E-book similarity calculation method, electronic device and storage medium

Abstract

The present invention discloses an e-book similarity calculation method, an electronic device and a storage medium. The method comprises: obtaining a first historical user of a first e-book and a second historical user of a second e-book, determining a user list containing common users between the first historical user and the second historical user; screening the same e-books between any two users according to the historical behavior information of the users in the user list, and determining the common e-book data; calculating the negative correlation weight factor of the user to the e-book according to the historical behavior information of the users in the user list to the e-book; performing user loop calculation on the user list according to the common e-book data and the negative correlation weight factor of any two users, and obtaining the similarity between the first e-book and the second e-book. Mining the similarity between e-books according to the common e-book data; calculating the negative correlation weight factor of the e-book using the user historical behavior information, and adjusting the influence of the user behavior on the e-book similarity to make it more accurate.

Description

Technical Field

[0001] The present invention relates to the field of electronic books, and in particular to an electronic book similarity calculation method, an electronic device and a storage medium. Background Art

[0002] E-books provide users with a convenient reading experience, and more and more users choose to read e-books. When reading, users will choose e-books based on their own preferences. E-book recommendations can help users find their favorite e-books more quickly and reduce their search time.

[0003] When recommending e-books to users, similar types of e-books are generally recommended to them based on the user's preference for e-books. Therefore, when recommending e-books, it is necessary to first determine the similarity between e-books. When determining the similarity of e-books in the prior art, most of them are based on the user's reading behavior. For example, if different users have read e-book 1 and e-book 2, it can be indirectly shown that there is a certain degree of similarity between e-book 1 and e-book 2. However, this method also has defects. For example, among the e-books read by different users, only e-book 1 and e-book 2 are the same, and there may be relatively large differences in the reading preferences of different users. The similarity between e-book 1 and e-book 2 determined only by the above method is inaccurate, resulting in the recommended e-books not meeting user needs and a low click-through rate. Summary of the invention

[0004] In view of the above problems, the present invention is proposed to provide an e-book similarity calculation method, electronic device and storage medium that overcome the above problems or at least partially solve the above problems.

[0005] According to one aspect of the present invention, there is provided a method for calculating electronic book similarity, comprising:

[0006] Acquire a first historical user of a first electronic book and a second historical user of a second electronic book, and determine a user list including common users between the first historical user and the second historical user;

[0007] According to the historical behavior information of any two users in the user list, the same e-books between any two users are screened to determine the common e-book data;

[0008] The negative correlation weight factor of the user to the e-book is calculated based on the historical behavior information of the user to the e-book in the user list;

[0009] According to the common e-book data of any two users and the negative correlation weight factor, a user loop calculation is performed on the user list to obtain the similarity between the first e-book and the second e-book.

[0010] According to another aspect of the present invention, there is provided an electronic device, comprising: a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus;

[0011] The memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform the following operations:

[0012] Acquire a first historical user of a first electronic book and a second historical user of a second electronic book, and determine a user list including common users between the first historical user and the second historical user;

[0013] According to the historical behavior information of any two users in the user list, the same e-books between any two users are screened to determine the common e-book data;

[0014] The negative correlation weight factor of the user to the e-book is calculated based on the historical behavior information of the user to the e-book in the user list;

[0015] According to the common e-book data of any two users and the negative correlation weight factor, a user loop calculation is performed on the user list to obtain the similarity between the first e-book and the second e-book.

[0016] According to another aspect of the present invention, a computer storage medium is provided, in which at least one executable instruction is stored, and the executable instruction enables a processor to execute operations corresponding to any of the above-mentioned methods for calculating electronic book similarity.

[0017] According to the e-book similarity calculation method, electronic device and storage medium provided by the present invention, when calculating the e-book similarity, the common users of the first e-book and the second e-book are first determined, and then the same e-books of two users are determined based on the historical behavior information of the common users. The similarity between e-books can be mined based on the common e-book data; the negative correlation weight factor of the e-book is calculated using the user historical behavior information, and the influence of the user historical behavior on the e-book similarity is adjusted to make it more accurate.

