A method and device for information recommendation

Abstract

The present invention discloses a method and device for information recommendation, and relates to the field of computer technology. A specific implementation of the method includes: determining the effective tag of the user based on the user tag reflecting the user portrait and the recommended tag reflecting the information to be recommended; and selecting a set number of target tags from the effective tags by calculating the recommended position tag value of the user tag corresponding to the recommended information position, and then determining the information to be recommended displayed to the user according to the target tag; overcoming the problem of low accuracy in determining the information to be recommended based only on the user tag, improving the accuracy of information recommendation, and then improving the personalization degree and input-output ratio of information recommendation.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a method and device for information recommendation. Background Art

[0002] User recommendation information has been widely used in Internet applications. Currently, the main method for recommending information to users is to obtain multiple user main tags contained in user feature information and recommend information to users based on the user main tags.

[0003] Due to the large number of user tags, the main user tags based on the existing information recommendation methods cannot accurately express the actual needs of users, or have a low match with the targeted needs of information recommendation, resulting in low accuracy of information recommendation. Summary of the invention

[0004] In view of this, an embodiment of the present invention provides a method and device for information recommendation, which can determine the user's effective tags based on the user tags that reflect the user portrait and the recommended tags that reflect the information to be recommended; and by calculating the recommended position tag value of the user tag corresponding to the recommended information position, a set number of target tags are selected from the effective tags, and then the information to be recommended displayed to the user is determined according to the target tags; thereby overcoming the problem of low accuracy in determining the information to be recommended based only on the user tags, improving the accuracy of information recommendation, and then improving the personalization level and input-output ratio of information recommendation.

[0005] To achieve the above-mentioned purpose, according to an aspect of an embodiment of the present invention, a method for information recommendation is provided, characterized in that it includes: in response to receiving a page acquisition request sent by a client, providing a page corresponding to the page acquisition request to the client, wherein the page acquisition request includes user information and the page includes an information recommendation position; determining a user tag corresponding to the user information based on the user information; selecting a valid tag from the user tags based on pre-stored information to be recommended and the recommended tag corresponding to the information to be recommended; based on the valid tag, selecting target recommended information from the stored information to be recommended, and sending the target recommended information to the information recommendation position, so that the client displays the target recommended information through the information recommendation position.

[0006] Optionally, the information recommendation method is characterized by:

[0007] Obtain user tags corresponding to multiple users who browse the information recommendation position; calculate the recommendation position tag value of the user tag based on the recommendation cost of multiple recommended information corresponding to the information recommendation position; select target recommended information from the stored information to be recommended based on the valid tag, including: selecting a set number of target tags for the information recommendation position from the valid tags according to the recommendation position tag value of the user tag; and determining the target recommended information according to the target tag.

[0008] Optionally, the information recommendation method is characterized by:

[0009] Determine a user tag corresponding to the user information based on the user information, including: obtaining a user tag corresponding to the user identifier included in the user information from a user feature data source based on the user identifier included in the user information, wherein the user feature data source includes a user tag generated by any one or more of user portrait data and user behavior data.

[0010] Optionally, the method for information recommendation is characterized in that calculating the recommendation position label value of the user tag includes: for each of the user tags corresponding to multiple users browsing the information recommendation position, executing: determining at least one target recommended information corresponding to the user tag from multiple recommended information corresponding to the information recommendation position; and counting the recommendation position label value of the user tag based on the recommendation cost of at least one of the target recommended information.

[0011] Optionally, the method for information recommendation is characterized in that calculating the recommendation position label value of the user tag includes: for each of the user tags corresponding to multiple users who browse the information recommendation position, executing: for the information recommendation position, counting the number of clicks and views of the user tag; and calculating the recommendation position label value of the user tag based on the number of clicks and views of the user tag.

[0012] Optionally, the method for information recommendation is characterized in that a set number of target tags are selected from the valid tags according to the recommendation position label value of the user tag, including: determining all first user tags with the recommendation position label value corresponding to the information recommendation position; selecting a second user tag matching the first user tag from the valid tags; when the number of the second user tags is greater than a set number, selecting a set number of target tags from the second user tags according to the sorting of the recommendation position label values ​​of the second user tags.

[0013] Optionally, the information recommendation method is characterized in that it further includes: when the number of the second user tags is less than a set number, using the second user tags as the target tags, and randomly selecting the remaining number of target tags from the valid tags, wherein the sum of the number of the second user tags and the remaining number is the set number.

