first tally set；The user includes user property, custom attribute and interest category
Property；
Attribute acquisition module 102, for obtaining the interest attribute of the user based on first tally set, and it is based on institute
State the custom attribute that user behaviors log obtains the user；
First recommending module 103, feature is carried out for the user property according to the user, interest attribute and custom attribute
Extraction, the classification of Dynamic Recognition user, corresponding information is recommended to the user according to the classification of the user.
When it is implemented, by the user behaviors log of the real-time collecting user of collection module 101, by label engine to the row
The information pointed to for daily record is entered row label and extracted, and so as to generate the first tally set, is then based on institute by attribute acquisition module 102
The interest attribute that the first tally set obtains the user is stated, and the custom attribute of the user is obtained based on the user behaviors log,
Feature extraction is carried out according to the user property of the user, interest attribute and custom attribute by the first recommending module 103 again, moved
State identifies the classification of the user, recommends corresponding information to the user according to the classification of the user, can realize based on use
Family Web daily records carry out information recommendation, and can be measured according to user interest and complete personalized recommendation.
Preferably, collection module 101 includes：
First identification module, for collecting the second tally set, second label is identified by document subject matter generation model
Concentrate the granularity attribute of each label；
First tally set acquisition module, for identifying feature of the label in second tally set in a large amount of articles,
Enter row label extraction to the information pointed to the user behaviors log according to the feature, so as to generate the first tally set；
Second identification module, for the granularity attribute according to each label in second tally set, obtain described first
The granularity attribute of each label in tally set.
In a preferred embodiment, the information recommendation apparatus 100 also includes:Second recommending module, for according to first
The granularity attribute of label in tally set, recommend the information of different grain size attribute to the user.
In another preferred embodiment, the collection module 100 is specifically used for carrying out by the message-oriented middleware of high capacity
Data transfer is with the user behaviors log of real-time collecting user.
In a preferred embodiment, the attribute acquisition module 102 is specifically used for entering the timestamp of the user behaviors log
Row time series analysis, so as to obtain the custom attribute of user.
To sum up, the embodiments of the invention provide a kind of information to recommend behavior of the method and device by first real-time collecting user
Daily record, the information pointed to by label engine to the user behaviors log enter row label extraction, so as to generate the first tally set, then
The interest attribute of the user is obtained based on first tally set, and the custom of the user is obtained based on the user behaviors log
Attribute, feature extraction, user described in Dynamic Recognition are carried out further according to the user property of the user, interest attribute and custom attribute
Classification, corresponding information is recommended to the user according to the classification of the user, can realize and be provided based on user's Web daily records
News are recommended, and can be measured according to user interest and be completed personalized recommendation.
Described above is the preferred embodiment of the present invention, it is noted that for those skilled in the art
For, under the premise without departing from the principles of the invention, some improvements and modifications can also be made, these improvements and modifications are also considered as
Protection scope of the present invention.
Claims (10)
1. a kind of information recommends method, it is characterised in that including step：
The user behaviors log of real-time collecting user, the information pointed to by label engine to the user behaviors log enter row label extraction,
So as to generate the first tally set；The user includes user property, custom attribute and interest attribute；
The interest attribute of the user is obtained based on first tally set, and obtains the user's based on the user behaviors log
It is accustomed to attribute；
Feature extraction, user described in Dynamic Recognition are carried out according to the user property of the user, interest attribute and custom attribute
Classification, corresponding information is recommended to the user according to the classification of the user.
2. information as claimed in claim 1 recommends method, it is characterised in that the user behaviors log is pointed to by label engine
Information enter row label extraction, be specially so as to generate the first tally set：
The second tally set is collected, the granularity category of each label in second tally set is identified by document subject matter generation model
Property；
Feature of the label in second tally set in a large amount of articles is identified, according to the feature to the user behaviors log
The information of sensing enters row label extraction, so as to generate the first tally set；
According to the granularity attribute of each label in second tally set, the granularity of each label in first tally set is obtained
Attribute.