[0018] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0020] Figure 1 A schematic diagram showing a flow chart of a method for calculating electronic book similarity according to an embodiment of the present invention;

[0021] Figure 2 A schematic diagram showing a flow chart of a method for calculating electronic book similarity according to another embodiment of the present invention;

[0022] Figure 3 A schematic structural diagram of an electronic device according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0023] Embodiment 1

[0024] Figure 1 FIG. 4 is a flow chart showing a method for calculating electronic book similarity according to an embodiment of the present invention. Figure 1 As shown, the method comprises the following steps:

[0025] Step S101 : obtaining a first historical user of a first electronic book and a second historical user of a second electronic book, and determining a user list including common users between the first historical user and the second historical user.

[0026] In this embodiment, the first electronic book and the second electronic book are any two different electronic books, and calculating the similarity between the first electronic book and the second electronic book is calculating the similarity between any two electronic books.

[0027] Specifically, obtain a first historical user who has a history of interactive behavior with the first e-book. The historical interactive behavior includes, for example, reading the first e-book, searching the first e-book, viewing the comment information and recommendation information related to the first e-book, adding the first e-book to the bookshelf, etc. By searching the historical behavior information of the user, it is possible to query whether the user has a history of interactive behavior with the first e-book, thereby determining the first historical user. Query whether the user has a history of interactive behavior with the second e-book, thereby determining the second historical user. Find common users that exist in both the first historical user and the second historical user, and obtain a user list consisting of common users. Each user in the user list is a user who has a history of interactive behavior with both the first e-book and the second e-book.

[0028] Step S102, based on the historical behavior information of any two users in the user list, the same e-books between any two users are screened to determine the common e-book data.

[0029] Different users have historical interactions with the first e-book and the second e-book, and it can be judged to a certain extent that the first e-book and the second e-book are similar. For example, if two different users have read the first e-book and the second e-book, and the two users read only the first e-book and the second e-book, it means that the two users have very different reading preferences, and there is similarity between the first e-book and the second e-book. Therefore, the first e-book and the second e-book are the same e-books read. That is, the similarity between the first e-book and the second e-book is related to the existence of the same e-books between the users.

[0030] The historical behavior information of each user in the user list is obtained, and the historical behavior information includes historical e-books, interaction information between users and e-books, etc. Any two users in the user list are selected, and the historical e-books in the historical behavior information of any two users are compared to see whether there is the same e-book, so as to obtain the same e-books of any two users and determine the common e-book data of any two users. The common e-book data may be the number and type of the same e-books of any two users. When the common e-book data is larger, the reading preferences of any two users are more similar, and the first e-book may also have similarities with other e-books, and the similarity with the second e-book may be relatively low.

[0031] Step S103, calculating the negative correlation weight factor of the user to the e-book according to the historical behavior information of the user to the e-book in the user list.

[0032] In addition to the common e-book data obtained in step S102, this embodiment also considers the negative impact of the user's historical behavior information on the e-book similarity, and calculates the negative correlation weight factor of the user's e-book based on the user's historical behavior information on the e-book in the user list, and adjusts the impact of the user's historical behavior on the e-book similarity. The negative correlation weight factor is based on the user's historical behavior information on the e-book, and calculates different data from different aspects of the user's behavior, such as the time of the user's interaction with the first e-book and the second e-book, and whether the e-book is read in a relatively recent time; the number of e-books in the common e-book data read by the user, different types, etc., to adjust the impact of the user's reading preferences on the e-book similarity, the impact of different types of e-books on the e-book similarity, etc.

[0033] Step S104: performing user loop calculation on the user list according to the common e-book data of any two users and the negative correlation weight factor to obtain the similarity between the first e-book and the second e-book.

[0034] When the number of e-books in the common e-book data is greater, that is, the more common e-books any two users have interacted with, the similarity between the first e-book and the second e-book is reduced. If the common e-books of user A and user B only include the first e-book and the second e-book, it means that the first e-book and the second e-book are similar. If the common e-books of user A and user B include the first e-book, the second e-book, e-book 3, e-book 4, e-book 5, etc., at this time, there are many common e-books between user A and user B, and it cannot be directly determined whether the first e-book is similar to the second e-book, or similar to e-book 3, e-book 4, e-book 5, etc., which affects the similarity between the first e-book and the second e-book. When the common e-books of any two users involve a variety of categories, the users' reading preferences are wide, which also affects the calculation of the similarity between the first e-book and the second e-book accordingly.