[0014] Optionally, the information recommendation method is characterized in that, according to the recommendation position tag value of the user tag, a set number of target tags are selected from the valid tags, including:

[0015] According to the recommendation position label value of the user label, multiple third user labels corresponding to the information recommendation position are sorted; for the third user labels with recommendation position label values ​​from high to low in the sorting result, the following steps are executed in a loop until the number of target labels reaches the set number or all third user labels in the sorting result have completed the following steps, and the loop ends: determine whether the third user label belongs to the valid label, and if so, determine that the third user label is the target label.

[0016] Optionally, the information recommendation method is characterized by:

[0017] The recommendation position label value of any user label corresponding to the information recommendation position includes: a recommendation cost statistical value and a click-through rate value; sorting multiple third user labels corresponding to the information recommendation position includes: sorting the user labels corresponding to the information recommendation position based on the order of the recommendation cost statistical values ​​from high to low; when there are multiple user labels with the same recommendation cost statistical value, sorting the multiple user labels with the same recommendation cost statistical value based on the order of the click-through rate values ​​from high to low.

[0018] To achieve the above object, according to a second aspect of an embodiment of the present invention, there is provided an information recommendation device, characterized in that it comprises: a module for receiving a page acquisition request, a module for determining a valid tag, and a module for determining a target recommended information; wherein:

[0019] The page acquisition request receiving module is used to provide a page corresponding to the page acquisition request to the client in response to receiving a page acquisition request sent by the client, wherein the page acquisition request includes user information and the page includes an information recommendation position;

[0020] The valid tag determination module is used to determine the user tag corresponding to the user information according to the user information; and select a valid tag from the user tags according to the pre-stored information to be recommended and the recommended tags corresponding to the information to be recommended;

[0021] The module for determining target recommended information is used to select target recommended information from the stored information to be recommended based on the valid tag, and send the target recommended information to the information recommendation position, so that the client can display the target recommended information through the information recommendation position.

[0022] Preferably, the module for determining the target recommended information is used to obtain user tags corresponding to multiple users who browse the information recommendation position; calculate the recommendation position label value of the user tag based on the recommendation cost of multiple recommended information corresponding to the information recommendation position; select target recommended information from the stored information to be recommended based on the valid label, including: selecting a set number of target tags for the information recommendation position from the valid labels according to the recommendation position label value of the user tag; and determining the target recommended information according to the target label.

[0023] Optionally, the module for determining the target recommended information is also used to execute, for each of the user tags corresponding to multiple users browsing the information recommendation position: determining at least one target recommended information corresponding to the user tag from multiple recommended information corresponding to the information recommendation position; and counting the recommended position tag value of the user tag based on the recommendation cost of at least one of the target recommended information.

[0024] Optionally, the module for determining the target recommended information is also used to calculate the recommendation position label value of the user tag, including: for each of the user tags corresponding to multiple users who browse the information recommendation position, executing: for the information recommendation position, counting the number of clicks and views of the user tag; based on the number of clicks and views of the user tag, calculating the recommendation position label value of the user tag.

[0025] Optionally, the module for determining target recommended information is also used to select a set number of target tags from the valid tags according to the recommendation position label value of the user tag, including: determining all first user tags with the recommendation position label value corresponding to the information recommendation position; selecting a second user tag matching the first user tag from the valid tags; when the number of the second user tags is greater than a set number, selecting a set number of target tags from the second user tags according to the sorting of the recommendation position label values ​​of the second user tags.

[0026] Optionally, the module for determining the target recommendation information is also used to, when the number of the second user tags is less than a set number, use the second user tags as the target tags and randomly select the remaining number of target tags from the valid tags, wherein the sum of the number of the second user tags and the remaining number is the set number.

[0027] Optionally, the module for determining target recommended information is also used to select a set number of target tags from the valid tags according to the recommendation position label value of the user tag, including: sorting multiple third user tags corresponding to the information recommendation position according to the recommendation position label value of the user tag; for the third user tags with recommendation position label values ​​from high to low in the sorting results, looping the following steps until the number of target tags reaches the set number or all third user tags in the sorting results have completed the following steps, ending the loop: determining whether the third user tag belongs to the valid tag, and if so, determining that the third user tag is the target tag.