3. information as claimed in claim 2 recommends method, it is characterised in that also including step：
The granularity attribute of label in the first tally set, recommend the information of different grain size attribute to the user.
4. information as claimed in claim 1 recommends method, it is characterised in that the user behaviors log of real-time collecting user is specially：
Carried out data transmission by the message-oriented middleware of high capacity with the user behaviors log of real-time collecting user.
5. information as claimed in claim 1 recommends method, it is characterised in that obtains the user's based on the user behaviors log
Being accustomed to attribute is specially：
Time series analysis is carried out to the timestamp of the user behaviors log, so as to obtain the custom attribute of user.
A kind of 6. information recommendation apparatus, it is characterised in that including：
Collection module, for the user behaviors log of real-time collecting user, the information pointed to by label engine to the user behaviors log
Enter row label extraction, so as to generate the first tally set；The user includes user property, custom attribute and interest attribute；
Attribute acquisition module, for obtaining the interest attribute of the user based on first tally set, and it is based on the behavior
Daily record obtains the custom attribute of the user；
First recommending module, feature extraction is carried out for the user property according to the user, interest attribute and custom attribute, is moved
State identifies the classification of user, recommends corresponding information to the user according to the classification of the user.
7. information recommendation apparatus as claimed in claim 6, it is characterised in that collection module includes：
First identification module, for collecting the second tally set, identified by document subject matter generation model in second tally set
The granularity attribute of each label；
First tally set acquisition module, for identifying feature of the label in second tally set in a large amount of articles, according to
The feature enters row label extraction to the information pointed to the user behaviors log, so as to generate the first tally set；
Second identification module, for the granularity attribute according to each label in second tally set, obtain first label
Concentrate the granularity attribute of each label.
8. information recommendation apparatus as claimed in claim 7, it is characterised in that described device also includes:
Second recommending module, for the granularity attribute of the label in the first tally set, recommend different grain size to the user
The information of attribute.
9. information recommendation apparatus as claimed in claim 6, it is characterised in that the collection module is specifically used for passing through high capacity
Message-oriented middleware carry out data transmission with the user behaviors log of real-time collecting user.
10. information recommendation apparatus as claimed in claim 6, it is characterised in that the attribute acquisition module is specifically used for institute
The timestamp for stating user behaviors log carries out time series analysis, so as to obtain the custom attribute of user.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710580717.1A CN107436930A (en) | 2017-07-17 | 2017-07-17 | Information recommends method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710580717.1A CN107436930A (en) | 2017-07-17 | 2017-07-17 | Information recommends method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107436930A true CN107436930A (en) | 2017-12-05 |
Family
ID=60460292
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710580717.1A Withdrawn CN107436930A (en) | 2017-07-17 | 2017-07-17 | Information recommends method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107436930A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108133011A (en) * | 2017-12-22 | 2018-06-08 | 新奥（中国）燃气投资有限公司 | A kind of message push method and device |
| CN108269196A (en) * | 2017-12-01 | 2018-07-10 | 优视科技有限公司 | Add in the method, apparatus and computer equipment of network social association |
| CN109684566A (en) * | 2018-11-08 | 2019-04-26 | 百度在线网络技术（北京）有限公司 | Label engine implementation method, device, computer equipment and storage medium |
| CN111090815A (en) * | 2019-12-31 | 2020-05-01 | 恩亿科(北京)数据科技有限公司 | Label generation method and device |
