HK1154667B - A retrieving method and system applied to online transaction platform - Google Patents

Description

Retrieval method and system applied to online trading platform

Technical Field

The application relates to the field of network data processing, in particular to a retrieval method and a retrieval system applied to an online transaction platform.

Background

The online trading platform is a computer information system provided by a platform service provider for carrying out online trading, and online trading of seller users and buyer users can be realized on the online trading platform. On the online transaction platform, a buyer user can correspondingly retrieve the advertisement source of a seller user by inputting a certain keyword, and under the bidding ranking mode in the internet field, the mapping relationship between the advertisement source (such as a text file, a picture and the like) provided by the seller user and the corresponding keyword (namely a purchasing word or unit) is called matching between the advertisement source and the keyword. The seller user can achieve the purpose of advertising the advertisement source by purchasing the keyword, for example, when the seller user purchases the keyword a, the specified advertisement source (e.g. text, picture, etc.) of the seller user is shown to the buyer user when the search term input by the buyer user is mapped to the keyword a under certain rules. And the seller users can put the advertisements and bring benefits to the seller users through the matching between the keywords and the advertisement sources, such as: clicking, consulting, or generating a transaction, etc., by the buyer user is referred to as the profit of the seller user.

In the prior art, in a retrieval method applied to an online transaction platform, after a buyer user inputs a keyword, an advertisement source of a seller user is retrieved according to a certain matching mapping relationship, and the matching mapping relationship in the prior art is generated as follows: when selecting the matching between the advertisement source and the keyword, generally considering the advertisement source which is matched with the latest advertisement source preferentially and whether the advertisement source is matched with a certain keyword or not to bring the maximum profit to the advertisement source. For example, a seller user has two different advertisement sources of "sell various mobile phones" and "sell foreign mobile phones", and the keywords include three, namely "mobile phone", "motorola mobile phone" and "nokia mobile phone", and it is assumed that each advertisement source can only be associated with two keywords, and there are various associations that bring the following benefits to the seller user: the income of matching the mobile phones for selling various mobile phones is as follows: 10; the income of matching various mobile phones with the Motorola mobile phone is as follows: 7; the income of matching the associated mobile phones for selling various mobile phones is as follows: 6; the income of matching the 'selling foreign mobile phone' with the 'mobile phone' is as follows: 9; the income of matching the 'foreign mobile phone' with the 'motorola mobile phone' is as follows: 7; the income of matching the 'associating mobile phone' for selling foreign mobile phones is as follows: 5. then the matching relationship between each advertisement source and the keyword generated according to the existing method is as follows: the advertisement source 'sell various mobile phones' matching keywords 'mobile phone' and 'motorola mobile phone', the advertisement source 'sell foreign mobile phone' matching keywords 'associated mobile phone', and the profit brought to the seller user by the matching relation is 22.

However, in the above example, there is also a way to match the advertisement source with the keyword: the matching relationship between the mobile phone selling various mobile phones and the associating mobile phone and the mobile phone selling foreign mobile phones and the motorola mobile phone brings 23 profits for the seller users. It can be seen that in practical application, when the number of keywords matched with the advertisement sources of the seller users is limited, if only the advertisement source with the latest extraction time is considered when the advertisement source is extracted, or the advertisement source and the keywords are matched, the maximum benefit can be brought to the current one or one group of similar advertisement sources only according to a certain matching relation, so that the obtained matching relation between the advertisement source and the keywords cannot represent all the fields operated by the seller users, and no or few advertisement sources in some operated fields of the seller users can be matched with the keywords, so that when the buyer users input the keywords to perform commodity retrieval, the corresponding advertisement sources meeting the requirements cannot be retrieved, and if the buyer users repeatedly perform retrieval, the search performance of the online transaction platform server is certainly reduced. Further, the prior art cannot generate the matching relationship between the advertisement source and the keyword in the manner of generating the matching relationship between the advertisement source and the keyword to bring the maximum profit to the seller user.

In summary, one of the technical problems that needs to be urgently solved by those skilled in the art is: how to creatively provide a retrieval method applied to an online trading platform to solve the problem of the reduction of the searching performance of the server of the online trading platform when repeated retrieval is carried out on the online trading platform based on the matching relationship between an advertisement source and a keyword in the prior art.

Disclosure of Invention

The technical problem to be solved by the application is to provide a retrieval method, which is used for solving the problem that the search performance of the server of the online trading platform is reduced when repeated retrieval is carried out on the online trading platform based on the matching relation between the advertisement source and the keyword in the prior art.

The application also provides a retrieval device used for ensuring the realization and the application of the method in practice.

In order to solve the above problems, the present application discloses a retrieval method applied to an online transaction platform, which includes:

extracting advertisement sources of the seller users meeting preset conditions according to categories and given seller users;

generating corresponding keywords by the extracted advertisement source of the seller user;

establishing a planning model according to preset constraint conditions, wherein the planning model represents the matching relation between the advertisement source extracted to the seller user and the generated keywords;

solving the planning model by using a search-based solving method to obtain a matching relation between an advertisement source needing to be matched and a keyword;

and searching corresponding advertisement sources according to the matching relation according to the received keywords input by the buyer user.

The application also provides a method for generating the matching relationship between the advertisement source and the keywords, which comprises the following steps:

extracting advertisement sources of the seller users meeting preset conditions according to categories and given seller users;

generating corresponding keywords by the extracted advertisement source of the seller user;

establishing a planning model according to preset constraint conditions, wherein the planning model represents the matching relation between the advertisement source extracted to the seller user and the generated keywords;

and solving the planning model by using a search-based solving method to obtain a matching relation between the advertisement source needing to be matched and the keyword.

The application also provides a retrieval system applied to the online trading platform, which comprises:

the matching relation generating device is used for extracting the advertisement sources of the seller users meeting the preset conditions according to the given seller users and categories; generating corresponding keywords by the extracted advertisement source of the seller user; establishing a planning model according to preset constraint conditions, wherein the planning model represents the matching relation between the advertisement source extracted to the seller user and the generated keywords; solving the planning model by using a search-based solving method to obtain a matching relation between an advertisement source needing to be matched and a keyword;

and the searching unit is used for searching corresponding advertisement sources according to the matching relation according to the received keywords input by the buyer user.