[0028] Optionally, the module for determining the target recommended information is also used for the recommendation position label value of any user label corresponding to the information recommendation position, including: a recommendation cost statistical value and a click-through rate value; sorting multiple third user labels corresponding to the information recommendation position, including: sorting the user labels corresponding to the information recommendation position based on the order of the recommendation cost statistical values ​​from high to low; when there are multiple user labels with the same recommendation cost statistical value, sorting the multiple user labels with the same recommendation cost statistical value based on the order of the click-through rate values ​​from high to low.

[0029] To achieve the above-mentioned purpose, according to the third aspect of an embodiment of the present invention, there is provided an electronic device for information recommendation, characterized in that it comprises: one or more processors; a storage device for storing one or more programs, and when the one or more programs are executed by the one or more processors, the one or more processors implement any of the methods described in the above-mentioned information recommendation methods.

[0030] To achieve the above-mentioned purpose, according to a fourth aspect of an embodiment of the present invention, a computer-readable medium is provided, on which a computer program is stored, characterized in that when the program is executed by a processor, any method described in the above-mentioned information recommendation method is implemented.

[0031] An embodiment of the above invention has the following advantages or beneficial effects: it can determine the user's effective tags based on the user tags that reflect the user portrait and the recommended tags that reflect the information to be recommended; and by calculating the recommended position tag value of the user tag corresponding to the recommended information position, a set number of target tags are selected from the effective tags, and then the recommended information to be displayed to the user is determined according to the target tags; it overcomes the problem of low accuracy in determining the information to be recommended based only on user tags, improves the accuracy of information recommendation, and then improves the personalization level and input-output ratio of information recommendation.

[0032] The further effects of the above-mentioned non-conventional optional manner will be described below in conjunction with the specific implementation manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The accompanying drawings are used to better understand the present invention and do not constitute an improper limitation of the present invention.

[0034] Figure 1 It is a flowchart of an existing information recommendation method provided by an embodiment of the present invention;

[0035] Figure 2 is a schematic diagram of a process of information recommendation provided by the first embodiment of the present invention;

[0036] Figure 3 is a schematic diagram of a process of information recommendation provided by the second embodiment of the present invention;

[0037] Figure 4 is a structural schematic diagram of an information recommendation device provided by an embodiment of the present invention;

[0038] Figure 5 is an exemplary system architecture diagram to which embodiments of the present invention may be applied;

[0039] Figure 6 It is a schematic diagram of the structure of a computer system of a terminal device or a server suitable for implementing an embodiment of the present invention. DETAILED DESCRIPTION

[0040] The following is a description of exemplary embodiments of the present invention in conjunction with the accompanying drawings, including various details of the embodiments of the present invention to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0041] User recommendation information has been widely used in Internet applications. One embodiment of the user recommendation information is advertisements recommended to users. The method of the present invention is described below using advertisements as an example.

[0042] At present, Internet application advertising providers have higher and higher requirements for accurate and personalized advertising push. User tag targeting technology is one of the key technologies to achieve personalized advertising push. Figure 1As shown in the schematic diagram, the data processing platform described in step S101 receives user portraits or behavior data, analyzes the user portrait data, user behavior data, etc. of each user to generate user tags, that is, by using big data and artificial intelligence technology, through the collection, extraction, cleaning and analysis of user portrait data and user behavior data, user tags are generated, and the information service module described in step S102 selects a set number of user tags from the user tags sorted based on time: that is, reverse sorting is performed based on the occurrence time corresponding to the user tags, and a set number of user tags are selected from the reverse sorted user tags according to the performance of the system and the number of advertisements; the advertising delivery system described in step S103 determines the corresponding advertisement according to the selected user tags; however, according to the reverse sorting of the occurrence time corresponding to the user tags and selecting a set number of user tags, higher-value user tags may not be selected, so that the selected advertisement set does not meet the requirements of the optimal input-output ratio; and when the recommended tags provided by the advertisement provider are not included in the selected user tags, the accuracy of the selected advertisement is reduced.

[0043] In view of this, if Figure 2 As shown, an embodiment of the present invention provides a method for information recommendation, which may include the following steps:

[0044] Step S201: in response to receiving a page acquisition request sent by a client, providing a page corresponding to the page acquisition request to the client, wherein the page acquisition request includes user information and the page includes an information recommendation position;

[0045] Specifically, the user sends a page acquisition request through the client, wherein the page is, for example, a page of an e-commerce mall, a page of a news application, a page of a social application, etc. Furthermore, the page contains an information recommendation position, that is, a position for displaying and recommending information, such as an advertisement position in a page. An application may contain multiple advertisement positions, and multiple advertisement positions may exist on the same page or on different pages. Therefore, when the page acquisition request sent by the user contains the identifier of the information recommendation position, the server returns information corresponding to the identifier of the information recommendation position (for example, an advertisement).

