Material screening method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention discloses a material screening method and device, electronic equipment and a storage medium. The method comprises the steps that click materials and interactive materials of a user group in a target time period and interest capture materials of an inactive user group are acquired; the inactive user groups comprise a plurality of users whose frequency of initiating service requests within a target time period is less than a preset frequency threshold in the user groups; performing material screening on the click material, the interactive material and the interest capture material according to a predetermined screening rule to obtain a basic material; generating a target material library based on the basic material and the aging material; the aging materials comprise materials meeting preset aging and heat conditions; the target material library is a material library used for carrying out material recommendation to the inactive user group. According to the embodiment of the invention, the material library of the recommendation system can contain the materials interested by the inactive user group, so that the recommendation system can recommend the materials of the inactive user group more accurately.

Description

Technical Field

[0001] The present invention relates to the field of Internet technology, and in particular to a material screening method, device, electronic equipment and storage medium. Background Art

[0002] In the material recommendation scenario, the material library plays a vital role. When a user makes a business request, the material recommendation system can select videos, pictures, short articles, etc. that the user is interested in from the material library and recommend them to the user based on the user's past behaviors or interests. However, the frequency of users making business requests varies. Some users are high-frequency users, while some are new users or low-frequency users. Due to incomplete attribute portraits of new and low-frequency users or less user behavior, the material library of the existing recommendation system lacks materials that new or low-frequency users are interested in, resulting in inaccurate materials recommended by the recommendation system to new or low-frequency users, and new or low-frequency users are unable to generate interest in the recommended materials. Summary of the invention

[0003] The purpose of the embodiments of the present application is to provide a material screening method, device, electronic device and storage medium to solve the problem in the prior art that the material library of the recommendation system lacks materials of interest to new users or low-frequency users, thereby causing the recommendation system to recommend inaccurate materials to new users or low-frequency users.

[0004] To solve the above technical problems, the embodiments of the present application are implemented as follows:

[0005] In a first aspect, an embodiment of the present application provides a material screening method, comprising:

[0006] Obtaining click materials and interaction materials of a user group within a target time period, and interest capture materials of an inactive user group; the inactive user group includes a plurality of users in the user group whose frequency of initiating service requests within the target time period is less than a preset frequency threshold;

[0007] According to a predetermined screening rule, the click material, the interactive material and the interest capture material are screened to obtain basic materials;

[0008] Based on the basic materials and time-sensitive materials, a target material library is generated; the time-sensitive materials include materials that meet preset timeliness and heat conditions; the target material library is a material library used to recommend materials to the inactive user group.

[0009] In a second aspect, an embodiment of the present application provides a material screening device, comprising:

[0010] An acquisition module, used to acquire click materials and interaction materials of a user group within a target time period, and interest capture materials of an inactive user group; the inactive user group includes a plurality of users in the user group whose frequency of initiating service requests within the target time period is less than a preset frequency threshold;

[0011] A screening module, used to screen the click material, the interactive material and the interest capture material according to a predetermined screening rule to obtain a basic material;

[0012] A generation module is used to generate a target material library based on the basic materials and time-sensitive materials; the time-sensitive materials include materials that meet preset timeliness and heat conditions; the target material library is a material library used to recommend materials to the inactive user group.

[0013] In a third aspect, an embodiment of the present application provides an electronic device, comprising a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the material screening method described in the first aspect.

[0014] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the material screening method described in the first aspect are implemented.

[0015] In a fifth aspect, an embodiment of the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of the material screening method described in the first aspect above.

[0016] In the embodiment of the present application, first, the click materials and interactive materials of the user group within the target time period, as well as the interest capture materials of the inactive user group, are obtained; the inactive user group includes multiple users in the user group whose frequency of initiating business requests within the target time period is less than the preset frequency threshold; then, according to the predetermined screening rules, the click materials, interactive materials and interest capture materials are screened to obtain basic materials; then, based on the basic materials and time-effective materials, a target material library is generated; the time-effective materials include materials that meet the preset time and heat conditions; the target material library is the material library used for recommending materials to the inactive user group. It can be seen that through the embodiment of the present application, the click materials and interactive materials of the user group within the target time period and the interest capture materials of the inactive user group are screened according to the predetermined screening rules, and the target material library is generated according to the basic materials and time-effective materials obtained by screening, and then the target material library can be used to recommend materials of interest to the inactive user group, so that the recommendation system can recommend materials to the inactive user group more accurately. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0018] Figure 1 A schematic diagram of a material screening method provided in an embodiment of the present application;

[0019] Figure 2 A schematic diagram of a workflow of a recommendation system provided in an embodiment of the present application;

[0020] Figure 3 A schematic diagram of the structure of a material screening device provided in an embodiment of the present application;

[0021] Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0022] Embodiments of the present application provide a material screening method, device, electronic device and storage medium.