| CN114637917A (en) * | 2022-03-28 | 2022-06-17 | 中国银行股份有限公司 | Information head bar recommendation method and device based on artificial intelligence |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110191217A1 (en) * | 2010-02-03 | 2011-08-04 | Oracle International Corporation | Approval workflow engine for services procurement timesheets, progress logs, and expenses |
| CN105976161A (en) * | 2016-04-29 | 2016-09-28 | 随身云（北京）信息技术有限公司 | Time axis-based intelligent recommendation calendar and user-based presentation method |
| CN106294730A (en) * | 2016-08-09 | 2017-01-04 | 百度在线网络技术（北京）有限公司 | The recommendation method and device of information |
-
2017
- 2017-07-17 CN CN201710580717.1A patent/CN107436930A/en not_active Withdrawn
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110191217A1 (en) * | 2010-02-03 | 2011-08-04 | Oracle International Corporation | Approval workflow engine for services procurement timesheets, progress logs, and expenses |
| CN105976161A (en) * | 2016-04-29 | 2016-09-28 | 随身云（北京）信息技术有限公司 | Time axis-based intelligent recommendation calendar and user-based presentation method |
| CN106294730A (en) * | 2016-08-09 | 2017-01-04 | 百度在线网络技术（北京）有限公司 | The recommendation method and device of information |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108269196A (en) * | 2017-12-01 | 2018-07-10 | 优视科技有限公司 | Add in the method, apparatus and computer equipment of network social association |
| CN108133011A (en) * | 2017-12-22 | 2018-06-08 | 新奥（中国）燃气投资有限公司 | A kind of message push method and device |
| CN108133011B (en) * | 2017-12-22 | 2022-05-24 | 新奥（中国）燃气投资有限公司 | Information pushing method and device |
| CN109684566A (en) * | 2018-11-08 | 2019-04-26 | 百度在线网络技术（北京）有限公司 | Label engine implementation method, device, computer equipment and storage medium |
| CN109684566B (en) * | 2018-11-08 | 2020-04-28 | 百度在线网络技术（北京）有限公司 | Label engine implementation method and device, computer equipment and storage medium |
| CN111090815A (en) * | 2019-12-31 | 2020-05-01 | 恩亿科(北京)数据科技有限公司 | Label generation method and device |
| CN114637917A (en) * | 2022-03-28 | 2022-06-17 | 中国银行股份有限公司 | Information head bar recommendation method and device based on artificial intelligence |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104239331B (en) | A kind of method and apparatus for realizing comment search engine sequence | |
| CN103678335B (en) | The method of method, apparatus and the commodity navigation of commodity sign label | |
| CN107436930A (en) | Information recommends method and device | |
| CN107944913B (en) | High-potential user purchase intention prediction method based on big data user behavior analysis | |
| CN103927309B (en) | A kind of method and device to business object markup information label | |
| CN103106259B (en) | A kind of mobile webpage content recommendation method based on situation | |
| CN107844565B (en) | Commodity searching method and device | |
| CN107894998B (en) | Video recommendation method and device | |
| US20120072466A1 (en) | Contents creating device and contents creating method | |
| CN105930429A (en) | Music recommendation method and apparatus | |
| CN110737821B (en) | Similar event query method, device, storage medium and terminal equipment | |
| CN101984437A (en) | Music resource individual recommendation method and system thereof | |
| CN101853299A (en) | Image searching result ordering method based on perceptual cognition | |
| CN106530001A (en) | Information recommending method and apparatus | |
| CN110059271A (en) | With the searching method and device of label knowledge network | |
| CN102402594A (en) | Rich media personalized recommendation method | |
| CN104317827A (en) | Picture navigation method of commodity | |
| CN116186372A (en) | A bibliographic system capable of providing personalized services | |
| CN107451216A (en) | The granularity attribute recognition methods of label and device | |
| CN106997350A (en) | A kind of method and device of data processing | |
| CN103823868A (en) | Event recognition method and event relation extraction method oriented to on-line encyclopedia | |
| CN110347922B (en) | Recommendation method, device, equipment and storage medium based on similarity | |
| Milicic et al. | Live topic generation from event streams | |
| CN114154063A (en) | Internet big data label analysis-based paired activity recommendation method | |
| KR101487871B1 (en) | Manual Auto-generating device for Crisis Management Response of Online-based. |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20171205 |