Method, device and equipment for pushing electronic book data

Abstract

The invention provides an electronic book data pushing method, device and equipment, belongs to the technical field of computer information processing, and solves the problem that a recommendation result is inaccurate when books are recommended to a user. The method comprises the following steps: acquiring log data and book feature data; processing the log data to obtain user preference data; processing the user preference data and the book feature data to obtain book pushing data; and sending the book pushing data to a client. According to the scheme, dynamic tracking and attenuation processing of user preferences are achieved, and the effect of recommending and displaying books to users in a personalized mode is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of computer information processing, and in particular to a method, device and equipment for pushing electronic book data. Background Art

[0002] At present, the borrowing of traditional paper books has dropped precipitously, while the digital reading rate has increased year by year. However, not all books have electronic versions and corresponding e-book resources. When reading e-books, the existing technology has the problem of poor e-book reading experience, which is mainly manifested in the lack of obvious personalization of user behavior. Conventional book recommendations do not have personalized algorithms, so the books exposed to users are very homogeneous, which in turn affects the diversity of user clicks and reading books. At the same time, due to the small number of users, it is difficult to learn the difference in book representation from such user behavior data. At the same time, the Gini coefficient of books recommended by relying on operational configuration is high, resulting in serious homogeneity of books recommended to users. Summary of the invention

[0003] The present invention provides a method, device and equipment for pushing electronic book data, which solves the problem of inaccurate recommendation results when recommending books to users.

[0004] In order to solve the above technical problems, the technical solution of the present invention is as follows:

[0005] The embodiment of the present invention provides a method for pushing electronic book data, comprising:

[0006] Get log data and book feature data;

[0007] Processing the log data to obtain user preference data;

[0008] Processing the user preference data and the book feature data to obtain book push data;

[0009] The book push data is sent to the client.

[0010] Optionally, obtain log data and book feature data, including:

[0011] Acquire first log data and second log data from a user database stored in a server, wherein the first log data includes at least one of click data, search data, and reading time data; and the second log data includes at least one of click data, search data, and reading behavior data;

[0012] The book characteristic data is obtained from a book database stored in the server.

[0013] Optionally, the log data is processed to obtain user preference data, including:

[0014] At least one of the click data, search data and reading time data in the first log data is weighted to generate user preference data.

[0015] Optionally, the user preference data and the book feature data are processed to obtain book push data, including:

[0016] Performing feature extraction processing on the user preference data to obtain user preference feature data;

[0017] Performing matching processing on the user preference feature data and the book feature data to obtain first book matching data, wherein the first book matching data includes at least one of an identification code, a title, a publisher, and an author of the book;

[0018] According to the first book matching data, a book database stored in a server is searched and processed to obtain book push data, wherein the book push data includes at least one of a book identification code, a book title, a publisher, an author, a cover image, and a content summary.

[0019] Optionally, the log data is processed to obtain user preference data, including:

[0020] The first log data is subjected to weight reduction processing, and the second log data is subjected to weight addition processing to generate user preference data.

[0021] Optionally, the user preference data and the book feature data are processed to obtain book push data, including:

[0022] Performing weighted processing on at least one of the click data, search data, and reading behavior data in the second log data to generate book preference data;

[0023] The book preference data, the user preference data and the book feature data are processed to obtain book push data.

[0024] Optionally, the book preference data, the user preference data and the book feature data are processed to obtain book push data, including:

[0025] Performing feature extraction processing on the book preference data and the user preference data to obtain book preference feature data;

[0026] Matching the book preference feature data with the book feature data to obtain second book matching data, wherein the second book matching data includes at least one of an identification code, a title, a publisher, and an author of the book;

[0027] According to the second book matching data, a book database stored in the server is searched and processed to obtain book push data.

[0028] The embodiment of the present invention also provides a device for pushing electronic book data, including:

[0029] The acquisition module is used to obtain log data and book feature data;

[0030] The processing module is used to process the log data to obtain user preference data; process the user preference data and the book feature data to obtain book push data; and send the book push data to the client.