[0046] Step S202: determining a user tag corresponding to the user information according to the user information; and selecting a valid tag from the user tags according to pre-stored information to be recommended and the recommended tags corresponding to the information to be recommended.

[0047] Specifically, the corresponding user tag is determined according to the user information (e.g., user ID), wherein the user tag is a user tag corresponding to the user ID included in the user information obtained from the user feature data source, and the user feature data source contains user tags generated by any one or more of the user portrait data and user behavior data. For example, the user tag of user abc is associated with the user portrait data, such as female, white-collar worker, mother, master, etc., and is also associated with the user behavior data, such as browsing, purchasing, renting, mobile phone, children's books, etc.; it can be understood that a user can correspond to 1,000 tags or 100 tags, and as the user portrait is segmented, the number of user tags corresponding to the user also increases.

[0048] Furthermore, according to the pre-stored information to be recommended and the recommendation tags corresponding to the information to be recommended, valid tags are selected from the user tags; specifically, the information to be recommended is the information to be displayed. Taking advertisements as an example, the advertisement data source can pre-store various advertisements to be displayed (i.e., information to be recommended) provided by the advertisement provider. For each advertisement, the advertisement provider can set multiple recommendation tags. For example, the recommendation tags set by the provider of a mobile phone advertisement (i.e., information to be recommended) include: mobile phone, white-collar worker, young people, ABC (brand name), etc.; then, according to the recommendation tags, valid tags are selected from the user tags, for example: User abc corresponds to 1000 user tags, and there are 500 recommended tags corresponding to the pre-stored information to be recommended. Then, according to the recommended tags, the user tags with low correlation with the recommended tags are filtered out, and the remaining user tags are valid tags. For example, if the user tags include yachts, and the recommended tags do not include yachts, the user tag "yacht" is filtered out, that is, valid tags are selected from the user tags. It can be understood that by selecting valid tags associated with recommended tags, the accuracy of pushing advertisements to user abc can be improved, thereby overcoming the defect of determining advertisements only based on user tags in the prior art.

[0049] Furthermore, the pre-stored information to be recommended is updated in real time. When an advertisement provider adds advertisements, removes advertisements, adds recommendation tags, or deletes recommendation tags, the information to be recommended will be changed accordingly.

[0050] Step S203: based on the valid tag, select target recommended information from the stored information to be recommended, and send the target recommended information to the information recommendation position, so that the client displays the target recommended information through the information recommendation position.

[0051] Specifically, the target recommendation information is the recommendation information selected according to the valid tags corresponding to the user, and is used to be displayed on the client page of the user. It can be understood that the user sends a page acquisition request containing information recommendation position information, and the target recommendation information is displayed in the corresponding information recommendation position, that is, the target recommendation information is sent to the information recommendation position so that the client displays the target recommendation information through the information recommendation position. For example: based on the 500 valid tags of user aaa, and advertising recommendation position 1 (that is, information recommendation position), the advertising delivery system selects advertising 1 for user aaa (advertising 1 is the target recommendation information), then sends the information of advertising 1 to the client of user aaa, and the client displays advertising 1 in advertising recommendation position 1.

[0052] like Figure 3 As shown, an embodiment of the present invention provides a method for recommending information, which may include the following steps:

[0053] Step S301: selecting a valid tag from the user tags according to the pre-stored information to be recommended and the recommendation tags corresponding to the information to be recommended.

[0054] Specifically, the description of selecting a valid tag is consistent with the description of step S201 and will not be repeated here.

[0055] Step S302: According to the recommendation position tag value of the user tag, a set number of target tags are selected from the valid tags for the information recommendation position.

