Article recommendation method and device, electronic equipment and storage medium

Abstract

The invention relates to an article recommendation method and device, electronic equipment and a storage medium. The article recommendation method comprises the steps that in response to article recommendation performed on a first user, a second user is determined, and the similarity between the second user and the first user meets a similarity condition; acquiring a to-be-recommended article set of the second user, wherein the to-be-recommended article set is determined on the basis of browsing chapters and browsing times of the second user on articles; and recommending articles to the first user based on the to-be-recommended article set. According to the invention, the efficiency of article reading recommendation between different users and the accuracy of article recommendation are improved.

Description

Technical Field

[0001] The present disclosure relates to the technical field of information recommendation, and in particular to an article recommendation method, device, electronic device and storage medium. Background Art

[0002] With the continuous development of information recommendation technology, people's requirements for recommendation accuracy are getting higher and higher.

[0003] In current recommendation systems, user-based collaborative filtering algorithms are widely used for personalized recommendations. However, in scenarios involving content-based article recommendations, most recommendation systems rely on user click behavior, ratings and other feedback to optimize recommendations, but fail to fully consider the actual level of user involvement in the reading process. Summary of the invention

[0004] In order to overcome the problems existing in the related art, the present disclosure provides an article recommendation method, device, electronic device and storage medium.

[0005] According to a first aspect of an embodiment of the present disclosure, there is provided an article recommendation method, comprising: in response to recommending an article to a first user, determining a second user, wherein a similarity between the second user and the first user satisfies a similarity condition; obtaining a set of articles to be recommended for the second user, wherein the set of articles to be recommended is determined based on the length and number of views of the articles by the second user; and recommending an article to the first user based on the set of articles to be recommended.

[0006] In one embodiment, the set of articles to be recommended is determined based on the browsing length and browsing times of the articles by the second user in the following manner: the browsing length and browsing times corresponding to each article browsed by the second user are determined, and based on the browsing length and browsing times, a target browsing length and a target browsing times are determined; based on the target browsing length and the target browsing times, the second user's preference for each article is determined; the articles are sorted from high to low according to the preference, and a preset number of articles ranked first are used as the set of articles to be recommended.

[0007] In one embodiment, determining the second user's preference for each article based on the target browsing length and the number of views includes: determining an initial preference based on the target browsing length, the initial preference representing the second user's preference for a single browse of each article; and determining the product of the initial preference and the number of views as the second user's preference for each article.

[0008] In one embodiment, determining the initial preference level based on the target browsing length includes: determining the exponent of an exp exponential function based on the target browsing length, wherein the exponent is the difference between a first constant and a ratio function, and the ratio function is the product of the target browsing length and a second constant; determining the sum of a third constant and the exp exponential function, and mapping the inverse of the sum to the initial preference level.

[0009] In one embodiment, the target browsing length and target browsing number are determined based on the browsing length and the browsing number, including: among the articles browsed by the second user, the articles whose browsing number is less than the browsing number threshold are filtered out to obtain the filtered articles; for each of the filtered articles, the browsing length of each browsing of the same article is obtained, and the maximum browsing length is determined, and the maximum browsing length is used as the target browsing length of the article, and the total number of browsing times of the article is used as the target browsing number of the article.

[0010] In one implementation, recommending an article to the first user based on the set of articles to be recommended includes: in response to the set of articles to be recommended including an article to be recommended and the articles browsed by the first user not including the article to be recommended, recommending the article to be recommended to the first user.

[0011] According to a second aspect of an embodiment of the present disclosure, there is provided an article recommendation device, comprising: a determination unit, for determining a second user in response to recommending an article to a first user, wherein the similarity between the second user and the first user satisfies a similarity condition; a processing unit, for obtaining a set of articles to be recommended for the second user, wherein the set of articles to be recommended is determined based on the length and number of views of the articles by the second user; and a recommendation unit, for recommending articles to the first user based on the set of articles to be recommended.

[0012] In one embodiment, the processing unit determines a set of articles to be recommended based on the browsing length and browsing times of the second user's articles in the following manner: determine the browsing length and browsing times corresponding to each article browsed by the second user, and based on the browsing length and browsing times, determine a target browsing length and a target browsing times; based on the target browsing length and the target browsing times, determine the second user's preference for each article; sort the articles from high to low according to the preference, and use a preset number of articles ranked first as a set of articles to be recommended.