The application also provides a device for generating the matching relationship between the advertisement source and the keywords, which comprises:

the extraction module is used for extracting the advertisement sources of the seller users meeting the preset conditions according to the given seller users and categories;

a keyword generation module for generating corresponding keywords from the extracted advertisement sources of the seller users;

a planning model establishing module for establishing a planning model according to preset constraint conditions, wherein the planning model represents the matching relation between the advertisement source extracted to the seller user and the generated keywords;

and the matching relation obtaining module is used for solving the planning model by utilizing a solving method based on search so as to obtain the matching relation between the advertisement source needing to be matched and the keyword.

Compared with the prior art, the method has the following advantages:

the method and the device for searching the advertisement source can avoid the problem that the searching speed and the performance of the server of the online transaction platform are reduced due to repeated searching under the condition that the buyer user cannot search the proper advertisement source after inputting the keyword. Of course, it is not necessary for any product to achieve all of the above-described advantages at the same time for the practice of the present application.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without inventive labor.

FIG. 1 is a flowchart of embodiment 1 of a method for generating a matching relationship between an advertisement source and a keyword according to the present application;

FIG. 2 is a flow chart of a retrieval method applied to an online trading platform according to the present application;

FIG. 3 is a flowchart of embodiment 2 of a method for generating a matching relationship between an advertisement source and a keyword according to the present application;

FIG. 4 is a schematic structural diagram of an embodiment 1 of a device for generating a matching relationship between an advertisement source and a keyword according to the present application;

FIG. 5 is a schematic structural diagram of a retrieval system applied to an online trading platform according to the present application;

fig. 6 is a schematic structural diagram of an embodiment 2 of the apparatus for generating a matching relationship between an advertisement source and a keyword according to the present application.

Detailed Description

The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.

The application is operational with numerous general purpose or special purpose computing device environments or configurations. For example: personal computers, server computers, hand-held or portable devices, tablet-type devices, multi-processor apparatus, distributed computing environments that include any of the above devices or equipment, and the like.

The application may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The application may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.

One of the main ideas of the present application may include, first, extracting a certain number of advertisement sources according to the distribution ratio of the advertisement sources of the seller users under each category, because the extraction is carried out according to the category distribution proportion of the advertisement sources, the extracted advertisement sources can distribute all commodity areas which can be traded by the seller users as much as possible, secondly, generating corresponding keywords according to the advertisement sources, adopting a method of combining establishment of a 0-1 planning model and a search-based solving mode, thereby obtaining the matching between the advertisement source needing to establish the matching relation and the corresponding keywords, the matching relation established by the method of the embodiment of the application can represent the maximum profit of the seller users, therefore, when the buyer user inputs the keyword to search the required goods, the advertisement source of the most relevant seller user can be searched according to the established matching relation. The method and the device for searching the advertisement source solve the problem that performance of a server of an online transaction platform is reduced due to repeated searching when a buyer user cannot search the appropriate advertisement source, and the embodiment of the application ensures that the most appropriate advertisement source can be obtained under the condition that the buyer user searches once, so that shopping experience of the buyer user and selling experience of a seller user are improved.

Referring to fig. 1, a flowchart of embodiment 1 of a method for generating a matching relationship between an advertisement source and a keyword according to the present application is shown, which may include the following steps:

step 101: and according to a given seller user, extracting the advertisement sources of the seller user meeting preset conditions according to categories.

In this embodiment, for all advertisement sources of the seller users, a certain rule may be adopted to classify all advertisement sources, for example: the category of the advertisement source may be generated according to a tree, and when the root node is a garment, the child nodes thereof may include: the sub-node ' men's clothing ' can also include different seasons and different styles of clothing, etc., and the categories are divided into a certain detailed degree according to the needs in practice. The method for classifying the advertisement sources belongs to the prior art, and the invention is not described in detail herein. The method for classifying the advertisement sources of the seller users can be various methods as long as the same type of advertisement sources meeting the actual requirements are in the same category.

Extracting advertisement sources of the seller users meeting preset conditions according to categories, wherein the preset conditions can be as follows: the number of the extracted advertisement sources is larger than that of the advertisement sources needing to establish the matching relationship, and the like, so that the situation that some advertisement sources cannot match the keywords is avoided. The number of the advertisement sources needing to establish the matching relationship is provided by the seller user, namely, the seller user determines the number of the advertisement sources needing to establish the matching relationship before the process of establishing the matching relationship. Specifically, under a certain category, when the advertisement source is extracted under the category, the advertisement source can be extracted according to whether the time is up to date and whether a certain advertisement source is extracted as a reference. The advertisement sources for the vendor users may be in various forms, such as: text, pictures, videos, etc., as long as they are satisfied that they can be displayed on the internet, are referred to as advertisement sources in the embodiments of the present application.

Step 102: and generating corresponding keywords according to the extracted advertisement source of the seller user. The method for generating the keywords from the advertisement source extracted in step 101 belongs to the prior art, and is not described in detail herein, wherein the method for generating the keywords from the advertisement source may use extraction of the central word or an expansion matching method when generating the keywords. It should be noted that the keyword criteria generated in this step are: the generated keywords need to be related to the advertisement source to some extent and can represent the purchasing intention of the buyer user. The fact that the keywords generated according to a certain advertisement source are related to the advertisement source indicates that the keywords should meet the requirements of the seller user, but cannot damage the experience of the buyer user, and can represent the purchasing intention of the buyer user. For example, when the advertisement source of the seller user is "sell various Nokia mobile phones", the extracted keyword may be "mobile phone", "Nokia mobile phone", or "various mobile phones", and the keyword may be a central word of a text corresponding to the advertisement source in general.

Step 103: and establishing a planning model according to preset constraint conditions, wherein the planning model represents the matching relation between the advertisement source extracted to the seller user and the generated keywords.