[0023] In order to enable those skilled in the art to better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0024] The embodiments of the present application provide a material screening method and device, which can be applied to a server, which can be an independent server or a server cluster composed of multiple servers, without specific limitation. The application scenarios of the above method and device include but are not limited to the scenario where the Internet platform recommends materials to users. Among them, the Internet platform can be an e-commerce platform, a streaming service platform, a social platform or an advertising platform, etc.; materials refer to objects recommended to users, including but not limited to goods, services or content, etc. Material information refers to information that can reflect the properties and characteristics of the material, and the display form includes but is not limited to text, pictures, audio or video, etc., without specific limitation. Through the above method and device, the problem of lack of materials that are of interest to new users and low-frequency users in the material library of the existing recommendation system is overcome, which leads to the problem that the recommendation system recommends inaccurate materials to new users and low-frequency users.

[0025] Figure 1 A schematic diagram of a material screening method provided in an embodiment of the present application is shown in FIG. Figure 1 As shown, the embodiment of the present application provides a material screening method, which may specifically include the following steps:

[0026] S102, obtaining click materials and interaction materials of the user group within the target time period, and interest capture materials of the inactive user group; the inactive user group includes a plurality of users in the user group whose frequency of initiating business requests within the target time period is less than a preset frequency threshold.

[0027] In the embodiment of the present application, click materials, interactive materials of the user group within the target time period and interest capture materials of the inactive group are obtained.

[0028] In one example, the Internet platform where the material recommendation system is located is a Weibo platform, and the click material of the user group can be the material corresponding to the click data of all Weibo users in the past 30 days, and the click data includes the user's click to forward, click to comment, click to like, follow the author of the graphic information and / or video data, click the author's personal homepage, etc. For example, user A likes a video titled "xx movie commentary" posted by user B on Weibo, then the video liked by user A is the click material.

[0029] Interactive materials can include materials corresponding to the interactive data of all Weibo users in the past 30 days, and the interactive data includes user forwarding, comments and likes. For example, user A forwards an article titled "xx restaurant review" published by user C to user C, then the article is an interactive material. It is understandable that user A can also forward and / or comment on a video titled "xx movie commentary" published by user B, then the video can also be an interactive material.

[0030] Interest-capturing materials can reflect the user's interests. By acquiring interest-capturing materials, the generated material library can be made more compatible with the user's interests, so as to better recommend accurate materials to the user.

[0031] The user group includes all users on the Internet platform, among which the user group includes active user groups and inactive user groups. Active users are users who initiate service requests more frequently than the preset frequency threshold during the target time period, i.e., high-frequency users. Inactive users are users who initiate service requests less frequently than the preset frequency threshold during the target time period, including new users and low-frequency users.

[0032] In one implementation, the frequency of the user initiating the service request may be a ratio of the number of times the user initiates the service request within a unit time to the length of the unit time, and the unit time may be a target time period.

[0033] In one example, the target time period is 30 days. User A initiates service requests 100 times within 30 days, that is, the average frequency of initiating service requests is 3 times per day. Assuming that the preset frequency threshold is 10 times per day, the frequency of user A initiating service requests within the target time period is less than the preset frequency threshold, and the user group to which user A belongs is an inactive user group.

[0034] In another implementation, the unit time is, for example, one day. Exemplarily, user A initiates service requests 2 times in one day, that is, the frequency of initiating service requests is 2 times a day. Assuming that the preset frequency threshold is 10 times a day, then the frequency of user A initiating service requests is less than the preset frequency threshold. Assuming that the target time period is 30 days, the number of days in which user A initiates service requests less than the preset frequency threshold within 30 days is 20 days, and the number of days in which the frequency of initiating service requests exceeds the preset frequency threshold is 10 days, then the user group to which user A belongs is an inactive user group.

[0035] In another implementation, the user type and the user group to which the user belongs can also be determined based on the ratio of the number of active days of the user in the target time period to the length of the target time period. The user is considered active on a day if he or she has at least one service request in a day. The service request can be an information flow request sent by the user to the Internet platform, or a login request sent to the Internet platform.

[0036] In an example, assuming that the target time period is 30 days, users are divided into new users, low-frequency users, and high-frequency users according to the number of active days in the past 30 days. Among them, new users can be subdivided into new users of the day, returning users, and new users in 30 days. High-frequency users can be subdivided into medium-low-frequency users, medium-high-frequency users, high-frequency users, and full-time users. Table 1 shows the correspondence between user types and active days. As shown in Table 1, returning users are users who have not been active in the past 30 days, new users in 30 days are new users in the past 30 days, low-frequency users are users who have been active for 1 to 4 days in the past 30 days, medium-low-frequency users are users who have been active for 5 to 15 days in the past 30 days, medium-high-frequency users are users who have been active for 16 to 24 days in the past 30 days, high-frequency users are users who have been active for 25 to 29 days in the past 30 days, and full-time users are users who have been active for 30 days in the past 30 days.