[0013] In one embodiment, the processing unit determines the second user's preference for each of the articles based on the target browsing length and the number of views in the following manner: based on the target browsing length, an initial preference is determined, wherein the initial preference represents the second user's preference for a single browse of each of the articles; and the product of the initial preference and the number of views is determined as the second user's preference for each of the articles.

[0014] In one embodiment, the processing unit determines the initial preference level based on the target browsing length in the following manner: based on the target browsing length, determining the exponent of the exp exponential function, the exponent being the difference between a first constant and a ratio function, the ratio function being the product of the target browsing length and a second constant; determining the sum of a third constant and the exp exponential function, and mapping the inverse of the sum to the initial preference level.

[0015] In one embodiment, the processing unit determines the target browsing length and target browsing number based on the browsing length and browsing number in the following manner: among the articles browsed by the second user, the articles whose browsing number is less than the browsing number threshold are filtered out to obtain the filtered articles; for each of the filtered articles, the browsing length of each browsing of the same article is obtained respectively, and the maximum browsing length is determined, and the maximum browsing length is used as the target browsing length of the article, and the total number of browsing times of the article is used as the target browsing number of the article.

[0016] In one embodiment, the recommendation unit recommends an article to the first user based on the set of articles to be recommended in the following manner: in response to the set of articles to be recommended including an article to be recommended and the articles browsed by the first user not including the article to be recommended, recommending the article to be recommended to the first user.

[0017] According to a third aspect of an embodiment of the present disclosure, an electronic device is provided, comprising: a processor; and a memory for storing processor-executable instructions; wherein the processor is configured to execute the executable instructions to execute the article recommendation method in the first aspect or any one of the implementations of the first aspect.

[0018] According to a fourth aspect of an embodiment of the present disclosure, a storage medium is provided, in which instructions are stored. When the instructions in the storage medium are executed by a processor of a terminal, the terminal device is enabled to execute the article recommendation method in the first aspect or any one of the implementations of the first aspect.

[0019] The technical solution provided by the embodiments of the present disclosure may include the following beneficial effects: by introducing the implicit feature of the second user's reading length of the article, the accuracy of capturing the user's reading preference is improved, and comprehensive consideration of the reading length and the number of views is conducive to reflecting the second user's real reading interest, and article recommendations are made to the first user based on the second user's set of articles to be recommended, without determining the articles to be recommended for each user, thereby improving the efficiency of article recommendations.

[0020] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0022] Figure 1 The figure is a flowchart of a method for recommending articles according to an exemplary embodiment.

[0023] Figure 2 The present invention is a flowchart showing a method of determining a set of articles to be recommended according to an exemplary embodiment.

[0024] Figure 3 The present invention is a flowchart showing a method for determining an initial preference degree according to an exemplary embodiment.

[0025] Figure 4 The present invention is a flowchart showing a method of determining a user's article preference according to an exemplary embodiment.

[0026] Figure 5 It is a schematic diagram of a reading situation according to an exemplary embodiment.

[0027] Figure 6 The present invention is a flowchart of determining target data according to an exemplary embodiment.

[0028] Figure 7 The figure is a block diagram of a device for recommending articles according to an exemplary embodiment.

[0029] Figure 8 It is a structural schematic diagram of an application embodiment of the electronic device disclosed in the present invention.

[0030] Fig. 9 A schematic structural diagram of an embodiment of an electronic device disclosed in the present invention is shown. DETAILED DESCRIPTION

[0031] Here, exemplary embodiments will be described in detail, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present disclosure.

[0032] The article recommendation method provided by the embodiment of the present disclosure is applied to information browsing scenarios. The article recommendation method provided by the embodiment of the present disclosure can be applied to the process of users recommending electronic articles, news or books, etc.

[0033] In the related art, in the current recommendation system, the user-to-user collaborative filtering algorithm (U2U2I) is widely used in personalized recommendation scenarios. However, in the content article recommendation scenario, the user-based U2U2I usually ignores the user's actual reading preference for the article content when processing user preferences. Therefore, the related art usually optimizes the recommendation results based on the user's click behavior, ratings, and other explicit feedback on the article. However, the use of explicit information feedback cannot fully reflect the user's actual preference for the article during the reading process.