In this embodiment, the preset constraint conditions are generally: the number of keywords mapped by each advertisement source, and the number of advertisement sources required to establish a matching relationship. The content of the constraint condition is generally preset by a seller user, when a matching relationship is established in practical application, because the number of the advertisement sources and the number of keywords mapped by each advertisement source are limited to a certain extent, in order to establish a planning model meeting the constraint condition, a 0-1 integer variable is required to be introduced to represent whether a certain advertisement source or a certain group of advertisement sources is mapped with a certain keyword or not, the mapping is represented as 1, the non-mapping is represented as 0, and a 0-1 integer planning model aiming at maximizing the profit of the seller user is established according to the predetermined constraint condition. The seller user profit can be understood as profit brought to the seller user after a certain advertisement source is matched with a certain keyword, and the profit can be expressed by click rate brought to the seller user in an actual project. The specific building process of the integer programming model will be described in detail in the next embodiment.

It should be noted that the mapping relationship between the advertisement source provided by the seller user and the corresponding keyword is matching between the advertisement source and the keyword in the embodiment of the present application.

Step 104: and solving the planning model by using a search-based solving method to obtain a matching relation between the advertisement source needing to be matched and the keyword.

After the planning model in step 103 is established, the 0-1 integer planning model may be solved by a search-based solving method, and the solved result is the matching relationship corresponding to the advertisement source and the keyword. The total number of the advertisement sources in the matching relationship is the number of the advertisement sources which are specified by the seller user in advance and need to be matched, and the seller user only needs to specify constraint conditions in advance, namely the number of the advertisement sources which need to be matched and the number of keywords which need to be matched by each advertisement source.

The matching relation solved in the application can be repeatedly utilized, and meanwhile, the matching of one advertisement source and the keyword can also be returned to the seller user after being generated, and the seller user can determine whether the requirement is met. After the matching between the advertisement source and the keyword is established, the method can be applied to various applications, and is generally applied to that after a buyer user inputs the keyword, the online transaction platform server can search the advertisement source of a certain seller user according to the pre-generated matching relationship and return the content of the advertisement source to the buyer user.

Referring to fig. 2, on the basis of embodiment 1 of the matching relationship generation method, after step 104, the embodiment of the retrieval method applied to the online transaction platform of the present application further needs to include:

step 201: and searching corresponding advertisement sources according to the matching relation according to the received keywords input by the buyer user.

That is, in this embodiment, after the matching relationship of the seller user is generated, the matching relationship is stored at the server of the online trading platform, after the buyer user inputs a keyword, that is, a keyword of a commodity or an article that the buyer user needs to purchase, the server can search a corresponding advertisement source according to the matching relationship obtained in step 104, according to the advertisement source searched according to the matching relationship in this embodiment, the product information of the seller user can be reflected more truly, meanwhile, the buyer user can search a product that the buyer user needs more accurately, and an advertisement source meeting the buying desire of the buyer user can be searched within as few times as possible without repeatedly searching or replacing the keyword; moreover, since the advertisement sources in the matching relationship established in the embodiment of the application are extracted as the categories, the online trading platform server only needs to search according to the categories where the advertisement sources in the matching relationship are located when searching, and based on the analysis, the embodiment of the application can reduce the searching times of the online trading platform server and improve the searching performance of the online trading platform server.

Preferably, this embodiment may further include:

step 202: and displaying the corresponding advertisement source obtained by searching to the buyer user on the website page.

Meanwhile, the server can display the retrieved advertisement sources to the buyer user on a website page, so that the buyer user can click to acquire the information and the purchasing mode of the specific product.

In the embodiment, the advertisement sources of the seller users are classified and then the advertisement sources are selected according to the category distribution proportion, so that the advertisement sources can cover all fields of articles sold by the seller users, in addition, the model which is optimally matched and selected for the seller users is established according to the 0-1 integer programming in the embodiment, the model which is established is solved according to the search greedy algorithm, so that the advertisement sources which are optimal for the seller users, namely the matching between the advertisement sources and the keywords in all fields operated by the seller users can be represented, so that the relevant advertisement sources can be expected to be searched when the keywords are input by the buyer users, the advertisement sources of the seller users which can meet the requirements are obtained according to the matching relationship generated in the embodiment of the application, the repeated searching operation of the buyer users is avoided, and the searching rate and the performance of the online transaction platform server are improved, meanwhile, the shopping experience of the buyer user in searching the advertisement source is improved.

Referring to fig. 3, a flow chart of embodiment 2 of the method for generating a matching relationship between an advertisement source and a keyword is shown, which may include the following steps:

step 301: and classifying the advertisement sources of all seller users in the current online trading platform system according to preset classification criteria.

In this embodiment, firstly, the advertisement sources of all the seller users in the current online trading platform system need to be classified according to preset classification criteria, and the classified advertisement sources of various categories are stored, so that the advertisement sources to be extracted can be subsequently acquired from certain advertisement sources according to various categories.

Step 302: and acquiring the distribution proportion of all advertisement sources of the seller users under various categories according to the given seller users.

For a certain seller user needing to generate a matching relationship, firstly, the advertisement sources belonging to the seller user are obtained from all the advertisement sources, and the distribution proportion of the advertisement sources of the seller user in each category is obtained. The distribution proportion is obtained for extracting advertisement sources with corresponding proportions in each category according to the proportion. In the embodiment, the advertisement sources are extracted according to the distribution proportion of the advertisement sources of the seller users in various categories, so that the corresponding advertisement sources can be extracted from all categories of the seller users, and one or more categories of the seller users cannot be missed in the generation process of the matching relationship, so that the matching can be targeted to represent the advertisement sources of the seller users when the matching relationship is generated according to the extracted advertisement sources, and the matching probability is relatively high when the buyer users input keywords for searching subsequently, and therefore the matching relationship generated based on the embodiment of the application also meets the maximum benefit of the seller users.

Step 303: and extracting a preset number of advertisement sources under each category according to the distribution proportion, wherein the preset number is larger than the number of the advertisement sources needing to be matched.