[0037] Table 1

[0038] User Type Active days New customers of the day New customers of the day Returning users Inactive in the past 30 days 30 days new account New users in the past 30 days Low frequency users Active for 1 to 4 days in the past 30 days Medium and low frequency users Active for 5-15 days in the past 30 days Medium and high frequency users Active for 16-24 days in the past 30 days High frequency users Active 25-29 days in the past 30 days Full attendance user Active for 30 days in the past 30 days

[0039] Among them, new users of the day, returning users, new users within 30 days, and low-frequency users can be determined as inactive user groups. It is understandable that although a user is active on the day if there is at least one service request in a day, the number of active days within 30 days for new users of the day, returning users, new users within 30 days, and low-frequency users is less than that for high-frequency users. Therefore, new users of the day, returning users, new users within 30 days, and low-frequency users can be determined as inactive user groups, and medium-low frequency users, medium-high frequency users, high frequency users, and full-time users can be determined as active user groups.

[0040] In some embodiments of the present application, the interest-capturing material of the inactive user group includes at least one of the following:

[0041] Materials related to the field to which the positive and negative feedback materials of each user in the inactive user group in the first time period belong;

[0042] Positive feedback materials from target users who meet the set conditions in the inactive user group during the second time period;

[0043] Materials whose exposure ranking improvement rate in two consecutive time periods among inactive user groups exceeds the preset ranking improvement rate threshold;

[0044] Among them, positive feedback materials refer to materials that are clicked, interacted or viewed by users, and the viewing time is greater than the preset time; the first time period and the second time period are within the target time period.

[0045] In the embodiment of the present application, the interest-capturing materials of the inactive user group include materials related to the field to which the positive feedback materials of each user in the inactive user group within the first time period belong.

[0046] For example, the first time period may be a day or a period of time two days or one day before the current time. The current time is, for example, September 5, and the target time period is, for example, August 6 to September 4. The first time period is, for example, September 2 to September 3. Positive feedback is that user A likes, reposts, and watches multiple articles or videos posted by multiple bloggers on the Weibo platform, and the viewing time of each article or video is greater than 5 seconds.

[0047] In the embodiment of the present application, the interest-capturing materials of the inactive user group include positive feedback materials of target users in the inactive user group that meet the set conditions within the second time period.

[0048] In some embodiments of the present application, the target user includes a user whose frequency of initiating service requests in the target time period changes, and the changed frequency is less than a preset frequency threshold; and / or a user whose overlap between each interest point in the third time period and each interest point in the fourth time period reaches a preset overlap threshold, and whose interest score in the fourth time period is greater than that in the third time period;

[0049] Among them, the points of interest are used to indicate the areas of interest of the user, and the interest score is used to indicate the degree of interest of the user in the points of interest; the third time period and the fourth time period are within the second time period, and the third time period and the fourth time period differ by a preset time interval.

[0050] In the embodiment of the present application, the target user in the inactive user group may be a user whose frequency of initiating service requests in the target time period changes, and the frequency after the change is still less than the preset frequency threshold. In a specific implementation, the target user may be a user whose frequency of initiating service requests in the target time period increases, and the frequency after the increase is still less than the preset frequency threshold.

[0051] In one example, the current time is September 5, and user B is a new user in the past 30 days. After that, user B has been browsing the information flow on the Weibo platform for 14 days. In the past 30 days, user B has changed from a new user to a medium-low frequency user, and the frequency of user B's service requests has increased.

[0052] In an embodiment of the present application, the target users in the inactive user group may also be users whose overlap between various interest points in the third time period and various interest points in the fourth time period reaches a preset overlap threshold, and whose interest score in the fourth time period is greater than the interest score in the third time period, that is, users with solidified interests.

[0053] In one example, the current time is, for example, September 5, and the target time period is, for example, August 6-September 4. The second time period is, for example, August 29 to September 4, the third time period is, for example, August 29-September 2, and the fourth time period is, for example, August 31-September 4. User C comes to Weibo to browse the information flow every day from August 29 to September 2, and the top five points of interest of user C from August 29 to September 2 overlap with the top five points of interest from August 31 to September 4, and the interest score of each point of interest from August 31 to September 4 increases compared to that from August 29 to September 2, so user C is a user with solidified interests. Points of interest are used to indicate the areas of interest of users, and interest scores are used to indicate the degree of interest of users in points of interest. Interest scores can be generated based on the positive feedback of the second user on materials in various fields. The specific generation process of interest scores can refer to the relevant technology and will not be repeated here. The materials on which the users B and C provide positive feedback from August 29 to September 4 can be used as interest-capturing materials for users B and C.

[0054] In some implementations of the present application, the interest capture material of the inactive user group further includes:

[0055] Materials whose exposure ranking improvement rate in two consecutive time periods among inactive user groups exceeds the preset ranking improvement rate threshold.