[0056] Preferably, based on the valid tag, target recommended information is selected from the stored information to be recommended, including: according to the recommendation position tag value of the user tag, a set number of target tags are selected from the valid tags for the information recommendation position; according to the target tag, the target recommended information is determined. The method comprises: obtaining user tags corresponding to multiple users who browse the information recommendation position; calculating the recommendation position tag value of the user tag based on the recommendation cost of multiple recommended information corresponding to the information recommendation position; specifically, when the user has a large number of valid tags, data query based on the valid tags imposes a large burden on system performance. Therefore, preferably, from multiple valid tags, a set number of tags are selected as target tags based on the tag value of the user tag and the recommendation cost, for example: 200 (i.e., a set number) of valid tags are selected as target tags from 500 valid tags; wherein the number of target tags corresponding to each information recommendation position is different, i.e., the set number is set for the information recommendation position; wherein the recommendation cost is obtained from the log data center of multiple recommended information (e.g., already displayed advertisements) corresponding to the information recommendation position, the recommended information position is information (e.g., advertisements) that has been displayed in the information recommendation position, and the recommendation position tag value is based on the user tag browsing the information recommendation position and the recommended information displayed in the information recommendation position. Statistics and calculation. For example: data is obtained from the log and then statistics are performed. The log can be a data table in the database, for example, including: a user request table for recording user information, information recommendation positions, etc.; an information display table for recording the display data of recommended information; an information click table for recording the data of recommended information being clicked; the log can be obtained and updated based on a set time to obtain user information and recommended information of the information recommendation position in real time.

[0057] Furthermore, there are two methods for calculating the recommended position label value:

[0058] The first method: for each of the user tags corresponding to multiple users browsing the information recommendation position, execute: determining at least one target recommended information corresponding to the user tag from multiple recommended information corresponding to the information recommendation position; and counting the recommendation position tag value of the user tag based on the recommendation cost of at least one of the target recommended information.

[0059] Specifically, the recommendation cost is taken as CPM (Cost Per Mille, CPM). The recommendation cost is the CPM of the recommended information. Still taking advertisement as an example: for example, there are 50 advertisements corresponding to the advertisement position (i.e., the information recommendation position), and the CPM of each advertisement may be different. The advertisement provider sets the CPM (i.e., recommendation cost) according to the content and number of the advertisements. Furthermore, after statistics, there are 1000 users browsing the advertisement position. Assume that the number of user tags corresponding to the users is a number in [1,1000], and the user tags of each user may be consistent, such as: female, white-collar worker, mobile phone, etc. Further, the statistical value of CPM (e.g., average value, percentile value, median value, etc.) is calculated for the user tag corresponding to each advertisement position. For example, the statistical value takes the average value as an example, and the calculation formula is shown in formula (1):

[0060] CPM(tag)=Average(act_price) (1)

[0061] Among them, tag is the user tag, act_price is the recommendation cost, and the recommendation position tag value of each user tag (such as the CPM average) is calculated based on this; for example: the recommendation cost of advertisement 1 corresponding to user tag 1 is 10 yuan, the recommendation cost of advertisement 2 corresponding to user tag 1 is 20 yuan, and the recommendation cost of advertisement 3 corresponding to user tag 1 is 15 yuan, then the average value corresponding to user tag 1 is calculated to be 15; by calculating the recommendation position tag value of user tags, it is helpful to determine user tags with higher values, so as to improve the input-output ratio of recommendation information of advertising providers.

[0062] The second method: for each of the user tags corresponding to multiple users who browse the information recommendation position, perform: for the information recommendation position, count the number of clicks and views of the user tag; based on the number of clicks and views of the user tag, calculate the recommendation position tag value of the user tag.

[0063] Specifically, still taking advertisement as an example, the value of the recommendation position label can be calculated based on the CTR of the advertisement position. CTR (Click-Through Rate, CTR = number of advertisement clicks / number of advertisement displays) is calculated as shown in formula (2):

[0064] CTR(tag)=Count(click tag ) / Count(impression tag ) (2)

[0065] That is, for the information recommendation position, the number of clicks and views of the user tag are counted; based on the number of clicks and views of the user tag, the recommendation position label value of the user tag is calculated; by calculating the recommendation position label value of the user tag, it is helpful to determine user tags with higher values ​​to improve the investment-output ratio of the advertising provider.