For example, if the distribution ratio of the advertisement sources of the seller users under the categories a and B is 1: 2, then in the process of extracting the advertisement sources, the advertisement sources under the categories a and B respectively with the ratio of 1: 2 are also required. Of course, the total number of the extracted advertisement sources needs to satisfy a certain requirement, in this embodiment, the number of the extracted advertisement sources is greater than the number of the advertisement sources needing to establish matching, and in general, 2 times of the number of the advertisement sources needing to establish matching can be directly extracted. In practical application, a plurality of advertisement sources often correspond to one keyword, but in general, one keyword generally corresponds to only one advertisement source or one class of advertisement sources based on the search speed and performance consideration of the online trading platform server, so that when the advertisement sources are searched subsequently according to the keywords, all the advertisement sources do not need to be traversed, each keyword can search different advertisement sources, and the search speed of the online trading platform server can be increased. For example: the advertisement source selling various mobile phones only establishes mapping with the keyword mobile phone, the advertisement source selling mobile phones also establishes mapping with the keyword mobile phone, if the two advertisement sources are independent, the two matching relations cannot exist at the same time, and the condition that a certain advertisement source does not have the keyword and can form matching can be caused.

Step 304: establishing a matrix X, the entries X of said matrix_i，jIndicating whether the ith keyword matches the jth ad source.

After extracting that the seller user needs to establish a matched advertisement source, establishing a 0-1 integer planning model for maximizing the seller user profit. In practice, a matrix X is first established, the matrix is a matrix with n rows and m columns, and the term X of the matrix is adopted_i，jTo indicate whether the ith keyword matches the jth advertisement source, i.e. whether the ith keyword has a mapping relationship with the jth advertisement source. Wherein, i 1.. and n, j 1.. and m.

Step 305: and converting the matrix into vectors according to rows, and establishing a 0-1 integer programming model according to the converted vectors.

When converting a matrix into a vector, a second column of the matrix needs to be connected to the bottom of a first column, a third column needs to be connected to the bottom of the second column, and the two columns are connected in sequence until the matrix is converted into a vectorIs connected to the bottom of the penultimate column, the matrix is converted to a total number n^*A vector of m. The planning model built from this vector is as follows:

wherein Pij is the income brought by the matching of the ith keyword and the jth advertisement source, and Xij represents whether the ith keyword is matched with the jth advertisement source, the matching is 1, and the mismatching is 0; nj represents the number of keywords that the jth ad source can accommodate. n is the number of keywords, and m is the number of advertisement sources.

Step 306: and respectively obtaining the advertisement source income, the similarity, the advertisement source breadth and the advertisement source depth which are corresponding to each matching between each keyword and each advertisement source.

In this embodiment, the advertisement source profit, the similarity, the advertisement source breadth and the advertisement source depth corresponding to the matching between the ith keyword and the jth advertisement source on each entry in the matrix X are obtained. The advertisement source breadth represents the degree that one keyword matches the number of advertisement sources as much as possible, and the advertisement source depth represents the degree that one advertisement source matches the number of keywords as much as possible. In practical application, the income, the similarity, the advertisement source breadth and the advertisement source depth are all considered as existing numerical values in the embodiment of the application, and the similarity represents the similarity between the advertisement source and the keyword and can be calculated by utilizing the prior art; in practical application, the advertisement source width and the advertisement source depth can be calculated by adopting the following calculation formulas: the depth of the advertisement source is the number of the keywords limited by the advertisement source; the advertisement source breadth is the number of the keywords limited by the advertisement source-the number of the keywords which establish the matching relationship with the advertisement source at present. Meanwhile, in practical application, the concepts expressed above can have different calculation results according to different situations.

Step 307: and respectively calculating a plurality of seller profits brought to the seller users by the plurality of matching combinations in a plurality of matching combinations with the maximum advertisement source profits.

Firstly, all matching combinations in the vector are sorted according to the advertisement source income, a plurality of matching combinations with the highest advertisement source income in the front row are obtained, and a plurality of seller income brought to a seller user by the matching combinations are respectively calculated. The calculation method of the seller user income comprises the following steps: the maximum revenue for a match of an ad source to a keyword minus the loss to the seller user from the match. In practical applications, the loss caused by a certain match to the seller user is defined as: the matched keywords in other advertisement sources generate the maximum profit value.

It should be noted that, the calculation of the profit of the seller user may be implemented by using a valuation function, that is, a matched profit brought to the seller user is obtained by constructing the valuation function in this step, for example, the valuation function is defined as: the profit brought to the seller user by a certain match is the maximum profit of the match of a certain advertisement source and the keyword minus the loss brought to the seller user by the match.

Step 308: judging whether the matching combination with the largest seller income is multiple or not, if so, sequencing the matching combinations with the largest seller income according to the similarity, the advertisement source breadth and/or the advertisement source depth, and selecting the top matching combination from the final sequencing result; if not, the matching combination with the maximum profit of the seller is directly obtained.

In the step, if the maximum matching combination of the seller profits is only one, directly acquiring the maximum matching combination of the seller profits, and entering the next step; if the matching combination with the largest seller income is multiple, the matching combinations with the largest seller income are required to be ranked according to the similarity, and the matching combination with the largest similarity is obtained according to the ranking result. In this step, an optimal matching combination relationship is obtained according to the profit, the similarity, the advertisement source breadth and/or the advertisement source depth.

Step 309: judging whether the number of keywords matched with the advertisement source in the currently acquired matching combination reaches the upper limit number, if so, deleting the mapping relation of the matching combination, and entering the next step; if not, the matching combination is saved, and the next step is entered.

In this embodiment, when the planning model is established, that is, a constraint condition is predetermined, that is, the number of keywords that can be matched by each advertisement source, then the matching obtained in the previous step needs to be detected in this step, it is detected whether the number of keywords that are matched by the matched advertisement source has reached the upper limit number, if the number has exceeded the upper limit number, the mapping relationship of the matching combination is deleted, and the next step is performed, and if the number has not exceeded the upper limit number of the keywords, the mapping relationship between the advertisement source and the keywords indicated by the matching relationship is stored, and the matching mapping relationship needs to be deleted, so that the matching relationship is not repeatedly searched in the subsequent traversal, and the next step is performed at the same time.

Step 310: and continuing to obtain a matching combination meeting the constraint condition in other matching between each keyword and each advertisement source until the extracted advertisement sources are matched with the keywords with the upper limit number.

After the matching relationship between an advertisement source and a keyword is obtained, other matching between other keywords and each advertisement source needs to be continuously traversed in the matrix X, and the traversing and obtaining of matching modes can be the same as those in steps 306-309 until the advertisement sources extracted in step 302 are matched with the keywords with the upper limit number. For example, if each advertisement source has matched a maximum of 5 keywords, the step of obtaining matches needs to be stopped.