[0056] The interest-capturing material of the inactive user group may be a material whose exposure ranking improvement rate in the fifth time period and the exposure ranking improvement rate in the sixth time period both exceed a preset ranking improvement rate threshold in the inactive user group. The fifth time period and the sixth time period are within the target time period.

[0057] In one example, the inactive user group includes user B and user C, the current time is, for example, September 5, the fifth time period is September 2-September 3, and the sixth time period is September 4. The exposure of all materials to user B and user C on September 1 can be counted and sorted, then the exposure to user B and user C from September 2 to September 3 is counted and sorted, and the exposure to user B and user C on September 4 is counted and sorted. Materials with an exposure ranking improvement rate of more than 5% to user B and user C from September 2 to September 3 and an exposure ranking improvement rate of more than 5% to user B and user C on September 4 are interest-capturing materials for user B and user C.

[0058] After the target material is determined as the interest-capturing material, the determined interest-capturing material can be classified to obtain the category to which each interest-capturing material belongs. For example, materials related to the field to which the positive feedback materials of each user in the inactive user group in the first time period belong can be determined as similar materials; positive feedback materials of the target user in the inactive user group in the second time period can be determined as positive feedback materials; materials in the inactive user group whose exposure ranking improvement rate in two consecutive time periods exceeds a preset ranking improvement rate threshold can be determined as hot-selling materials.

[0059] S104, screening the click materials, interactive materials and interest-capturing materials according to predetermined screening rules to obtain basic materials.

[0060] In some implementations of the present application, the click materials, interactive materials, and interest-capturing materials are screened according to predetermined screening rules to obtain basic materials, including:

[0061] Generate corresponding click material library, interactive material library and interest capture material library according to click materials, interactive materials and interest capture materials;

[0062] According to the exposure style of the inactive user group, the proportion base of materials in various fields included in the click material library, interactive material library and interest capture material library are determined respectively; the exposure style is used to characterize the field distribution characteristics of the materials that the inactive user group is exposed to;

[0063] According to the proportion base of materials in various fields contained in each material library and the preset screening ratio, the click material library, interactive material library and interest capture material library are screened to obtain basic materials.

[0064] In some implementations of the application, according to the click material, the interactive material and the interest capture material, a corresponding click material library, an interactive material library and an interest capture material library are generated, including:

[0065] According to the click rates corresponding to the click materials, the click materials are sorted, and a click material library is generated based on the sorted click materials;

[0066] According to the number of interactions corresponding to each interactive material, each interactive material is sorted, and an interactive material library is generated based on the sorted interactive materials;

[0067] According to the category to which each interest capturing material belongs, each interest capturing material is assigned a corresponding value, and each interest capturing material after the value assignment is sorted, and an interest capturing material library is generated according to each sorted interest capturing material.

[0068] When generating a corresponding click material library based on click materials, the click materials can be sorted according to the click rate corresponding to each click material, and the corresponding click material library can be generated based on the sorted click materials. The click rate can be calculated based on the number of clicks and the real exposure number corresponding to the click material according to the formula click rate = number of clicks / real exposure number. The number of clicks is the number of times the material is clicked by users, and the real exposure number is the number of times the material is actually viewed by users.

[0069] When generating a corresponding interactive material library based on interactive materials, each interactive material can be sorted according to the number of interactions corresponding to each interactive material, such as the number of reposts and / or the number of comments, and a corresponding interactive material library can be generated based on the sorted interactive materials.

[0070] When generating a corresponding interest capture material library based on the interest capture materials, each interest capture material can be assigned a corresponding value according to the category to which each interest capture material belongs, and then the assigned interest capture materials can be sorted. Based on the sorted interest capture materials, a corresponding interest capture material library can be generated.

[0071] In one example, the material belongs to the similar material category and is assigned a value of 2.5; the material belongs to the positive feedback material category and is assigned a value of 3.3; the material belongs to the hot-selling material category and is assigned a value of 5. After assigning values ​​to each interest capture material, they are sorted from low to high according to the scores, with low-score materials sorted first and high-score materials sorted last. Considering that there may be multiple materials with the same score, the sorting results of each material with the same score can be confirmed according to the actual sum of the number of reposts, comments, and likes of the material, and the materials with a larger sum of reposts, comments, and likes are sorted first. Then, according to each interest capture material after sorting, the corresponding interest capture material library is generated.

[0072] In the embodiment of the present application, the exposure style of the inactive user group is used to characterize the field distribution characteristics of the materials that the inactive user group is exposed to. Through the exposure style of the inactive user group, the material preference style or characteristics presented by the inactive user group and the field to which the materials that the user is interested belong can be obtained, and then the proportion base of the materials in each field contained in the click material library, interactive material library and interest capture material library can be determined. Among them, the fields to which the materials belong include entertainment stars, sports, society, games and many others. The screening ratio can be predetermined based on the specific material recommendation scenario, material recommendation results and user behavior feedback. The specific determination process can refer to the relevant technology, and this application will not repeat it here.