[0035] The negative correlation weight factor further adjusts the impact of the user's historical behavior on the e-book similarity. For example, if the time when the user interacts with the first e-book and the second e-book is similar, the similarity between the first e-book and the second e-book will be improved; if the user interacts with different types of e-books, the similarity between the first e-book and the second e-book will be reduced.

[0036] When calculating the similarity between the first electronic book and the second electronic book, for the common electronic book data of any two users, the common electronic book data can be processed in a reciprocal manner, and then multiplied by the negative correlation weight factor to calculate the similarity of the electronic books. For multiple users included in the user list, multiple loop calculations are performed on any two users, and the calculation results of any two users are accumulated to finally obtain the similarity between the first electronic book and the second electronic book.

[0037] According to the e-book similarity calculation method provided by the present invention, when calculating the e-book similarity, the common users of the first e-book and the second e-book are first determined, and then the same e-books of two users are determined based on the historical behavior information of the common users. The similarity between e-books can be mined based on the common e-book data; the negative correlation weight factor of the e-book is calculated using the user historical behavior information, and the influence of the user historical behavior on the e-book similarity is adjusted to make it more accurate.

[0038] Embodiment 2

[0039] Figure 2 FIG. 4 is a flow chart showing a method for calculating electronic book similarity according to another embodiment of the present invention. Figure 2 As shown, the method comprises the following steps:

[0040] Step S201: Acquire a first historical user of a first electronic book and a second historical user of a second electronic book, and determine a user list including common users between the first historical user and the second historical user.

[0041] Step S202, based on the historical behavior information of any two users in the user list, the same e-books between any two users are screened to determine the common e-book data.

[0042] The above steps refer to the description of steps S101-S102 in the first embodiment, and will not be repeated here.

[0043] Step S203: Calculate a time decay factor based on the historical behavior information of the users in the user list on the e-book.

[0044] In this embodiment, the negative correlation weight factor may include, for example, a time decay factor, which is calculated based on time information in the interaction information between the user and the e-book.

[0045] Specifically, the interaction information between the user and the e-book includes, for example, the reading time of the e-book. When obtaining the historical behavior information of the user, the e-books read by the user can be sorted according to the reading time, and the first reading time of the user reading the first e-book and the second reading time of the user reading the second e-book can be obtained. The reading time difference can be calculated based on the first reading time and the second reading time. When the reading time difference is smaller, it means that the interval between the user reading the first e-book and the second e-book is shorter. For example, the reading time difference between the first e-book and the second e-book is 1 day, and the user reads the two books in a short period of time, then the similarity of the two books is higher; when the reading time difference is larger, it means that the interval between the user reading the first e-book and the second e-book is longer. For example, the reading time difference between the first e-book and the second e-book is 1 year, and the user reads the two books 1 year apart, then the similarity of the two books is lower. According to the reading time difference and the reading time coefficient (pre-set according to the implementation situation, not limited here), the time decay factor of the e-book is calculated. The reading time difference is inversely proportional to the time decay factor, that is, the larger the reading time difference, the smaller the time decay factor; the smaller the reading time difference, the larger the time decay factor, and the corresponding adjustment of the obtained negative correlation weight factor has an impact on the similarity of the e-books.

[0046] Alternatively, further considering that if the user adds an e-book to the bookshelf, it indicates that the book is more in line with the user's preferences, the interaction information between the user and the e-book may also include the bookshelf time when the e-book is added to the bookshelf. When obtaining the user's historical behavior information, the bookshelf time when the user operates the e-book to add it to the bookshelf is obtained, and the first bookshelf time when the user adds the first e-book to the bookshelf and the second bookshelf time when the user adds the second e-book to the bookshelf are obtained. The bookshelf time difference between the first e-book and the second e-book can be calculated according to the first bookshelf time and the second bookshelf time. When the bookshelf time difference is smaller, it means that the time interval between the user adding the first e-book and the second e-book to the bookshelf is shorter. For example, if the bookshelf time difference between the first e-book and the second e-book is 1 day, the user operates the two books in a relatively close time, and the similarity between the two books is high; when the bookshelf time difference is larger, it means that the time interval between the user adding the first e-book and the second e-book to the bookshelf is longer. For example, if the bookshelf time difference between the first e-book and the second e-book is 1 year, the user adds the second book to the bookshelf 1 year after the first book, and the similarity between the two books is lower. The time decay factor of the e-book is calculated based on the bookshelf time difference and the bookshelf time coefficient (pre-set according to the implementation situation, not limited here). The bookshelf time difference is inversely proportional to the time decay factor, that is, the larger the bookshelf time difference, the smaller the time decay factor; the smaller the bookshelf time difference, the larger the time decay factor, thereby obtaining a negative correlation weight factor to adjust the impact on the similarity of the e-book.