[0034] In view of this, the present disclosure provides an article recommendation method, which takes into account the user's explicit feedback on the article, such as the number of clicks, and the user's browsing length for each article. By analyzing the browsing length of each article, the user's reading behavior is comprehensively determined, and then article recommendations are achieved based on the similarity between different users.

[0035] Figure 1 FIG. 1 is a flowchart of an article recommendation method according to an exemplary embodiment. Figure 1 As shown, the following steps are included.

[0036] In step S11 , in response to recommending an article to the first user, a second user is determined.

[0037] In the disclosed embodiment, it is determined that the first user and the second user have similar reading preferences. When the first user has an article reading need, an article recommendation list for the second user is determined, and articles are recommended to the first user based on the article recommendation list of the second user.

[0038] The similarity of the reading preferences between the second user and the first user satisfies the similarity condition. The first user can be understood as a user with article recommendation needs, and the second user can be understood as a user who can recommend articles to users with similar reading preferences based on his or her own list of recommended articles.

[0039] In step S12, a set of articles to be recommended of the second user is obtained, where the set of articles to be recommended is determined based on the length and number of views of the articles by the second user.

[0040] In the embodiment of the present disclosure, the set of articles to be recommended for the second user is determined based on the browsing length and browsing times of the articles by the second user.

[0041] Among them, the set of articles to be recommended can be a set of articles browsed by the second user after a certain moment in history. The number of views of each article in the article set is accumulated and determined after a certain moment in history. For example, if the articles to be recommended by the second user are collected from three days ago to the current moment, the number of views of each article in the set of articles to be recommended should be accumulated and summed from this time period.

[0042] In step S13, based on the set of articles to be recommended, articles are recommended to the first user.

[0043] According to the exemplary embodiment of the present disclosure, by introducing the length of the article browsed by the user, the accuracy of determining the user's reading preference is improved. In addition, the user's reading preference is determined based on the two dimensions of the length of the article browsed by the user and the number of times the article is browsed, so as to more accurately reflect the user's real reading interest.

[0044] In one example, for example, by constructing a user profile, a first user whose similarity with a second user satisfies a similarity condition is pre-analyzed.

[0045] In the disclosed embodiment, user portraits of the first user and the second user are constructed based on the attribute characteristics and behavioral characteristics of the users. The various characteristics of the users are converted into numerical values, and the numerical values ​​of different types are normalized or standardized. The different characteristic values ​​are mapped to a unified dimension, and finally the characteristic vector of the first user and the characteristic vector of the second user are generated. The obtained characteristic vectors of the users are used to calculate the user similarity using the cosine similarity calculation method, and the user similarity matrix is ​​calculated to find the first user who is most similar to the second user.

[0046] Among them, attribute characteristics include user's personal information, such as age, gender, etc. Behavioral characteristics include analyzing user's behavioral data, including browsing history, clicks, etc.

[0047] In the disclosed embodiment, based on the similarity between the first user and the second user, if the second user's collection of articles to be recommended includes the article to be recommended and the articles browsed by the first user do not include the article to be recommended, the article to be recommended is recommended to the first user.

[0048] According to an exemplary embodiment of the present disclosure, by using the similarity between a first user and a second user, articles are recommended to the first user based on a set of articles to be recommended determined by utilizing the reading preferences of the second user, thereby reducing the process of re-analyzing the reading preferences of the first user and improving the efficiency of article recommendations between different users.

[0049] In the disclosed embodiment, based on the set of articles to be recommended of the second user, when the second user closes the current article reading and enters the next article reading, articles whose similarity with the articles in the set of articles to be recommended meets the preset article similarity can be recommended to the second user based on the existing set of articles to be recommended.

[0050] In the disclosed embodiment, it is further described that in the article recommendation process, the set of articles to be recommended is determined according to the browsing length and browsing times of the articles by the second user.

[0051] Figure 2 FIG. 1 is a flowchart of determining a set of articles to be recommended according to an exemplary embodiment. Figure 2 As shown, the following steps are included.

[0052] In step S21, the browsing length and browsing times corresponding to each article browsed by the second user are determined, and based on the browsing length and browsing times, a target browsing length and target browsing times are determined.

[0053] In step S22, based on the target browsing length and the target browsing times, the second user's preference for each article is determined.

[0054] In the disclosed embodiment, the user's target browsing length and target browsing times for each article are mapped to the user's preference for each article.