[0073] After determining the proportion base of materials in each field, the click material library, interactive material library and interest capture material library can be screened according to the proportion base of materials in each field contained in each material library and the preset screening ratio to obtain basic materials.

[0074] In one example, the screening ratio is click material library: interactive material library: interest capture material library = 6:6:4. For the field of entertainment stars, the ratio base is 1000. Then, in the click material library, for the field of entertainment stars, the number of field materials screened out = ratio base * screening ratio, and the number of field materials is 6000, that is, 6000 materials are selected. Similarly, in the interactive material library, 6000 materials are selected for the field of entertainment stars, and in the interest capture material library, 4000 materials are selected for the field of entertainment stars. Similarly, for other fields, according to the field ratio base, based on the same screening ratio, the click material library, interactive material library and interest capture material library are screened. Table 2 shows each field and the corresponding ratio base of each field. As shown in Table 2, the ratio base of the entertainment star field is 1000, the ratio base of the sports field is 540, the ratio base of the social field is 950, and the ratio base of the game field is 1189.

[0075] Table 2

[0076] Field Name Field Number Proportional base Entertainment Stars Category_001 1000 physical education Category_002 540 society Category_003 950 game Category_004 1189

[0077] It should be noted that the above-mentioned ratio bases of various fields and specific values ​​of the screening ratios are only used for examples and are not used for specific limitations. They can be set or adjusted according to actual needs in specific application scenarios.

[0078] S106: Generate a target material library based on basic materials and time-sensitive materials; time-sensitive materials include materials that meet preset timeliness and popularity conditions; the target material library is a material library used to recommend materials to inactive user groups.

[0079] In the embodiment of the present application, after materials are screened in the click material library, the interactive material library, and the interest capture material library, a target material library is generated based on the basic materials and time-sensitive materials obtained after the screening.

[0080] Timely materials include materials that meet preset timeliness and popularity conditions. Materials that meet timeliness conditions are materials that are published relatively close to the current time, for example, the top 10 news in the real-time news list in chronological order. Popularity conditions can be determined based on the attention, spread speed and scope, or interactivity of the material. Materials that meet popularity conditions are hot materials. For example, hot materials are materials corresponding to the top 10 entries on the Weibo hot search list.

[0081] After generating the target material library, the recommendation system can use the target material library to recommend materials to inactive user groups.

[0082] Considering that the data in the Internet platform is constantly updated and the user access situation is different every day, after the target material library is generated, the target material library will be updated periodically. For example, new materials can be added to the target material library every 24 hours. The newly added materials can come from the above-mentioned click material library, interactive material library and interest capture material library. Correspondingly, the click material library, interactive material library and interest capture material library will also be updated regularly.

[0083] In some embodiments of the present application, after generating the target material library, the method further includes:

[0084] Receive a service request from a service requester and confirm the user group to which the service requester belongs;

[0085] In the case that the user group to which the business requester belongs is an inactive user group, corresponding materials are selected from the target material library according to preset configuration items and recommended to the business requester.

[0086] Figure 2 A workflow diagram of a recommendation system provided in an embodiment of the present application is shown in FIG. Figure 2 As shown, the target material library and the main material library are independent of each other and output in parallel. When the business requester sends a business request, the business requester is judged to take the main recall or the new user recall according to the user group to which the business requester belongs. The main recall uses the main material library, and the new user recall uses the target material library. If it is an inactive user group, the new user recall is taken. Among them, the main recall includes other recalls, new target recalls and DR (Deep Retrieval) recalls. In a specific case, if the business requester is a low-frequency user in the inactive user group or a new user in 30 days, other recalls, new target recalls and DR (Deep Retrieval) recalls can also be taken. New target recall refers to recall for a newly set or updated target user group, which involves redefining or updating the target user, and then formulating a suitable recall strategy. If it is other users, there is no new user recall, only other recalls, new target recalls and DR recalls.

[0087] In one example, if it is determined that the user group to which the business requester belongs is an inactive user group and a low-frequency user, the recommendation system reads the relevant ab experiment configuration items, such as recall ratio, recall domain, rough ranking service pool, recall model name, recall model corresponding multiplier, etc. Then, the recommendation system calculates the score of each material in the target material library in the recall stage, and selects the top N materials and the top R materials selected from the main material library, and passes them to the rough ranking stage together. After scoring the N+R materials in the rough ranking stage, the top M materials are selected and passed to the fine ranking stage. After scoring the M materials in the fine ranking stage, the top Q materials are selected according to the truncation and quota strategies and recommended to the business requester.

[0088] If the user group to which the business requester belongs is determined to be an active user group, there is no new user recall path. Then, in the recall phase, the scores of each material in the main material library are calculated, and the top N materials are selected and passed to the rough ranking phase. After scoring the N materials in the rough ranking phase, the top M materials are selected and passed to the fine ranking phase. After scoring the M materials in the fine ranking phase, the top Q materials are selected and recommended to the user.