Step 311: and sequencing the obtained matching combinations according to the total income value of each advertisement source, and taking the matching combinations which meet the constraint conditions after sequencing as solving results.

And then, calculating a total profit value of the extracted advertisement sources, for example, when one advertisement source can match two keywords, the total profit value is the sum of profit values when the advertisement source and the two keywords are respectively matched, sequencing the obtained matching combinations according to the total profit value, and taking a certain number of matching combinations from the sequencing result as a solving result, wherein the certain number is the number of the advertisement sources needing to be matched.

Certainly, in embodiment 2 of the retrieval method for the online transaction platform of the present application, after step 311, step 201 and step 202 may be further included, so as to retrieve the advertisement source corresponding to the seller according to the matching relationship generated in embodiment 2 of the method for generating a matching relationship and the keyword input by the buyer user, and this retrieval manner may enable the displayed advertisement source to better cover the sales field of the seller user, and meanwhile, to be closer to the purchase intention of the buyer user, so that the buyer user performing a transaction on the online transaction platform does not need to repeatedly input the keyword for querying, which enables the online transaction platform server to reduce the number of retrieval times, thereby improving the retrieval performance and speed.

To better understand the present application, the following will further explain the present application with respect to an example of an actual implementation, which focuses on the building of an integer programming model and its solving process, which may include the following steps:

step A1: and according to a given seller user, extracting the advertisement sources of the seller user meeting preset conditions according to categories.

In this example, the preset conditions are that the number of the advertisement sources needing to establish matching is 3, and the number of the keywords required by each advertisement source is 2, then the number of the advertisement sources needing to be extracted in this step needs to be greater than 3, and is assumed to be 6 in this example. For the seller user a, assuming that the total number of advertisement sources owned by the seller user a is 100, the advertisement sources extracted according to the category distribution proportion are: 1,2,3,4,5,6.

Step A2: and generating corresponding keywords by using the extracted advertisement source of the seller user.

In this step, a process of generating keywords is performed for the advertisement source extracted in the previous step, and the corresponding keywords are respectively: the keywords generated by the advertisement source 1 are: a, b, c, the keywords generated by the advertisement source 2 are: c, d, e, the keywords generated by the advertisement source 3 are: d, e, the keywords generated by the advertisement source 4 are: a, d, the keywords generated by the advertisement source 5 are: a, d, e, the keywords generated by the advertisement source 6 are: a, b and d.

Step A3: establishing a matrix X, the entries X of said matrix_i，jIndicating whether the ith keyword matches the jth ad source.

Step A4: converting the matrix into vectors according to columns, and establishing a planning model as shown in the following according to the vectors:

it should be noted that the planning model is established according to the converted vector in the present application, the result of the planning model cannot be obtained by using the data solving method in the prior art, and the relatively optimized result can be calculated by using the solving method based on the search only by combining the specific matching situation between each keyword and the advertisement source in the present application, and the specific calculating process of the solving method based on the search will be described in detail below.

Step A5: and respectively obtaining the advertisement source income, the similarity, the advertisement source breadth and the advertisement source depth which are corresponding to each matching between each keyword and each advertisement source.

In this example, the profit, the similarity, the advertisement source width, and the advertisement source depth of each match between the above 6 advertisement sources and 5 keywords are respectively shown in table 1, wherein four consecutive numbers in each small square respectively represent the profit, the similarity, the advertisement source width, and the advertisement source depth of the corresponding match:

TABLE 1

Advertising source 1

Advertisement source 2

Advertisement source 3

Advertisement source 4

Advertisement source 5

Advertisement source 6

Keyword a

5、3、2、1

0、0、0、0

0、0、0、0

3、3、5、1

5、2、1、4

5、2、1、3

Keyword b

4、3、3、1

0、0、0、0

0、0、0、0

0、0、0、0

0、0、0、0

3、2、5、2

Keyword c

5、3、2、4

4、2、5、1

0、0、0、0

0、0、0、0

0、0、0、0

0、0、0、0

Keyword d

0、0、0、0

5、3、5、3

4、3、2、1

4、2、1、3

4、3、2、3

4、2、1、3

Keyword e

0、0、0、0

5、3、3、1

4、3、2、1

0、0、0、0

4、2、2、1

0、0、0、0

Step A6: and respectively calculating a plurality of seller profits brought to the seller users by the plurality of matching combinations in a plurality of matching combinations with the maximum advertisement source profits.

And sequencing the matching combinations according to the value of the advertisement source income to obtain a plurality of matching combinations with the maximum advertisement source income as follows: keyword a matches advertisement source 1: 5. 3, 2, 1, keyword a matches advertisement source 5: 5. 2, 1, 4, keyword a matches ad source 6: 5. 2, 1, 3, keyword c matches advertisement source 1: 5. 3, 2, 4, keyword d matches advertisement source 2: 5. 3, 5, 3, keyword e matches advertisement source 2: 5. 3, 3 and 1. Calculating the seller profit of each keyword for the seller user according to the matching combination, wherein the seller profit is calculated by the following method: the value of the maximum revenue for a match of an ad source to a keyword minus the value of the loss this match incurs to the seller user. Then in this step, the seller profit brought to the seller user by each matching relationship is calculated as follows:

keyword a matches advertisement source 1: 5-5 ═ 0;

keyword a matches advertisement source 5: 5-5 ═ 0;

keyword a matches advertisement source 6: 5-5 ═ 0;

keyword c matches ad source 1: 5-4 ═ 1;

keyword d matches ad source 2: 5-4 ═ 1;

keyword e matches ad source 2: 5-4 ═ 1.

Step A7: and sequencing the matching combinations with the maximum profit of the plurality of sellers in sequence according to the similarity, the advertisement source breadth and/or the advertisement source depth, and selecting the top matching combination from the final sequencing result.

And according to the sequencing of the seller profits in the last step, taking the matching combination with the maximum seller profits, and only leaving the following steps: 1. d: 2. e: 2 for 3 matches; and then sequencing according to other three values to obtain the following results in sequence: matching of keyword d with advertisement Source 2: 1. 3, 5, 3, matching 1, 3, 1 of keyword e with advertisement source 2, matching of keyword c with advertisement source 1: 1. 3, 2 and 4.