[0047] The time decay factor is calculated based on the time information difference between the user's interaction behavior with the first e-book and the second e-book, and the e-book similarity is adjusted by negative correlation weight. The time information difference in the above example can be adjusted according to the implementation situation, and is not limited here.

[0048] Step S204, calculating a user's high-heat penalty factor according to the historical behavior information of the user on the e-book in the user list.

[0049] There are many kinds of e-books. When users interact with different kinds of e-books, a large number of e-books interacted by users will also affect the calculation of the similarity of e-books. Therefore, in this embodiment, the negative correlation weight factor also includes a user high heat penalty factor, and the negative correlation weight is adjusted according to the number of e-books interacted by the user. Specifically, the number of e-books interacted in the historical behavior information of the user in the user list is counted in units of users. According to the number of interacted e-books, the user high heat penalty factor of the e-book is obtained by weight reduction calculation. For example, the user high heat penalty factor is obtained by calculating the inverse of the number, or the user high heat penalty factor is obtained by calculating the inverse after adjusting the number (such as calculating the square root of the number, etc.), which is not limited here. The number of interacted e-books is inversely proportional to the user high heat penalty factor, that is, the larger the number, the smaller the user high heat penalty factor, and the smaller the number, the larger the user high heat penalty factor. Correspondingly, if the more e-books a user reads, the smaller the similarity between e-books, and the calculated value of the e-book similarity is reduced by a smaller user high heat penalty factor.

[0050] The execution order of step S203 and step S204 is not limited here and can be adjusted according to the implementation situation.

[0051] Step S205, for any two users in the user list, calculate the reciprocal of the sum of the number of e-books in the common e-book data of any two users and the preset similarity coefficient, multiply the reciprocal with the time decay factor and / or the user high heat penalty factor to obtain the similarity component between any two users.

[0052] Step S206: Perform user loop calculation on any two users in the user list, and accumulate the obtained similarity components to obtain the similarity between the first electronic book and the second electronic book.

[0053] Select any two users in the user list, user A and user B, and calculate the reciprocal of the sum of the number of e-books in the common e-book data of user A and user B and a preset similarity coefficient (set according to the implementation situation, such as 1, which is not limited here). The reciprocal is used to indicate the frequency of the common e-book data between any two users. The similarity between the first e-book and the second e-book is related to the number of e-books in the common e-book data of any two users. If the number of e-books is large, the similarity between the first e-book and the second e-book will be lower. After the reciprocal is calculated, the reciprocal is further multiplied by the time decay factor, or by the user's high heat penalty factor, or by the time decay factor and the user's high heat penalty factor, to adjust the impact of the user's historical interaction behavior on the e-book similarity, and obtain the similarity component between any two users.

[0054] For the user list, two users are randomly selected each time for user loop calculation until all users in the user list are calculated with other users to obtain similarity components, and the similarity components between any two users are accumulated to obtain the similarity between the first e-book and the second e-book.

[0055] Step S207, obtaining historical e-books in the historical behavior information of the user, and screening e-books whose similarity with the historical e-books is higher than a preset similarity threshold as the e-books to be recommended.

[0056] The above-mentioned e-book similarity can be used to help users recommend e-books. Specifically, historical e-books in the user's historical behavior information are obtained, and historical e-books are e-books that the user has interacted with. According to the e-book similarity, e-books with similarities to historical e-books that are higher than a preset similarity threshold are screened out, and the e-books are recommended to the user as e-books to be recommended. The e-book similarity is higher than the preset similarity threshold, which ensures that the e-books recommended to the user are highly similar to the historical e-books that the user has read and interacted with, and are in line with the user's reading preferences. This can improve the user's click-through rate for recommended e-books, shorten the user's time to search for e-books, and improve the user's reading experience.