[0089] In another embodiment of the present application, a high-quality material library can be added, and corresponding materials can be selected from the high-quality material library and recommended to the business requester. The high-quality material library can be generated based on multiple materials that meet preset quality conditions, or it can be generated by screening text, pictures, videos and other information posted by multiple high-quality users.

[0090] In some embodiments of the present application, it also includes:

[0091] In the case that the user group to which the business requester belongs is not an inactive user group, corresponding materials are selected from the main material library and recommended to the business requester according to the corresponding preset configuration items.

[0092] In an embodiment of the present application, when a service request is received from a service requester, the user group to which the service requester belongs can be confirmed first. If the user group to which the service requester belongs is an inactive user group, then the target material library generated above can be used to recommend materials to the service requester. If the user group to which the service requester belongs is not an inactive user group but an active user group, then the main material library of the recommendation system can be used to recommend materials to the service requester.

[0093] In some implementations of the present application, confirming the user type of the target user includes:

[0094] Count the frequency of service requests initiated by the service requester within the target time period;

[0095] Based on the statistical results, confirm the user group to which the service requester belongs.

[0096] When confirming the user group to which the service requester belongs, the user group to which the service requester belongs can be confirmed based on the frequency of service requests initiated by users in the target time period. The specific method for confirming the user group to which the service requester belongs can refer to the above step S102, which will not be repeated in this application.

[0097] It should be noted that the various values ​​in the above examples are only for illustration and not for limitation, and they can be set as needed in actual applications.

[0098] In the embodiment of the present application, first, the click materials and interactive materials of the user group within the target time period, as well as the interest capture materials of the inactive user group, are obtained; the inactive user group includes multiple users in the user group whose frequency of initiating business requests within the target time period is less than the preset frequency threshold; then, according to the predetermined screening rules, the click materials, interactive materials and interest capture materials are screened to obtain basic materials; then, based on the basic materials and time-effective materials, a target material library is generated; the time-effective materials include materials that meet the preset time and heat conditions; the target material library is the material library used for recommending materials to the inactive user group. It can be seen that through the embodiment of the present application, the click materials and interactive materials of the user group within the target time period and the interest capture materials of the inactive user group are screened according to the predetermined screening rules, and the target material library is generated according to the basic materials and time-effective materials obtained by screening, and then the target material library can be used to recommend materials of interest to the inactive user group, so that the recommendation system can recommend materials to the inactive user group more accurately.

[0099] The above is a material screening method provided in an embodiment of the present application. Based on the same idea, an embodiment of the present application also provides a material screening device. Figure 3 A schematic diagram of the structure of a material screening device provided in an embodiment of the present application is shown in FIG. Figure 3 As shown, the material screening device comprises:

[0100] The acquisition module 301 is used to acquire click materials and interaction materials of a user group within a target time period, and interest capture materials of an inactive user group; the inactive user group includes a plurality of users in the user group whose frequency of initiating service requests within the target time period is less than a preset frequency threshold;

[0101] A screening module 302 is used to screen the click material, the interactive material and the interest capture material according to a predetermined screening rule to obtain a basic material;

[0102] The generation module 303 is used to generate a target material library based on the basic materials and time-sensitive materials; the time-sensitive materials include materials that meet preset timeliness and popularity conditions; the target material library is a material library used to recommend materials to the inactive user group.

[0103] In the embodiment of the present application, the screening module 302 is specifically used to generate corresponding click material libraries, interactive material libraries and interest capture material libraries according to the click materials, the interactive materials and the interest capture materials;

[0104] According to the exposure style of the inactive user group, the proportion bases of materials in various fields contained in the click material library, the interactive material library and the interest capture material library are determined respectively; the exposure style is used to characterize the field distribution characteristics of the materials exposed to the inactive user group;

[0105] According to the proportion base of materials in various fields contained in each material library and the preset screening ratio, the click material library, the interactive material library and the interest capture material library are screened to obtain basic materials.

[0106] In the embodiment of the present application, the screening module 302 is further specifically used to sort the click materials according to the click rates corresponding to the click materials, and generate the click material library according to the sorted click materials;

[0107] Sorting the interactive materials according to the number of interactions corresponding to the interactive materials, and generating the interactive material library according to the sorted interactive materials;

[0108] According to the category to which each interest capturing material belongs, each interest capturing material is assigned a corresponding value, and each interest capturing material after the assignment is sorted, and the interest capturing material library is generated according to each sorted interest capturing material.

[0109] In the embodiment of the present application, the interest-capturing material of the inactive user group includes at least one of the following:

[0110] Materials related to the field to which the positive feedback materials of each user in the inactive user group within the first time period belong;

[0111] Positive feedback materials from target users in the inactive user group who meet the set conditions within the second time period;

[0112] Materials whose exposure ranking improvement rate in two consecutive time periods in the inactive user group exceeds the preset ranking improvement rate threshold;

[0113] The positive feedback material refers to the material that is clicked, interacted or viewed by the user, and the viewing time is greater than the preset time; the first time period and the second time period are within the target time period.