Step A8: judging whether the number of keywords matched with the advertisement source in the currently acquired matching combination reaches the upper limit number, if so, deleting the mapping relation of the matching combination, and entering the next step; if not, the matching combination is saved, and the next step is entered.

It can be seen from the result of the previous step that the first matching selection is to match the advertisement source 2 with the keyword d, and simultaneously determine whether the number of the keyword words that have been matched by the advertisement source 2 reaches 2, if not, the matching relationship can be stored, and then delete the matching relationship, and if the matching number of the advertisement source 2 reaches 2, set all the corresponding values of the keyword d and the advertisement source 2 to 0, that is, delete the matching relationship between the keyword and the advertisement source.

Step A9: and sequencing the obtained matching combinations according to the total income value of each advertisement source, and taking the matching combinations which meet the constraint conditions after sequencing as solving results.

Finally, all the obtained matching combinations of the 6 advertisement sources are sequenced according to the total profit value of the 6 advertisement sources, namely the profit sum of the 2 keywords corresponding to the 6 advertisement sources; then, the top 3 ordered advertisement sources are taken out as a solution result, and the solution result can be further returned to the seller user as a matching relation between the advertisement sources and the keywords.

Because the algorithm performance is considered in practical application, a greedy algorithm can be adopted to select the maximum matching combination of the seller profits. The specific algorithms and implementations used in actual applications may vary depending on the actual implementation. In practical applications, after the buyer user inputs the keyword, the corresponding advertisement source is retrieved according to the matching relationship generated in this example, which belongs to the prior art, so detailed retrieving process is not repeated herein. However, the basis for searching the advertisement sources is the matching relation specific to the application, so that the searched advertisement sources can also enable the buyer user to search the more accurate advertisement sources within the least times, thereby reducing the searching times of the online trading platform and improving the searching performance of the online trading platform server.

While, for purposes of simplicity of explanation, the foregoing method embodiments have been described as a series of acts or combination of acts, it will be appreciated by those skilled in the art that the present application is not limited by the order of acts or acts described, as some steps may occur in other orders or concurrently with other steps in accordance with the application. Further, those skilled in the art should also appreciate that the embodiments described in the specification are preferred embodiments and that the acts and modules referred to are not necessarily required in this application.

Referring to fig. 4, the present application further provides an embodiment 1 of a device for generating a matching relationship between an advertisement source and a keyword, where in this embodiment, the device may include:

an extracting module 401, configured to extract, according to a given seller user, an advertisement source of the seller user that meets a preset condition according to a category;

a generate keywords module 402, configured to generate corresponding keywords from the extracted advertisement sources of the seller users;

a planning model establishing module 403, configured to establish a planning model according to preset constraint conditions, where the planning model represents a matching relationship between the advertisement source extracted to the seller user and the generated keyword;

and an obtaining matching relationship module 404, configured to solve the planning model by using a search-based solution method to obtain a matching relationship between the advertisement source and the keyword that need to be matched.

Referring also to fig. 5, in an embodiment of the retrieval system applied to the online transaction platform, the system may include:

the matching relation generating device is used for extracting the advertisement sources of the seller users meeting the preset conditions according to the given seller users and categories; generating corresponding keywords by the extracted advertisement source of the seller user; establishing a planning model according to preset constraint conditions, wherein the planning model represents the matching relation between the advertisement source extracted to the seller user and the generated keywords; solving the planning model by using a search-based solving method to obtain a matching relation between an advertisement source needing to be matched and a keyword;

and the searching unit 501 is configured to search a corresponding advertisement source according to the matching relationship according to the received keyword input by the buyer user.

A display unit 502 for displaying the searched corresponding advertisement source to the buyer user on the website page.

The apparatus described in this embodiment may be integrated into a server of a search engine, or may be connected to the search engine server separately as an entity, and in addition, when the method described in this application is implemented by software, the apparatus may be used as a function newly added to the server of the search engine, or may write a corresponding program separately, and the application does not limit the implementation manner of the method or the apparatus.

Corresponding to the method provided by embodiment 2 of the method for generating a matching relationship between an advertisement source and a keyword in the present application, referring to fig. 6, the present application further provides a preferred embodiment 2 of a device for generating a matching relationship between an advertisement source and a keyword, and in this embodiment, the device may specifically include:

the classification module 601 is configured to classify advertisement sources of all seller users in the current online transaction platform system according to preset classification criteria.

The obtaining proportion sub-module 602 is configured to obtain a distribution proportion of all advertisement sources of the seller user in each category.

The extracting sub-module 603 is configured to extract a preset number of advertisement sources in each category according to the distribution ratio, where the preset number is greater than the number of the advertisement sources that need to be matched.

The establish matrix submodule 604 is configured to establish a matrix X, the entries X of which_i，jIndicating whether the ith keyword matches the jth ad source.

The converting submodule 605 is configured to convert the matrix into a vector in columns, and establish a planning model as shown below according to the vector:

wherein Q is_iWhether the keywords expressed before the ith position are matched with the advertisement source or not is represented; p_iIndicating the profit brought to the seller user by the keyword and the advertisement source matching at the ith position on the corresponding matrix. N is a radical of_jThe number of keywords that the jth advertisement source can hold is shown, m is the number of advertisement sources, and n is the number of keywords.

The parameter obtaining sub-module 606 is configured to obtain advertisement source revenue, similarity, advertisement source breadth, and advertisement source depth corresponding to each match between each keyword and each advertisement source, respectively.

The calculating sub-module 607 is configured to calculate, among the plurality of matching combinations with the highest advertisement source profit, a plurality of seller profits brought to the seller user by the plurality of matching combinations, respectively.

The first obtaining and matching sub-module 608 is configured to determine whether there are multiple matching combinations with the largest seller profit, if so, sort the multiple matching combinations with the largest seller profit in sequence according to the similarity, the advertisement source breadth and/or the advertisement source depth, and select a top matching combination from the final sorting result; if not, the matching combination with the maximum profit of the seller is directly obtained.