[0057] According to the e-book similarity calculation method provided by the present invention, a time decay factor and / or a user high heat penalty factor are introduced to further adjust the influence of the user's historical behavior information on the e-book similarity, thereby improving the accuracy of the e-book similarity calculation. E-book recommendations for users based on e-book similarity can make the recommended e-books more in line with user preferences, facilitate users to find their favorite e-books more quickly, and improve the user's reading experience.

[0058] Embodiment 3

[0059] Embodiment 3 of the present application provides a non-volatile computer storage medium, wherein the computer storage medium stores at least one executable instruction, and the computer executable instruction can execute the e-book similarity calculation method in any of the above method embodiments.

[0060] The executable instructions may be specifically used to cause the processor to perform the following operations:

[0061] The first historical user of the first electronic book and the second historical user of the second electronic book are obtained, and a user list containing common users between the first historical user and the second historical user is determined; based on the historical behavior information of any two users in the user list, the same electronic books between any two users are screened to determine the common electronic book data; based on the historical behavior information of the users in the user list on the electronic books, a negative correlation weight factor of the users on the electronic books is calculated; based on the common electronic book data and the negative correlation weight factor of any two users, a user loop calculation is performed on the user list to obtain the similarity between the first electronic book and the second electronic book.

[0062] In an optional implementation, the historical behavior information includes historical e-books and interaction information between the user and the e-book; the negative correlation weight factor includes a time decay factor, which is calculated based on time information in the interaction information between the user and the e-book.

[0063] In an optional embodiment, the interaction information between the user and the e-book includes the reading time of the e-book; the executable instructions further cause the processor to perform the following operations: obtain a first reading time when the user reads a first e-book, and a second reading time when the user reads a second e-book; calculate a time decay factor of the e-book based on a reading time difference between the first reading time and the second reading time, and a reading time coefficient; wherein the reading time difference is inversely proportional to the time decay factor.

[0064] In an optional embodiment, the interaction information between the user and the e-book includes the shelf time when the e-book is added to the bookshelf; the executable instructions further cause the processor to perform the following operations: obtain the first bookshelf time when the user adds the first e-book to the bookshelf, and the second bookshelf time when the user adds the second e-book to the bookshelf; calculate the time decay factor of the e-book based on the shelf time difference between the first bookshelf time and the second bookshelf time, and the bookshelf time coefficient; wherein the bookshelf time difference is inversely proportional to the time decay factor.

[0065] In an optional embodiment, the negative correlation weight factor also includes a user high heat penalty factor; the executable instructions further cause the processor to perform the following operations: count the number of e-books interacted in the historical behavior information of users in the user list on a user basis; based on the number, calculate the user high heat penalty factor of the e-book by degrading the weight; wherein the number is inversely proportional to the user high heat penalty factor.

[0066] In an optional embodiment, the executable instructions further cause the processor to perform the following operations: for any two users in the user list, calculate the inverse of the sum of the number of e-books in the common e-book data of any two users and a preset similarity coefficient, multiply the inverse by the time decay factor and / or the user high heat penalty factor to obtain the similarity component between any two users; perform user loop calculation on any two users in the user list, and accumulate the obtained similarity components to obtain the similarity between the first e-book and the second e-book.

[0067] In an optional implementation, the executable instructions further cause the processor to perform the following operations: obtain historical e-books in the user's historical behavior information, and select e-books whose similarity with the historical e-books is higher than a preset similarity threshold as e-books to be recommended.

[0068] Embodiment 4

[0069] Figure 3A schematic structural diagram of an electronic device according to a fourth embodiment of the present invention is shown. The specific embodiment of the present invention does not limit the specific implementation of the electronic device.

[0070] like Figure 3 As shown, the electronic device may include: a processor (processor) 302 , a communication interface (Communications Interface) 304 , a memory (memory) 306 , and a communication bus 308 .

[0071] in:

[0072] The processor 302 , the communication interface 304 , and the memory 306 communicate with each other via the communication bus 308 .

[0073] The communication interface 304 is used to communicate with other devices such as clients or other servers.

[0074] The processor 302 is used to execute the program 310, and specifically can execute the relevant steps in the above-mentioned embodiment of the method for calculating the similarity of electronic books.