[0055] In step S23, the articles are sorted from high to low according to the preference level, and a preset number of articles ranked first are used as a set of articles to be recommended.

[0056] In the disclosed embodiment, the obtained preference levels of the articles are sorted in descending order, and a preset number of articles ranked first are used as a set of articles to be recommended.

[0057] The preset number of articles may be all articles sorted according to the preference of each user, or may be a certain number of articles set by the user according to the user's needs.

[0058] In the disclosed embodiment, the number of views and the length of views are mapped to the user's preference for reading articles, and there is no need to consider the reading characteristics of different dimensions separately, which improves the convenience of the recommendation process, and recommends subsequent articles based on the articles that the second user has viewed, thereby reducing invalid clicks and bounce rates. By recommending articles based on the preference of each article, the accuracy of article recommendations is further improved.

[0059] In the embodiment of the present disclosure, it is further described that the second user's preference for each article is determined according to the target browsing length and the number of browsing times. Figure 3 The initial preference level is determined in the manner shown.

[0060] Figure 3 The invention is a flow chart showing a method of determining an initial preference degree according to an exemplary embodiment, and includes the following steps.

[0061] In step S31, the exponent of the exp exponential function is determined based on the target browsing length.

[0062] In one example, m represents the target browsing length, and the exponent of the exp exponential function can be 10-20m. The exp exponential function can be, for example:

[0063] e 10-20m

[0064] The exponent is the difference between the first constant and the ratio function, and the ratio function is the product of the target browsing length and the second constant.

[0065] In step S32, the sum of the third constant and the exp exponential function is determined, and the inverse of the sum is mapped to the initial preference degree.

[0066] For example, the initial preference formula can be as follows:

[0067]

[0068] In the embodiment of the present disclosure, when the number of times a user browses different articles is multiple times, the following method is used: Figure 4 The method shown is further explained.

[0069] Figure 4 The invention is a flowchart showing a method for determining a user's article preference according to an exemplary embodiment, and includes the following steps.

[0070] In step S41, the initial preference degree is determined based on the target browsing length.

[0071] Among them, the initial preference level represents the user's preference level corresponding to the maximum browsing length of each article during multiple browsing processes.

[0072] In step S42, the product of the initial preference level and the number of views is determined as the second user's preference level for each article.

[0073] In one example, n represents the number of times a user browses, and the preference level can be:

[0074]

[0075] Among them, f(x) represents the degree of preference.

[0076] In one example, for example, the browsing data of a user at a certain historical moment can be collected to determine the number of views and the length of views mapped to the user's preferences. The number of views and the length of views included in the browsing data of a user at a certain historical moment are input into the data processing model, and an image that conforms to the user's actual browsing situation is fitted through the data processing model. The image features can be as follows: Figure 5 As shown, Figure 5 It is a schematic diagram of a reading situation according to an exemplary embodiment.

[0077] by Figure 5 For example, for an article that a user browses once, as the length of the article that the user browses becomes longer and longer, it can be understood that the user's preference for the article is higher. That is, according to the image with this characteristic change, the function fitting result meets the requirements. Figure 5 The changing trend of the image in the figure. When the value range of the article browsing length ratio f(x) is [0, 1], when the browsing length is close to 0, the preference is almost close to 0, and when the browsing length is close to 1, the preference is almost close to 1. This trend is close to the actual reading habits of users: when a user clicks on an article, if he or she browses a short section and then exits, it means that the user is not very interested in the article; when the user exits when he or she is about to read the entire article, it is possible that the user has roughly understood the content of the entire article, and at this time, the user's needs are satisfied without reading the entire article.

[0078] According to an exemplary embodiment of the present disclosure, by mapping the browsed length and the number of browsed times to the user's preference level, a unified measurement standard is used to determine the user's preference level for the article in accordance with the user's actual browsing habits, thereby improving the convenience of determining the user's reading preference.

[0079] In the embodiment of the present disclosure, in order to avoid interference data in which the user clicks multiple times but browses almost zero content, the following method can be used: Figure 6 The method shown here filters out interfering data.

[0080] Figure 6 A flowchart of determining target data according to an exemplary embodiment is shown, comprising the following steps:

[0081] In step S51, among the articles that have been browsed by the second user, the articles whose browsing times are less than the browsing times threshold are filtered out, and the filtered articles are obtained.

[0082] The browsing count threshold may be understood as the maximum number of times all articles have been read, or may be understood as a value set based on experience.