[0114] In the embodiment of the present application, the target user includes a user whose frequency of initiating service requests in the target time period changes, and the changed frequency is less than a preset frequency threshold; and / or a user whose overlap between each interest point in the third time period and each interest point in the fourth time period reaches a preset overlap threshold, and whose interest score in the fourth time period is greater than that in the third time period;

[0115] Among them, the points of interest are used to indicate the areas of interest of the user, and the interest score is used to indicate the degree of interest of the user in the points of interest; the third time period and the fourth time period are within the second time period, and the third time period and the fourth time period differ by a preset time interval.

[0116] In the embodiment of the present application, the generating module 303 is further configured to receive a service request sent by a service requester and confirm the user group to which the service requester belongs;

[0117] In the case that the user group to which the service requester belongs is the inactive user group, corresponding materials are selected from the target material library according to preset configuration items and recommended to the service requester.

[0118] In the embodiment of the present application, the generating module 303 is further specifically used to count the frequency of the service request initiated by the service requester within the target time period;

[0119] According to the statistical results, the user group to which the service requester belongs is confirmed.

[0120] The above-mentioned device provided in the embodiment of the present application can execute the material screening method provided in the above-mentioned various method embodiments. The specific process is detailed in the description of the above-mentioned method embodiments and will not be repeated here.

[0121] The above-mentioned device provided by the embodiment of the present application, first, obtains the click materials and interactive materials of the user group within the target time period, and the interest capture materials of the inactive user group; the inactive user group includes multiple users in the user group whose frequency of initiating business requests within the target time period is less than the preset frequency threshold; then, according to the predetermined screening rules, the click materials, interactive materials and interest capture materials are screened to obtain basic materials; then, based on the basic materials and time-effective materials, a target material library is generated; the time-effective materials include materials that meet the preset time and heat conditions; the target material library is the material library used for recommending materials to the inactive user group. It can be seen that through the embodiment of the present application, by screening the click materials and interactive materials of the user group within the target time period and the interest capture materials of the inactive user group according to the predetermined screening rules, and generating the target material library based on the basic materials and time-effective materials obtained by screening, the target material library can be used to recommend materials of interest to the inactive user group, so that the recommendation system can recommend materials to the inactive user group more accurately.

[0122] The embodiment of the present application also provides an electronic device, Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. Figure 4 As shown, the device includes: a memory 401, a processor 402, a bus 403 and a communication interface 404. The memory 401, the processor 402 and the communication interface 404 communicate through the bus 403, and the communication interface 404 may include an input and output interface, which includes but is not limited to a keyboard, a mouse, a display, a microphone, a loudspeaker, etc.

[0123] Figure 4 In the embodiment, the memory 401 stores computer executable instructions that can be run on the processor 402. When the computer executable instructions are executed by the processor 402, the following process is implemented:

[0124] Obtaining click materials and interaction materials of a user group within a target time period, and interest capture materials of an inactive user group; the inactive user group includes a plurality of users in the user group whose frequency of initiating service requests within the target time period is less than a preset frequency threshold;

[0125] According to a predetermined screening rule, the click material, the interactive material and the interest capture material are screened to obtain basic materials;

[0126] Based on the basic materials and time-sensitive materials, a target material library is generated; the time-sensitive materials include materials that meet preset timeliness and heat conditions; the target material library is a material library used to recommend materials to the inactive user group.

[0127] From the technical solution provided by the above embodiment of the present application, it can be seen that in the embodiment of the present application, first, the click materials and interactive materials of the user group within the target time period, and the interest capture materials of the inactive user group are obtained; the inactive user group includes multiple users in the user group whose frequency of initiating business requests within the target time period is less than the preset frequency threshold; then, according to the predetermined screening rules, the click materials, interactive materials and interest capture materials are screened to obtain basic materials; then, based on the basic materials and time-effective materials, a target material library is generated; the time-effective materials include materials that meet the preset time and heat conditions; the target material library is the material library used for material recommendation to the inactive user group. It can be seen that through the embodiment of the present application, the click materials and interactive materials of the user group within the target time period and the interest capture materials of the inactive user group are screened according to the predetermined screening rules, and the target material library is generated according to the basic materials and time-effective materials obtained by screening, and then the target material library can be used to recommend materials of interest to the inactive user group, so that the recommendation system can recommend materials to the inactive user group more accurately.

[0128] The electronic device provided in the embodiment of the present application can implement each process in the aforementioned method embodiment and achieve the same functions and effects, which will not be repeated here.

[0129] An embodiment of the present application also provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the various processes of the above-mentioned method embodiment are implemented and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0130] The embodiment of the present application also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, each process of the above method embodiment is implemented, and the same technical effect can be achieved. To avoid repetition, it is not repeated here. The computer-readable storage medium is, for example, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0131] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-readable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0132] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0133] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0134] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0135] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0136] Memory may include non-permanent storage in a computer-readable storage medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. Memory is an example of a computer-readable storage medium.