[0075] Specifically, the program 310 may include program codes, which include computer operation instructions.

[0076] The processor 302 may be a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiment of the present invention. The one or more processors included in the server may be processors of the same type, such as one or more CPUs, or processors of different types, such as one or more CPUs and one or more ASICs.

[0077] The memory 306 is used to store the program 310. The memory 306 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0078] The program 310 may be specifically configured to enable the processor 302 to perform the following operations:

[0079] The first historical user of the first electronic book and the second historical user of the second electronic book are obtained, and a user list containing common users between the first historical user and the second historical user is determined; based on the historical behavior information of any two users in the user list, the same electronic books between any two users are screened to determine the common electronic book data; based on the historical behavior information of the users in the user list on the electronic books, a negative correlation weight factor of the users on the electronic books is calculated; based on the common electronic book data and the negative correlation weight factor of any two users, a user loop calculation is performed on the user list to obtain the similarity between the first electronic book and the second electronic book.

[0080] In an optional implementation, the historical behavior information includes historical e-books and interaction information between the user and the e-book; the negative correlation weight factor includes a time decay factor, which is calculated based on time information in the interaction information between the user and the e-book.

[0081] In an optional embodiment, the interaction information between the user and the e-book includes the reading time of the e-book; the program 310 is used to enable the processor 302 to obtain the first reading time of the user reading the first e-book and the second reading time of the user reading the second e-book; based on the reading time difference between the first reading time and the second reading time and the reading time coefficient, the time decay factor of the e-book is calculated; wherein the reading time difference is inversely proportional to the time decay factor.

[0082] In an optional embodiment, the interaction information between the user and the e-book includes the shelf time when the e-book is added to the bookshelf; the program 310 is used to enable the processor 302 to obtain the first bookshelf time when the user adds the first e-book to the bookshelf, and the second bookshelf time when the user adds the second e-book to the bookshelf; based on the shelf time difference between the first bookshelf time and the second bookshelf time, and the bookshelf time coefficient, the time decay factor of the e-book is calculated; wherein the bookshelf time difference is inversely proportional to the time decay factor.

[0083] In an optional embodiment, the negative correlation weight factor also includes a user high heat penalty factor; the program 310 is used to enable the processor 302 to count the number of e-books interacted in the historical behavior information of users in the user list on a user basis; based on the number, the user high heat penalty factor of the e-book is obtained by de-weighting calculation; wherein the number is inversely proportional to the user high heat penalty factor.

[0084] In an optional embodiment, the program 310 is used to enable the processor 302 to calculate the inverse of the sum of the number of e-books in the common e-book data of any two users and a preset similarity coefficient, multiply the inverse by the time decay factor and / or the user high heat penalty factor to obtain the similarity component between any two users; perform user loop calculation on any two users in the user list, and accumulate the obtained similarity components to obtain the similarity between the first e-book and the second e-book.

[0085] In an optional implementation, the program 310 is used to enable the processor 302 to obtain historical e-books in the historical behavior information of the user, and select e-books whose similarity with the historical e-books is higher than a preset similarity threshold as the e-books to be recommended.

[0086] The specific implementation of each step in program 310 can refer to the corresponding description of the corresponding steps in the above-mentioned e-book similarity calculation embodiment, which will not be repeated here. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working process of the above-described devices and modules can refer to the corresponding process description in the above-mentioned method embodiment, which will not be repeated here.

[0087] Through the solution provided by this embodiment, when calculating the similarity of e-books, the common users of the first e-book and the second e-book are first determined, and then the common e-books of two users are determined based on the historical behavior information of the common users. The similarity between e-books can be mined based on the common e-book data; the negative correlation weight factor of the e-book is calculated using the user's historical behavior information, and the influence of the user's historical behavior on the e-book similarity is adjusted to make it more accurate.

[0088] In the description provided herein, a large number of specific details are described. However, it is understood that embodiments of the present invention can be practiced without these specific details. In some instances, well-known methods, structures and techniques are not shown in detail so as not to obscure the understanding of this description.