[0066] Preferably, according to the recommendation position tag value of the user tag, a set number of target tags are selected from the valid tags. There are two methods:

[0067] The first method: determining all first user tags corresponding to the information recommendation position and having the recommendation position tag value; selecting a second user tag matching the first user tag from the valid tags;

[0068] 1) When the number of the second user tags is greater than the set number, a set number of target tags are selected from the second user tags according to the sorting of the recommended position tag values ​​of the second user tags. For example: user aaa has 500 valid tags. Based on the performance of the system and the number of advertisements, the target tags are determined to be 200, and 200 is the set number. When 200 target tags are selected from 500 valid tags, they are determined according to user tags with recommended position tag values ​​(there may be user tags without recommended position tag values). It can be understood that user tags with higher values ​​can be selected in this way; that is, all first user tags with the recommended position tag values ​​corresponding to the information recommendation position are determined; the second user tags matching the first user tags are selected from the valid tags; when the number of the second user tags is greater than the set number (for example: 200), according to the sorting of the recommended position tag values ​​of the second user tags, for example: sorting from high to low based on the CPM value, when the CPMs are equal, sorting based on the CTR, a set number of target tags are selected from the second user tags; for example: 200 target tags matching the valid tags are selected from high to low.

[0069] 2) When the number of the second user tags (for example, 100) is less than the set number (for example, 200), the second user tag is used as the target tag, and the remaining number of target tags is randomly selected from the valid tags (that is, 100 are randomly selected from the valid tags so that the number of target tags is the set number), wherein the sum of the number of the second user tags and the remaining number is the set number.

[0070] The second method: sort the multiple third user tags corresponding to the information recommendation position according to the recommendation position label value of the user tag; for the third user tags with the recommendation position label value from high to low in the sorting result, loop the following steps until the number of the target tags reaches the set number or all the third user tags in the sorting result have completed the following steps, and end the loop: determine whether the third user tag belongs to the valid tag, and if so, determine that the third user tag is the target tag. Specifically, sort each user tag (i.e., the third user tag) based on the recommendation position label value, obtain the third user tag from high to low, match the valid tag, and if a matching valid tag is found, use the third user tag as the target tag; loop until the set number of target tags are obtained. Among them, the method of sorting according to the recommendation position label value is: the recommendation position label value of any user label corresponding to the information recommendation position includes: the recommendation cost statistical value and the click-through rate value; sorting the multiple third user labels corresponding to the information recommendation position includes: sorting the user labels corresponding to the information recommendation position based on the recommendation cost statistical value from high to low; when there are multiple user labels with the same recommendation cost statistical value, sorting the multiple user labels with the same recommendation cost statistical value based on the click-through rate value from high to low. For example: for each third user label, sort from high to low based on the CPM value (that is, the recommendation cost statistical value), and when the CPM is equal, sort from high to low based on the CTR (that is, the click-through rate value).

[0071] Step S303: Determine the target recommendation information according to the target tag.

[0072] Specifically, the target recommended information is selected from the stored information to be recommended, and the target recommended information is sent to the information recommendation position, so that the client displays the target recommended information through the information recommendation position. The description is consistent with step S203 and will not be repeated here.

[0073] like Figure 4 As shown, the embodiment of the present invention provides an information recommendation device 400, including: a page acquisition request receiving module 401, a valid tag determining module 402 and a target recommended information determining module 403; wherein,

[0074] The page acquisition request receiving module 401 is used to provide a page corresponding to the page acquisition request to the client in response to receiving a page acquisition request sent by the client, wherein the page acquisition request includes user information and the page includes an information recommendation position;

[0075] The valid tag determination module 402 is used to determine the user tag corresponding to the user information according to the user information; select a valid tag from the user tags according to the pre-stored information to be recommended and the recommended tags corresponding to the information to be recommended;

[0076] The target recommendation information determination module 403 is used to select target recommendation information from the stored information to be recommended based on the valid tag, and send the target recommendation information to the information recommendation position, so that the client can display the target recommendation information through the information recommendation position.

[0077] An embodiment of the present invention also provides an electronic device for information recommendation, comprising: one or more processors; a storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the method provided in any of the above embodiments.

[0078] An embodiment of the present invention further provides a computer-readable medium on which a computer program is stored. When the program is executed by a processor, the method provided in any of the above embodiments is implemented.

[0079] Figure 5 An exemplary system architecture 500 is shown to which the method or apparatus for information recommendation according to an embodiment of the present invention may be applied.

[0080] like Figure 5 As shown, system architecture 500 may include terminal devices 501, 502, 503, a network 504 and a server 505. Network 504 is used to provide a medium for communication links between terminal devices 501, 502, 503 and server 505. Network 504 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0081] Users can use terminal devices 501, 502, 503 to interact with server 505 through network 504 to receive or send messages, etc. Various client applications can be installed on terminal devices 501, 502, 503, such as e-commerce client applications, web browser applications, search applications, instant messaging tools, and email clients.