[0083] In the disclosed embodiment, among the articles browsed by the second user, the articles having a browsing count less than a browsing count threshold and a browsing length greater than a browsing length threshold can be filtered out to obtain the filtered articles.

[0084] In step S52, for each of the filtered articles, the browsing length of each browsing of the same article is obtained, and the maximum browsing length is determined, and the maximum browsing length is used as the target browsing length of the article, and the total number of browsing times of the article is used as the target number of browsing times of the article.

[0085] In the embodiment of the present disclosure, if the number of views is greater than the view count threshold, the article can be preliminarily considered to be an article where the user has made multiple accidental clicks. For example, if the number of views of the article is greater than the view count threshold and the page length is less than the page length threshold, the number of views and page length data corresponding to the article can be understood as interference data, that is, the user may have made multiple accidental clicks on the article.

[0086] In the embodiment of the present disclosure, the article recommendation process is further explained by taking the user's reading articles as an example.

[0087] In the disclosed embodiment, all articles clicked by users after a certain historical moment are obtained, the number of clicks corresponding to each article and the user's browsing length corresponding to each click of the article are saved, and the publication time of all articles clicked by the user is before the historical moment.

[0088] Compare the browsing length of each article, and determine the maximum browsing length of each article as the final browsing length of the article in the reading set. Articles that have been browsed multiple times but each browsing length is almost zero are considered interference articles and are not included in the reading set. For the articles in the reading set, the browsing times and maximum browsing length of each article are mapped to the user's preference for the article.

[0089] According to the order of interest level corresponding to each article in the reading set from high to low, determine the user's article reading preference list. Obtain the reading preference list of the first user and the reading preference list of the second user whose reading preference is similar to that of the first user, and compare the reading preference list of the first user with the reading preference list of the second user. Articles that exist in the reading preference list of the first user but not in the reading preference list of the second user are recommended to the second user.

[0090] Among them, browsing length refers to the ratio of the content of the article read by the user to the total content of the article.

[0091] According to the exemplary embodiment of the present disclosure, the user's actual reading effect dimension of the article is calculated to better represent the user's preference for the article. The user's preference for different articles can be analyzed more accurately, and the user's satisfaction with the recommendation process can be improved. At the same time, this method can be applied to other similar recommendation scenarios and has a certain generalization effect.

[0092] Based on the same concept, the embodiment of the present disclosure also provides an article recommendation device.

[0093] It is understandable that in order to achieve the above functions, the article recommendation device provided by the embodiment of the present disclosure includes hardware structures and / or software modules corresponding to the execution of each function. In combination with the units and algorithm steps of the various examples disclosed in the embodiment of the present disclosure, the embodiment of the present disclosure can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art may use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the technical solution of the embodiment of the present disclosure.

[0094] Figure 7 is a block diagram of an article recommendation device according to an exemplary embodiment. Figure 7 The device 100 includes a determining unit 101, a processing unit 102 and a recommending unit 103.

[0095] The determining unit 101 is configured to determine a second user in response to recommending an article to a first user, wherein the similarity between the second user and the first user satisfies a similarity condition.

[0096] The processing unit 102 is used to obtain a set of articles to be recommended for the second user, where the set of articles to be recommended is determined based on the browsing length and browsing times of the articles by the second user.

[0097] The recommendation unit 103 is used to recommend articles to the first user based on the set of articles to be recommended.

[0098] In one embodiment, the processing unit 102 determines a set of articles to be recommended based on the browsing length and browsing times of the second user's articles in the following manner: determine the browsing length and browsing times corresponding to each article that the second user has browsed, and based on the browsing length and browsing times, determine the target browsing length and target browsing times; based on the target browsing length and target browsing times, determine the second user's preference for each article; sort each article from high to low according to the preference, and use a preset number of articles ranked first as the set of articles to be recommended.

[0099] In one embodiment, the processing unit 102 determines the second user's preference for each article based on the target browsing length and the number of views in the following manner: based on the target browsing length, an initial preference is determined, the initial preference represents the second user's preference for each article for a single browse; the product of the initial preference and the number of views is determined as the second user's preference for each article.

[0100] In one embodiment, the processing unit 102 determines the initial preference level based on the target browsing length in the following manner: based on the target browsing length, determine the exponent of the exp exponential function, the exponent is the difference between the first constant and the ratio function, and the ratio function is the product of the target browsing length and the second constant; determine the sum of the third constant and the exp exponential function, and map the inverse of the sum to the initial preference level.