[0089] Similarly, it should be understood that in order to streamline the present disclosure and aid in understanding one or more of the various inventive aspects, in the above description of exemplary embodiments of the present invention, various features of the present invention are sometimes grouped together into a single embodiment, figure, or description thereof. However, this disclosed method should not be interpreted as reflecting the intention that the claimed invention requires more features than those expressly recited in each claim. Rather, as reflected in the claims, inventive aspects lie in less than all of the features of the individual embodiments previously disclosed. Therefore, the claims that follow the detailed description are hereby expressly incorporated into the detailed description, with each claim itself serving as a separate embodiment of the present invention.

[0090] Those skilled in the art will appreciate that the modules in the devices in the embodiments may be adaptively changed and arranged in one or more devices different from the embodiments. The modules or units or components in the embodiments may be combined into one module or unit or component, and in addition they may be divided into a plurality of submodules or subunits or subcomponents. Except that at least some of such features and / or processes or units are mutually exclusive, all features disclosed in this specification (including the accompanying claims, abstracts and drawings) and all processes or units of any method or device disclosed in this manner may be combined in any combination. Unless otherwise expressly stated, each feature disclosed in this specification (including the accompanying claims, abstracts and drawings) may be replaced by an alternative feature providing the same, equivalent or similar purpose.

[0091] In addition, those skilled in the art will appreciate that, although some embodiments described herein include certain features included in other embodiments but not other features, the combination of features of different embodiments is meant to be within the scope of the present invention and form different embodiments. For example, in the claims, any one of the claimed embodiments may be used in any combination.

[0092] It should be noted that the above embodiments illustrate the present invention rather than limit it, and that those skilled in the art may design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference symbol between brackets should not be constructed as a limitation to the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "one" or "an" preceding an element does not exclude the presence of a plurality of such elements. The present invention can be implemented by means of hardware including a number of different elements and by means of a suitably programmed computer. The use of the words first, second, and third, etc. does not indicate any order. These words can be interpreted as names.

Claims

1. A method for calculating e-book similarity, comprising: Obtaining a first historical user of a first e-book and a second historical user of a second e-book, and determining a user list including common users between the first historical user and the second historical user; screening the same e-books between any two users based on historical behavior information of any two users in the user list to determine the common e-book data; Calculating a negative correlation weight factor of the user for the e-book based on the historical behavior information of the user for the e-book in the user list; Based on the common e-book data of any two users, the reciprocal of the sum of the number of e-books in the common e-book data of any two users and a preset similarity coefficient is calculated. Based on the reciprocal and a negative correlation weight factor, a user loop calculation is performed on the user list to obtain the similarity between the first e-book and the second e-book.

2. The method according to claim 1, wherein The historical behavior information includes historical e-books and information about the interaction between the user and the e-books; The negative correlation weight factor includes a time decay factor, and the time decay factor is calculated based on time information in the interaction information between the user and the e-book.

3. The method according to claim 2, wherein: The interaction information between the user and the e-book includes the reading time of the e-book; The step of calculating the negative correlation weight factor of the user's e-book behavior based on the historical e-book behavior information of the user in the user list further includes: Obtaining a first reading time of a user reading a first e-book and a second reading time of a user reading a second e-book; A time decay factor of the e-book is calculated based on the reading time difference between the first reading time and the second reading time and a reading time coefficient; wherein the reading time difference is inversely proportional to the time decay factor.

4. The method according to claim 2, wherein: The interaction information between the user and the e-book includes the shelf time when the e-book is added to the bookshelf; The step of calculating the negative correlation weight factor of the user's e-book behavior based on the historical e-book behavior information of the user in the user list further includes: Obtaining a first bookshelf time when a user adds a first e-book to a bookshelf, and a second bookshelf time when a user adds a second e-book to a bookshelf; A time decay factor of the e-book is calculated based on a shelf time difference between the first bookshelf time and the second bookshelf time and a bookshelf time coefficient; wherein the shelf time difference is inversely proportional to the time decay factor.

5. The method according to any one of claims 1 to 4, wherein The negative correlation weight factor also includes a user high fever penalty factor; The step of calculating the negative correlation weight factor of the user's e-book behavior based on the historical e-book behavior information of the user in the user list further includes: Counting the number of e-books interacted with in the historical behavior information of users in the user list on a user-by-user basis; According to the number, a user high-heat penalty factor of the e-book is obtained by weight reduction calculation; wherein the number is inversely proportional to the user high-heat penalty factor.