The judging sub-module 609 is configured to judge whether the number of the keywords matched with the advertisement source in the currently obtained matching combination exceeds the upper limit number, and if so, delete the mapping relationship of the matching combination and enter the next step; if not, the matching combination is saved, and the next step is entered.

And a second obtaining and matching sub-module 610, configured to continue to obtain, from other matches between each keyword and each advertisement source, a matching combination that meets a preset constraint condition until all the extracted advertisement sources have been matched with the upper-limit number of keywords.

And the sorting submodule 611 is configured to sort the obtained matching combinations according to the total revenue value of each advertisement source, and use the sorted matching combinations that satisfy the constraint condition as solution results.

It should be noted that, in the present specification, the embodiments are all described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments may be referred to each other. For the device-like embodiment, since it is basically similar to the method embodiment, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiment.

Finally, it should also be noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. The term "comprising", without further limitation, means that the element so defined is not excluded from the group consisting of additional identical elements in the process, method, article, or apparatus that comprises the element.

The above detailed description is provided for a retrieval method and system applied to an online transaction platform, and a specific example is applied in the detailed description to explain the principle and implementation of the present application, and the description of the above embodiment is only used to help understanding the method and core ideas of the present application; meanwhile, for a person skilled in the art, according to the idea of the present application, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present application.

Claims (10)

1. A retrieval method applied to an online transaction platform is characterized by comprising the following steps:

extracting advertisement sources of the seller users meeting preset conditions according to categories and given seller users;

generating corresponding keywords by the extracted advertisement source of the seller user;

establishing a planning model according to preset constraint conditions, wherein the planning model represents the matching relation between the advertisement source extracted to the seller user and the generated keywords;

solving the planning model by using a search-based solving method to obtain a matching relation between an advertisement source needing to be matched and a keyword;

searching corresponding advertisement sources according to the matching relation according to the received keywords input by the buyer user;

the constraint conditions comprise the number of advertisement sources needing to be matched and the upper limit number of keywords required to be matched by each advertisement source; the establishing of the planning model representing the matching relationship between each advertisement source and each keyword according to the preset constraint condition specifically includes:

establishing a matrix X, wherein an item Xij of the matrix represents whether the ith keyword is matched with the jth advertisement source or not;

and then converting the matrix into vectors according to the rows, and establishing a planning model as shown in the following according to the vectors:

wherein Pij is the matching income of the ith keyword and the jth advertisement source, and Xij is used for indicating whether the ith keyword is matched with the jth advertisement source or not; nj represents the number of keywords that the jth advertisement source can accommodate; n is the number of keywords, and m is the number of advertisement sources;

the solving of the planning model by using the search-based solving method specifically includes:

respectively obtaining advertisement source income, similarity, advertisement source breadth and advertisement source depth corresponding to each matching between each keyword and each advertisement source;

respectively calculating a plurality of seller profits brought to a seller user by a plurality of matching combinations in a plurality of matching combinations with the maximum advertisement source profits;

judging whether the matching combination with the largest seller income is multiple or not, if so, sequencing the matching combinations with the largest seller income according to the similarity, the advertisement source breadth and/or the advertisement source depth, and selecting the top matching combination from the final sequencing result; if not, directly acquiring the matching combination with the maximum profit of the seller;

judging whether the number of keywords matched with the advertisement source in the currently acquired matching combination exceeds the upper limit number, if so, deleting the mapping relation of the matching combination, and entering the next step; if not, storing the matching combination, and entering the next step;

continuing to obtain matching combinations meeting constraint conditions in other matching between each keyword and each advertisement source until the extracted advertisement sources are matched with the keywords with the upper limit number;

and sequencing the obtained matching combinations according to the total income value of each advertisement source, and taking the matching combinations which meet the constraint conditions after sequencing as solving results.

2. The method of claim 1, further comprising:

and displaying the corresponding advertisement source obtained by searching to the buyer user on the website page.

3. A method for generating a matching relationship between an advertisement source and a keyword, the method comprising:

extracting advertisement sources of the seller users meeting preset conditions according to categories and given seller users;

generating corresponding keywords by the extracted advertisement source of the seller user;

establishing a planning model according to preset constraint conditions, wherein the planning model represents the matching relation between the advertisement source extracted to the seller user and the generated keywords;

solving the planning model by using a search-based solving method to obtain a matching relation between an advertisement source needing to be matched and a keyword;

the constraint conditions comprise the number of advertisement sources needing to be matched and the upper limit number of keywords required to be matched by each advertisement source; the establishing of the planning model representing the matching relationship between each advertisement source and each keyword according to the preset constraint condition specifically includes:

establishing a matrix X, wherein an item Xij of the matrix represents whether the ith keyword is matched with the jth advertisement source or not;

and then converting the matrix into vectors according to the rows, and establishing a planning model as shown in the following according to the vectors:

wherein Pij is the matching income of the ith keyword and the jth advertisement source, and Xij is used for indicating whether the ith keyword is matched with the jth advertisement source or not; nj represents the number of keywords that the jth advertisement source can accommodate; n is the number of keywords, and m is the number of advertisement sources;

the solving of the planning model by using the search-based solving method specifically includes:

respectively obtaining advertisement source income, similarity, advertisement source breadth and advertisement source depth corresponding to each matching between each keyword and each advertisement source;

respectively calculating a plurality of seller profits brought to a seller user by a plurality of matching combinations in a plurality of matching combinations with the maximum advertisement source profits;

judging whether the matching combination with the largest seller income is multiple or not, if so, sequencing the matching combinations with the largest seller income according to the similarity, the advertisement source breadth and/or the advertisement source depth, and selecting the top matching combination from the final sequencing result; if not, directly acquiring the matching combination with the maximum profit of the seller;

judging whether the number of keywords matched with the advertisement source in the currently acquired matching combination exceeds the upper limit number, if so, deleting the mapping relation of the matching combination, and entering the next step; if not, storing the matching combination, and entering the next step;

continuing to obtain matching combinations meeting constraint conditions in other matching between each keyword and each advertisement source until the extracted advertisement sources are matched with the keywords with the upper limit number;

and sequencing the obtained matching combinations according to the total income value of each advertisement source, and taking the matching combinations which meet the constraint conditions after sequencing as solving results.