[0101] In one embodiment, the processing unit 102 determines the target browsing length and the target browsing number based on the browsing length and the browsing number in the following manner: among the articles that the second user has browsed, the articles whose browsing number is less than the browsing number threshold are filtered out to obtain the filtered articles; for each of the filtered articles, the browsing length of each browsing of the same article is obtained respectively, and the maximum browsing length is determined, and the maximum browsing length is used as the target browsing length of the article, and the total number of browsing times of the article is used as the target browsing number of the article.

[0102] In one implementation, the recommendation unit 103 recommends articles to the first user based on the set of articles to be recommended in the following manner: in response to the set of articles to be recommended including the article to be recommended and the articles browsed by the first user not including the article to be recommended, recommending the article to be recommended to the first user.

[0103] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0104] If necessary, Figure 8The architecture shown is only one possible implementation. Figure 8 This is a schematic diagram of the structure of an application embodiment of the electronic device disclosed in the present invention. Figure 8 , which shows a schematic diagram of the structure of an electronic device suitable for implementing a terminal device or a server in an embodiment of the present application. Figure 8 As shown, the electronic device includes a memory for storing a computer program and one or more processors for executing the computer program stored in the memory. In one example, the memory may be a read-only memory (ROM) and / or a random access memory (RAM).

[0105] In one example, one or more processors may be one or more central processing units (CPUs), and / or one or more graphics processing units (GPUs), etc. The processors may perform various appropriate actions and processes according to executable instructions stored in ROM or executable instructions loaded from the storage part to RAM. In one example, the electronic device may also include a communication unit, which may include but is not limited to a network card, which may include but is not limited to an IB (Infiniband) network card. The processor may communicate with the ROM and / or RAM to execute executable instructions, connect to the communication unit through a bus, and communicate with other target devices through the communication unit, thereby completing the operations corresponding to any method provided in the embodiments of the present application, for example, based on the time attenuation coefficient, the user activity is calculated; based on the number of recognitions of the contributed content, the user popularity is calculated; based on the user activity and user popularity, the user's knowledge contribution ability is calculated.

[0106] In addition, various programs and data required for the operation of the device can also be stored in the RAM. The CPU, ROM and RAM are connected to each other via a bus. In the case of RAM, the ROM is an optional module. The RAM stores executable instructions, or writes executable instructions into the ROM at runtime, and the executable instructions enable the processor to perform operations corresponding to any of the above methods of the present invention. The input / output (I / O) interface is also connected to the bus. The communication unit can be integrated or can be configured to have multiple submodules (e.g., multiple IB network cards) and be linked on the bus.

[0107] The following components are connected to the I / O interface: an input part including a keyboard, a mouse, etc.; an output part including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker; a storage part including a hard disk, etc.; and a communication part including a network interface card such as a LAN card, a modem, etc. The communication part performs communication processing via a network such as the Internet. A drive is also connected to the I / O interface as needed. Removable media, such as magnetic disks, optical disks, magneto-optical disks, semiconductor memories, etc., are installed on the drive as needed, so that the computer program read therefrom can be installed into the storage part as needed. It should be noted that Figure 8 The architecture shown is only an optional implementation method. In the specific practice, the above can be modified according to actual needs. Figure 7 The number and type of components can be selected, deleted, added or replaced; different functional components can be set up by separate settings or integrated settings, for example, the GPU and CPU can be set up separately or the GPU can be integrated on the CPU, the communication unit can be set up separately or integrated on the CPU or GPU, etc. These replaceable implementations all fall within the protection scope disclosed in the present invention.

[0108] Fig. 9 FIG. 1 is a schematic diagram showing a structure of an embodiment of an electronic device disclosed in the present invention. Fig. 9 , which shows a schematic diagram of the structure of an electronic device suitable for implementing a terminal device or a server in an embodiment of the present application. Fig. 9 As shown, the electronic device includes a processor and a memory. The electronic device may also include an input / output device. The memory and the input / output device are connected to the processor via a bus. The memory is used to store instructions executed by the processor; the processor is used to call the instructions stored in the memory and execute the method recommended by the article involved in the above embodiment.