[0137] Computer-readable storage media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer-readable instructions, data structures, program modules or other data. Examples of computer-readable storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable storage media does not include temporary computer-readable media (transitory media), such as modulated data signals and carrier waves.

[0138] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0139] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0140] The above description is only an embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent substitution, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of the claims of the present invention.

Claims

1. A material screening method, characterized in that: The method comprises: Obtaining click materials and interaction materials of a user group within a target time period, and interest capture materials of an inactive user group; the inactive user group includes a plurality of users in the user group whose frequency of initiating service requests within the target time period is less than a preset frequency threshold; According to a predetermined screening rule, the click material, the interactive material and the interest capture material are screened to obtain basic materials; Based on the basic materials and time-sensitive materials, a target material library is generated; the time-sensitive materials include materials that meet preset timeliness and heat conditions; the target material library is a material library used to recommend materials to the inactive user group.

2. The method according to claim 1, characterized in that: The step of screening the click material, the interactive material and the interest capture material according to a predetermined screening rule to obtain basic materials includes: Generate a corresponding click material library, an interactive material library and an interest capturing material library according to the click material, the interactive material and the interest capturing material; According to the exposure style of the inactive user group, the proportion bases of materials in various fields contained in the click material library, the interactive material library and the interest capture material library are determined respectively; the exposure style is used to characterize the field distribution characteristics of the materials exposed to the inactive user group; According to the proportion base of materials in various fields contained in each material library and the preset screening ratio, the click material library, the interactive material library and the interest capture material library are screened to obtain basic materials.

3. The method according to claim 2, characterized in that The generating corresponding click material library, interactive material library and interest capture material library according to the click material, the interactive material and the interest capture material comprises: Sorting the click materials according to the click rates corresponding to the click materials, and generating the click material library according to the sorted click materials; Sorting the interactive materials according to the number of interactions corresponding to the interactive materials, and generating the interactive material library according to the sorted interactive materials; According to the category to which each interest capturing material belongs, each interest capturing material is assigned a corresponding value, and each interest capturing material after the assignment is sorted, and the interest capturing material library is generated according to each sorted interest capturing material.

4. The method according to claim 1, characterized in that: The interest-capturing material of the inactive user group includes at least one of the following: Materials related to the field to which the positive feedback materials of each user in the inactive user group within the first time period belong; Positive feedback materials from target users in the inactive user group who meet the set conditions within the second time period; Materials whose exposure ranking improvement rate in two consecutive time periods in the inactive user group exceeds the preset ranking improvement rate threshold; The positive feedback material refers to the material that is clicked, interacted or viewed by the user, and the viewing time is greater than the preset time; the first time period and the second time period are within the target time period.

5. The method according to claim 4, characterized in that The target users include users whose frequency of initiating service requests in the target time period changes, and the changed frequency is less than a preset frequency threshold; and / or users whose overlap between each interest point in the third time period and each interest point in the fourth time period reaches a preset overlap threshold, and whose interest score in the fourth time period is greater than that in the third time period; Among them, the points of interest are used to indicate the areas of interest of the user, and the interest scores are used to indicate the degree of interest of the user in the points of interest; the third time period and the fourth time period are within the second time period, and the third time period and the fourth time period differ by a preset time interval.

6. The method according to claim 1, characterized in that After generating the target material library, it also includes: Receive a service request sent by a service requester, and confirm the user group to which the service requester belongs; In the case that the user group to which the service requester belongs is the inactive user group, corresponding materials are selected from the target material library according to preset configuration items and recommended to the service requester.

7. The method according to claim 6, characterized in that The confirming the user group to which the service requester belongs includes: Counting the frequency of service requests initiated by the service requester within the target time period; According to the statistical results, the user group to which the service requester belongs is confirmed.

8. A material screening device, characterized in that: The device comprises: An acquisition module, used to acquire click materials and interaction materials of a user group within a target time period, and interest capture materials of an inactive user group; the inactive user group includes a plurality of users in the user group whose frequency of initiating service requests within the target time period is less than a preset frequency threshold; A screening module, used to screen the click material, the interactive material and the interest capture material according to a predetermined screening rule to obtain a basic material; A generation module is used to generate a target material library based on the basic materials and time-sensitive materials; the time-sensitive materials include materials that meet preset timeliness and heat conditions; the target material library is a material library used to recommend materials to the inactive user group.

9. An electronic device, characterized in that: The method comprises a processor, a memory and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the material screening method according to any one of claims 1 to 7.

10. A computer-readable storage medium having computer-executable instructions stored therein, characterized in that: When the computer executable instructions are executed by a processor, the steps of the material screening method according to any one of claims 1 to 7 can be implemented.

Images