4. The method of claim 3, further comprising:

and classifying the advertisement sources of all seller users in the current online trading platform system according to preset classification criteria.

5. The method as claimed in claim 3, wherein said extracting advertisement sources of the seller users satisfying preset conditions by category includes:

acquiring the distribution proportion of all advertisement sources of the seller user in each category;

and extracting a preset number of advertisement sources under each category according to the distribution proportion, wherein the preset number is larger than the number of the advertisement sources needing to be matched.

6. A search system for use with an online trading platform, the system comprising:

a matching relationship generating device for extracting the preset satisfaction according to the category according to the given seller user

An advertisement source for the seller user of the condition; generating corresponding keywords by the extracted advertisement source of the seller user; establishing a planning model according to preset constraint conditions, wherein the planning model represents the matching relation between the advertisement source extracted to the seller user and the generated keywords; solving the planning model by using a search-based solving method to obtain a matching relation between the advertisement sources needing to be matched and the keywords, wherein the constraint conditions comprise the number of the advertisement sources needing to be matched and the upper limit number of the keywords required to be matched by each advertisement source; the establishing of the planning model representing the matching relationship between each advertisement source and each keyword according to the preset constraint condition specifically includes:

establishing a matrix X, wherein an item Xij of the matrix represents whether the ith keyword is matched with the jth advertisement source or not;

and then converting the matrix into vectors according to the rows, and establishing a planning model as shown in the following according to the vectors:

wherein Pij is the matching income of the ith keyword and the jth advertisement source, and Xij is used for indicating whether the ith keyword is matched with the jth advertisement source or not; nj represents the number of keywords that the jth advertisement source can accommodate; n is the number of keywords, and m is the number of advertisement sources;

the solving of the planning model by using the search-based solving method specifically includes:

respectively obtaining advertisement source income, similarity, advertisement source breadth and advertisement source depth corresponding to each matching between each keyword and each advertisement source;

respectively calculating a plurality of seller profits brought to a seller user by a plurality of matching combinations in a plurality of matching combinations with the maximum advertisement source profits;

judging whether the matching combination with the largest seller income is multiple or not, if so, sequencing the matching combinations with the largest seller income according to the similarity, the advertisement source breadth and/or the advertisement source depth, and selecting the top matching combination from the final sequencing result; if not, directly acquiring the matching combination with the maximum profit of the seller;

judging whether the number of keywords matched with the advertisement source in the currently acquired matching combination exceeds the upper limit number, if so, deleting the mapping relation of the matching combination, and entering the next step; if not, storing the matching combination, and entering the next step;

continuing to obtain matching combinations meeting constraint conditions in other matching between each keyword and each advertisement source until the extracted advertisement sources are matched with the keywords with the upper limit number;

sequencing the obtained matching combinations according to the total income value of each advertisement source, and taking the matching combinations which meet the constraint conditions after sequencing as solving results;

and the searching unit is used for searching corresponding advertisement sources according to the matching relation according to the received keywords input by the buyer user.

7. The system of claim 6, further comprising:

and the display unit is used for displaying the searched corresponding advertisement source to the buyer user on the website page.

8. An apparatus for generating a matching relationship between an advertisement source and a keyword, the apparatus comprising:

the extraction module is used for extracting the advertisement sources of the seller users meeting the preset conditions according to the given seller users and categories;

a keyword generation module for generating corresponding keywords from the extracted advertisement sources of the seller users;

a planning model establishing module for establishing a planning model according to preset constraint conditions, wherein the planning model represents the matching relation between the advertisement source extracted to the seller user and the generated keywords;

the matching relation obtaining module is used for solving the planning model by utilizing a solving method based on search so as to obtain the matching relation between the advertisement source needing to be matched and the keyword;

wherein, the module for establishing the planning model specifically comprises:

establishing a matrix submodule for establishing a matrix X, the entries X of said matrix_i，jIndicating whether the ith keyword is matched with the jth advertisement source;

the conversion submodule is used for converting the matrix into vectors according to columns and establishing a planning model shown as the following according to the vectors:

wherein Pij is the matching income of the ith keyword and the jth advertisement source, and Xij is used for indicating whether the ith keyword is matched with the jth advertisement source or not; nj represents the number of keywords that the jth advertisement source can accommodate; n is the number of keywords, and m is the number of advertisement sources;

the matching relationship obtaining module specifically includes:

the parameter obtaining sub-module is used for respectively obtaining advertisement source income, similarity, advertisement source breadth and advertisement source depth which are corresponding to each matching between each keyword and each advertisement source;

the calculation sub-module is used for respectively calculating a plurality of seller profits brought to the seller users by the plurality of matching combinations in the plurality of matching combinations with the maximum advertisement source profits;

the first obtaining matching submodule is used for judging whether the matching combination with the largest seller income is multiple or not, if so, sequencing the matching combinations with the largest seller income according to the similarity, the advertisement source breadth and/or the advertisement source depth, and selecting the top matching combination from the final sequencing result; if not, directly acquiring the matching combination with the maximum profit of the seller;

the judging submodule is used for judging whether the number of the keywords matched with the advertisement source in the currently acquired matching combination exceeds the upper limit number, if so, deleting the mapping relation of the matching combination, and entering the next step; if not, storing the matching combination, and entering the next step;

the second obtaining and matching sub-module is used for continuously obtaining matching combinations meeting preset constraint conditions in other matching between each keyword and each advertisement source until the extracted advertisement sources are matched with the keywords with the upper limit number;

and the sequencing submodule is used for sequencing the obtained matching combinations according to the total income value of each advertisement source and taking the matching combinations which meet the constraint conditions after sequencing as solving results.

9. The apparatus of claim 8, further comprising:

and the classification module is used for classifying the advertisement sources of all the seller users in the current online transaction platform system according to preset classification standards.

10. The apparatus according to claim 8, wherein the extraction module specifically comprises:

the obtaining proportion sub-module is used for obtaining the distribution proportion of all the advertisement sources of the seller users under various categories;

and the extraction submodule is used for extracting a preset number of advertisement sources under each category according to the distribution proportion, and the preset number is larger than the number of the advertisement sources needing to be matched.