[0109] The present disclosure also provides a computer-readable storage medium storing computer-executable instructions. When the computer-executable instructions are executed on a computer, the method for article recommendation in the above embodiment is executed.

[0110] The embodiments of the present disclosure also provide a computer program product including instructions. When the computer program product including the instructions is run on a computer, the computer is enabled to execute the method for article recommendation according to the above embodiments.

[0111] In one or more optional implementations, the disclosed embodiment further provides a computer-readable storage medium for storing computer-readable instructions, which, when executed, causes a computer to execute the method for article recommendation in any of the above possible implementations. In another optional example, the computer program product is specifically embodied as a software product, such as a software development kit (SDK), etc.

[0112] Although operations are described in a particular order in the drawings, this should not be understood as requiring that the operations be performed in the particular order shown or in serial order, or that all shown operations be performed to achieve desired results. In certain circumstances, multitasking and parallel processing may be advantageous.

[0113] The methods and apparatus of the present disclosure can be implemented using standard programming techniques, using rule-based logic or other logic to implement various method steps. It should also be noted that the words "apparatus" and "module" used herein and in the claims are intended to include implementations using one or more lines of software code and / or hardware implementations and / or devices for receiving input.

[0114] Any steps, operations or procedures described herein may be performed or implemented using one or more hardware or software modules, either alone or in combination with other devices. In one embodiment, the software module is implemented using a computer program product including a computer-readable medium containing computer program code, which can be executed by a computer processor to perform any or all of the described steps, operations or procedures.

[0115] The foregoing description of the implementation of the present disclosure has been given for the purpose of illustration and description. The foregoing description is not exhaustive nor is it intended to limit the present disclosure to the exact form disclosed, and various variations and modifications may exist in accordance with the above teachings or may be obtained from the practice of the present disclosure. These embodiments are selected and described in order to illustrate the principles of the present disclosure and its practical application, so that those skilled in the art can utilize the present disclosure in various embodiments and various modifications suitable for the specific purpose contemplated.

[0116] It is to be understood that in the present disclosure, "plurality" refers to two or more than two, and other quantifiers are similar. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the associated objects before and after are in an "or" relationship. The singular forms "a", "the" and "the" are also intended to include plural forms, unless the context clearly indicates other meanings.

[0117] It is further understood that the terms "first", "second", etc. are used to describe various information, but such information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other, and do not indicate a specific order or degree of importance. In fact, the expressions "first", "second", etc. can be used interchangeably. For example, without departing from the scope of the present disclosure, the first information can also be referred to as the second information, and similarly, the second information can also be referred to as the first information.

[0118] It will be further understood that the terms “center”, “longitudinal”, “lateral”, “front”, “back”, “up”, “down”, “left”, “right”, “vertical”, “horizontal”, “top”, “bottom”, “inside”, “outside”, etc., indicating orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, are only for the convenience of describing the present embodiment and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation.

[0119] It can be further understood that, unless otherwise specified, “connection” includes a direct connection without other components between the two, and also includes an indirect connection with other components between the two.

[0120] It is further understood that, although the operations are described in a specific order in the drawings in the embodiments of the present disclosure, it should not be understood as requiring the operations to be performed in the specific order shown or in a serial order, or requiring the execution of all the operations shown to obtain the desired results. In certain environments, multitasking and parallel processing may be advantageous.

[0121] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. This application is intended to cover any modifications, uses or adaptations of the present disclosure, which follow the general principles of the present disclosure and include common knowledge or customary technical means in the art that are not disclosed in the present disclosure.

[0122] It should be understood that the present disclosure is not limited to the precise structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the scope of the appended claims.

Claims

1. A method for recommending articles, characterized in that: The method comprises: In response to recommending an article to the first user, determining a second user, wherein a similarity between the second user and the first user satisfies a similarity condition; Acquire a set of articles to be recommended by the second user, where the set of articles to be recommended is determined based on the length and number of views of the articles by the second user; Based on the set of articles to be recommended, recommend articles to the first user.

2. The method according to claim 1, characterized in that The set of articles to be recommended is determined based on the browsing length and browsing times of the articles by the second user in the following manner: Determine the browsing length and browsing times corresponding to each article browsed by the second user, and determine the target browsing length and target browsing times based on the browsing length and browsing times; Determining the second user's preference for each of the articles based on the target browsing length and the target browsing number; The articles are sorted from high to low according to the preference degree, and a preset number of articles ranked first are used as a set of articles to be recommended.