[0082] The terminal devices 501 , 502 , and 503 may be various electronic devices having a display screen and supporting various client applications, including but not limited to smart phones, tablet computers, laptop computers, desktop computers, and the like.

[0083] The server 505 may be a server that provides various services, such as a server that provides support for client applications used by users using the terminal devices 501, 502, and 503. The server may process the received request for page acquisition and feed back the corresponding page containing recommendation information to the terminal device.

[0084] It should be noted that the information recommendation method provided in the embodiment of the present invention is generally executed by the server 505 , and accordingly, the information recommendation device is generally disposed in the server 505 .

[0085] It should be understood that Figure 5 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0086] Reference below Figure 6 , which shows a schematic diagram of the structure of a computer system 600 of a terminal device suitable for implementing an embodiment of the present invention. Figure 6 The terminal device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0087] like Figure 6 As shown, the computer system 600 includes a central processing unit (CPU) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage part 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the system 600 are also stored. The CPU 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0088] The following components are connected to the I / O interface 605: an input section 606 including a keyboard, a mouse, etc.; an output section 607 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 608 including a hard disk, etc.; and a communication section 609 including a network interface card such as a LAN card, a modem, etc. The communication section 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to the I / O interface 605 as needed. A removable medium 611, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 610 as needed, so that a computer program read therefrom is installed into the storage section 608 as needed.

[0089] In particular, according to the embodiments disclosed in the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 609, and / or installed from the removable medium 611. When the computer program is executed by the central processing unit (CPU) 601, the above-mentioned functions defined in the system of the present invention are executed.

[0090] It should be noted that the computer-readable medium shown in the present invention may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present invention, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0091] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0092] The modules and / or units involved in the embodiments of the present invention may be implemented in software or hardware. The modules and / or units described may also be provided in a processor. For example, they may be described as follows: a processor includes a module for receiving a page acquisition request, a module for determining a valid tag, and a module for determining a target recommendation information. The names of these modules do not, in some cases, constitute limitations on the modules themselves. For example, a module for determining target recommendation information may also be described as a "module for determining target recommendation information to be sent to a user based on the user's valid tags."

[0093] As another aspect, the present invention also provides a computer-readable medium, which may be included in the device described in the above embodiment; or it may exist independently and not be assembled into the device. The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by a device, the device includes: in response to receiving a page acquisition request sent by a client, providing a page corresponding to the page acquisition request to the client, wherein the page acquisition request includes user information, and the page includes an information recommendation position; according to the user information, determining the user tag corresponding to the user information; according to the pre-stored information to be recommended and the recommended tag corresponding to the information to be recommended, selecting a valid tag from the user tags; based on the valid tag, selecting target recommended information from the stored information to be recommended, and sending the target recommended information to the information recommendation position, so that the client displays the target recommended information through the information recommendation position.

[0094] It can be seen that one embodiment of the present invention can determine the user's effective tags based on the user tags that reflect the user portrait and the recommended tags that reflect the information to be recommended; and by calculating the recommended position tag value of the user tag corresponding to the recommended information position, select a set number of target tags from the effective tags, and then determine the recommended information to be displayed to the user based on the target tags; thereby overcoming the problem of low accuracy in determining the information to be recommended based only on user tags, improving the accuracy of information recommendation, and thereby improving the personalization level and input-output ratio of information recommendation.

[0095] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may occur depending on design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for information recommendation, characterized in that: include: In response to receiving a page acquisition request sent by a client, providing a page corresponding to the page acquisition request to the client, wherein the page acquisition request includes user information and the page includes an information recommendation position; Determine, according to the user information, a user tag corresponding to the user information; Selecting a valid tag from the user tags according to the pre-stored information to be recommended and the recommendation tags corresponding to the information to be recommended; Based on the valid tag, select target recommended information from the stored information to be recommended, and send the target recommended information to the information recommendation position, so that the client displays the target recommended information through the information recommendation position; The method further comprises: Obtaining user tags corresponding to multiple users browsing the information recommendation position; Calculating the recommendation position tag value of the user tag based on the recommendation costs of the plurality of recommended information corresponding to the information recommendation position; Selecting target recommended information from the stored information to be recommended based on the valid tag includes: According to the recommendation position tag value of the user tag, a set number of target tags are selected from the valid tags for the information recommendation position; The target recommendation information is determined according to the target tag.