6. The method according to any one of claims 1 to 4, wherein The step of calculating the reciprocal of the sum of the number of e-books in the common e-book data of any two users and a preset similarity coefficient based on the common e-book data of any two users, and performing user loop calculation on the user list based on the reciprocal and a negative correlation weight factor to obtain the similarity between the first e-book and the second e-book further includes: For any two users in the user list, calculating the reciprocal of the sum of the number of e-books in the common e-book data of the two users and a preset similarity coefficient, and multiplying the reciprocal by a time decay factor and / or a user high-heat penalty factor to obtain a similarity component between the two users; Perform user loop calculation on any two users in the user list, and accumulate the obtained similarity components to obtain the similarity between the first e-book and the second e-book.

7. The method according to any one of claims 1 to 4, wherein The method further comprises: The historical e-books in the historical behavior information of the user are obtained, and e-books having similarity with the historical e-books higher than a preset similarity threshold are screened as the e-books to be recommended.

8. An electronic device comprising: A processor, a memory, a communication interface, and a communication bus, wherein the processor, the memory, and the communication interface communicate with each other via the communication bus; The memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform the following operations: Obtaining a first historical user of a first e-book and a second historical user of a second e-book, and determining a user list including common users between the first historical user and the second historical user; screening the same e-books between any two users based on historical behavior information of any two users in the user list to determine the common e-book data; Calculating a negative correlation weight factor of the user for the e-book based on the historical behavior information of the user for the e-book in the user list; Based on the common e-book data of any two users, the reciprocal of the sum of the number of e-books in the common e-book data of any two users and a preset similarity coefficient is calculated. Based on the reciprocal and a negative correlation weight factor, a user loop calculation is performed on the user list to obtain the similarity between the first e-book and the second e-book.

9. The electronic device according to claim 8, wherein the historical behavior information includes historical e-books and information about the user's interactions with the e-books; The negative correlation weight factor includes a time decay factor, and the time decay factor is calculated based on time information in the interaction information between the user and the e-book.

10. The electronic device according to claim 9, wherein the interaction information between the user and the e-book includes the reading time of the e-book; The executable instructions further cause the processor to: Obtaining a first reading time of a user reading a first e-book and a second reading time of a user reading a second e-book; The time decay factor of the e-book is calculated based on the reading time difference between the first reading time and the second reading time and the reading time coefficient; wherein, The reading time difference is inversely proportional to the time decay factor.

11. The electronic device according to claim 9, wherein the interaction information between the user and the e-book includes the time when the e-book was added to the bookshelf; The executable instructions further cause the processor to: Obtaining a first bookshelf time when a user adds a first e-book to a bookshelf, and a second bookshelf time when a user adds a second e-book to a bookshelf; A time decay factor of the e-book is calculated based on a shelf time difference between the first bookshelf time and the second bookshelf time and a bookshelf time coefficient; wherein the shelf time difference is inversely proportional to the time decay factor.

12. The electronic device according to any one of claims 8 to 11, wherein the negative correlation weight factor further comprises a user high fever penalty factor; The executable instructions further cause the processor to: Counting the number of e-books interacted with in the historical behavior information of users in the user list on a user-by-user basis; According to the number, the user high heat penalty factor of the e-book is obtained by weight reduction calculation; wherein, The number is inversely proportional to the user's high fever penalty factor.

13. The electronic device according to any one of claims 8 to 11, wherein the executable instructions further cause the processor to perform the following operations: For any two users in the user list, calculating the reciprocal of the sum of the number of e-books in the common e-book data of the two users and a preset similarity coefficient, and multiplying the reciprocal by a time decay factor and / or a user high-heat penalty factor to obtain a similarity component between the two users; Perform user loop calculation on any two users in the user list, and accumulate the obtained similarity components to obtain the similarity between the first e-book and the second e-book.

14. The electronic device according to any one of claims 8 to 11, wherein the executable instructions further cause the processor to perform the following operations: The historical e-books in the historical behavior information of the user are obtained, and e-books having similarity with the historical e-books higher than a preset similarity threshold are screened as the e-books to be recommended.

15. A computer storage medium, wherein at least one executable instruction is stored in the storage medium, wherein the executable instruction enables a processor to execute operations corresponding to the e-book similarity calculation method according to any one of claims 1 to 7.

Images