3. The method according to claim 2, characterized in that The determining, based on the target browsing length and the browsing times, the second user's preference for each of the articles includes: Based on the target browsing length, determining an initial preference degree, wherein the initial preference degree represents the second user's preference degree for a single browsing of each article; The product of the initial preference degree and the number of views is determined as the second user's preference degree for each article.

4. The method according to claim 3, characterized in that The determining of the initial preference degree based on the target browsing length includes: Based on the target browsing length, determining the exponent of the exp exponential function, the exponent is the difference between the first constant and the ratio function, and the ratio function is the product of the target browsing length and the second constant; A sum value of a third constant and the exp exponential function is determined, and the inverse of the sum value is mapped to an initial preference degree.

5. The method according to claim 2, characterized in that: The determining of a target browsing length and a target browsing number based on the browsing length and the browsing number includes: Filtering the articles whose browsing times are less than a browsing times threshold among the articles browsed by the second user, and obtaining the filtered articles; For each of the articles after the screening, the browsing length of each browsing of the same article is obtained respectively, and the maximum browsing length is determined, and the maximum browsing length is used as the target browsing length of the article, and the total number of browsing times of the article is used as the target number of browsing times of the article.

6. The method according to claim 1, characterized in that The recommending articles to the first user based on the set of articles to be recommended includes: In response to the to-be-recommended article being included in the to-be-recommended article set and the to-be-recommended article not being included in the articles browsed by the first user, the to-be-recommended article is recommended to the first user.

7. An article recommendation device, characterized in that: The device comprises: a determining unit, configured to determine a second user in response to recommending an article to the first user, wherein a similarity between the second user and the first user satisfies a similarity condition; A processing unit, configured to obtain a set of articles to be recommended for the second user, wherein the set of articles to be recommended is determined based on the length and number of views of the articles by the second user; A recommendation unit is used to recommend articles to the first user based on the set of articles to be recommended.

8. The device according to claim 7, characterized in that The processing unit determines the set of articles to be recommended based on the browsing length and browsing times of the articles by the second user in the following manner: Determine the browsing length and browsing times corresponding to each article browsed by the second user, and determine the target browsing length and target browsing times based on the browsing length and browsing times; Determining the second user's preference for each of the articles based on the target browsing length and the target browsing number; The articles are sorted from high to low according to the preference degree, and a preset number of articles ranked first are used as a set of articles to be recommended.

9. The device according to claim 8, characterized in that The processing unit determines the second user's preference for each of the articles based on the target browsing length and the number of browsing times in the following manner: Based on the target browsing length, determining an initial preference degree, wherein the initial preference degree represents the second user's preference degree for a single browsing of each article; The product of the initial preference degree and the number of views is determined as the second user's preference degree for each article.

10. The device according to claim 9, characterized in that The processing unit determines the initial preference degree based on the target browsing length in the following manner: Based on the target browsing length, determining the exponent of the exp exponential function, the exponent is the difference between the first constant and the ratio function, and the ratio function is the product of the target browsing length and the second constant; A sum value of a third constant and the exp exponential function is determined, and the inverse of the sum value is mapped to an initial preference degree.

11. The device according to claim 8, characterized in that The processing unit determines the target browsing length and the target browsing number based on the browsing length and the browsing number in the following manner: Filtering the articles that have been browsed by the second user and whose browsing times are less than a browsing times threshold, and obtaining the filtered articles; For each of the articles after the screening, the browsing length of each browsing of the same article is obtained respectively, and the maximum browsing length is determined, and the maximum browsing length is used as the target browsing length of the article, and the total number of browsing times of the article is used as the target number of browsing times of the article.

12. The device according to claim 7, characterized in that The recommendation unit recommends articles to the first user based on the set of articles to be recommended in the following manner: In response to the to-be-recommended article being included in the to-be-recommended article set and the to-be-recommended article not being included in the articles browsed by the first user, the to-be-recommended article is recommended to the first user.

13. An electronic device, characterized in that: include: processor; a memory for storing processor-executable instructions; The processor is configured to execute the method according to any one of claims 1 to 6.

14. A storage medium, characterized in that: The storage medium stores instructions, and when the instructions in the storage medium are executed by a processor, the method according to any one of claims 1 to 6 can be executed.

Images