2. The method according to claim 1, characterized in that Determining, according to the user information, a user tag corresponding to the user information, including: According to the user identifier included in the user information, a user tag corresponding to the user identifier included in the user information is obtained from a user feature data source, and the user feature data source includes a user tag generated by any one or more of user portrait data and user behavior data.

3. The method according to claim 1, characterized in that Calculating the recommendation position label value of the user label includes: For each of the user tags corresponding to the multiple users browsing the information recommendation position, execute: Determining at least one target recommended information corresponding to the user tag from a plurality of recommended information corresponding to the information recommendation position; According to the recommendation cost of at least one of the target recommended information, the recommendation position tag value of the user tag is counted.

4. The method according to claim 1, characterized in that: Calculating the recommendation position label value of the user label includes: For each of the user tags corresponding to the multiple users browsing the information recommendation position, execute: For the information recommendation position, count the number of clicks and views of the user tag; Based on the number of clicks and views of the user tag, a recommendation position tag value of the user tag is calculated.

5. The method according to claim 1, characterized in that According to the recommendation position tag value of the user tag, a set number of target tags are selected from the valid tags, including: Determine all first user tags corresponding to the information recommendation position and having the recommendation position tag value; Selecting a second user tag matching the first user tag from the valid tags; When the number of the second user tags is greater than a set number, a set number of target tags are selected from the second user tags according to the order of the recommended tag values ​​of the second user tags.

6. The method according to claim 5, characterized in that Further including: When the number of the second user tags is less than a set number, the second user tags are used as the target tags, and the remaining number of target tags are randomly selected from the valid tags, wherein the sum of the number of the second user tags and the remaining number is the set number.

7. The method according to claim 1, characterized in that According to the recommendation position tag value of the user tag, a set number of target tags are selected from the valid tags, including: sorting the plurality of third user tags corresponding to the information recommendation position according to the recommendation position tag value of the user tag; For the third user tags with the recommended position tag values ​​from high to low in the sorting results, the following steps are executed in a loop until the number of the target tags reaches the set number or all the third user tags in the sorting results have completed the following steps, and the loop ends: It is determined whether the third user tag belongs to the valid tag, and if so, the third user tag is determined to be the target tag.

8. The method according to claim 7, characterized in that The recommendation position tag value of any user tag corresponding to the information recommendation position includes: a recommendation cost statistical value and a click-through rate value; Sorting the plurality of third user tags corresponding to the information recommendation positions includes: Sort the user tags corresponding to the information recommendation positions based on the order of the recommendation cost statistics from high to low; When there are multiple user tags with the same recommendation cost statistical value, the multiple user tags with the same recommendation cost statistical value are sorted based on the order of the click rate values ​​from high to low.

9. An information recommendation device, characterized in that: include: A module for receiving page acquisition requests, a module for determining valid tags, and a module for determining target recommendation information; wherein, The page acquisition request receiving module is used to provide a page corresponding to the page acquisition request to the client in response to receiving a page acquisition request sent by the client, wherein the page acquisition request includes user information and the page includes an information recommendation position; The valid tag determination module is used to determine the user tag corresponding to the user information according to the user information; and select a valid tag from the user tags according to the pre-stored information to be recommended and the recommended tags corresponding to the information to be recommended; The target recommendation information determination module is used to select target recommendation information from the stored information to be recommended based on the valid tag, and send the target recommendation information to the information recommendation position, so that the client can display the target recommendation information through the information recommendation position; The device is further used to obtain user tags corresponding to multiple users browsing the information recommendation position; based on the recommendation costs of multiple recommended information corresponding to the information recommendation position, calculate the recommendation position tag value of the user tag; The module for determining target recommended information is used to select target recommended information from the stored information to be recommended based on the valid tag, including: selecting a set number of target tags for the information recommendation position from the valid tags according to the recommendation position tag value of the user tag; and determining the target recommended information according to the target tag.

10. The device according to claim 9, characterized in that The target recommendation information determination module is further used to obtain user tags corresponding to multiple users browsing the information recommendation position; based on the recommendation cost of multiple recommended information corresponding to the information recommendation position, calculate the recommendation position tag value of the user tag; Selecting target recommended information from the stored information to be recommended based on the valid tag includes: According to the recommendation position tag value of the user tag, a set number of target tags are selected from the valid tags for the information recommendation position; The target recommendation information is determined according to the target tag.

11. An electronic device, characterized in that: include: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 8.

12. A computer readable medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

Images