[0031] An embodiment of the present invention further provides a computing device, comprising: a processor and a memory storing a computer program, wherein the computer program executes the above method when executed by the processor.

[0032] An embodiment of the present invention further provides a computer-readable storage medium storing instructions, which, when executed on a computer, enable the computer to execute the above method.

[0033] The technical solution of the present invention includes at least the following effects:

[0034] The above scheme of the present invention obtains log data and book feature data; processes the log data to obtain user preference data; processes the user preference data and the book feature data to obtain book push data; and sends the book push data to the client, thereby realizing dynamic tracking and attenuation processing of user preferences, and achieving the effect of personalized recommendation and display of books to users. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 is a flow chart of a method for pushing electronic book data provided by an embodiment of the present invention;

[0036] Figure 2 It is a structure diagram of a label system of a method for pushing electronic book data provided by an embodiment of the present invention;

[0037] Figure 3 It is a flow chart of generating book push data of the method for pushing electronic book data provided by an embodiment of the present invention;

[0038] Figure 4 is a structural diagram of a device for pushing electronic book data provided by an embodiment of the present invention;

[0039] Figure 5 It is a schematic diagram of the structure of a computing device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0040] The exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present invention and to enable the scope of the present invention to be fully communicated to those skilled in the art.

[0041] like Figure 1 As shown, an embodiment of the present invention provides a method for pushing electronic book data, comprising:

[0042] Step 11, obtaining log data and book feature data;

[0043] Step 12, processing the log data to obtain user preference data;

[0044] Step 13, processing the user preference data and the book feature data to obtain book push data;

[0045] Step 14: Send the book push data to the client.

[0046] In this embodiment, the log data is derived from various behavior records generated when the user accesses the book database stored in the server; the log data is obtained by collecting the above-mentioned behavior data of the user within a specified time period; the book feature data includes basic information of the book to reflect the theme and core content of the book, and assigns one or more categories (such as science fiction, literature, history, etc.) and tags (such as recommended, popular, new books, etc.) to each book so that the user can filter according to interests and needs; the log data is processed, including analyzing the user's behavior to obtain user preference data, specifically, analyzing the type of books browsed by the user, the dwelling time, the page turning frequency, etc., to judge the user's initial interest, analyzing the user's browsing history, understanding the type and frequency of the book actually browsed by the user, and the influencing factors of the purchase decision, analyzing the user's evaluation content and rating of the book, etc., to understand the user's satisfaction and preference for the book; based on the user's browsing, purchasing and evaluation behaviors, generating the user's preference data for specific books, including Including favorite book lists, frequently read book types, etc., analyze the user's preference for different categories of books, generate user preference data for book categories, analyze the user's preference for books with specific tags, and generate user preference data for book tags; match user preference data with book feature data to find books that meet user preferences. At the same time, according to the user's preference for different types of books, assign different weights to the matching books to ensure the accuracy and personalization of the push results; sort the matching books according to the weights, give priority to pushing the books that the user is most interested in, and encapsulate the sorted book data into push data, including the basic information of the book, cover pictures, introductions, etc., so as to be displayed on the client; finally, convert the push data into a format that the client can recognize, such as JSON, XML, etc., and send the push data to the client through the data transmission protocol. According to the format and content of the push data, the client displays the recommended book list, book details and other information on the interface for users to view and select.

[0047] The above-mentioned embodiments of the present invention realize dynamic tracking and attenuation processing of user preferences, thereby achieving the effect of personalized recommendation and display of books to users.

[0048] In an optional embodiment of the present invention, step 11 may include:

[0049] Step 111, obtaining first log data and second log data from a user database stored in a server, wherein the first log data includes at least one of click data, search data, and reading time data; and the second log data includes at least one of click data, search data, and reading behavior data;

[0050] Step 112, obtaining book feature data from a book database stored in the server.

[0051] In this embodiment, the first log data is obtained by connecting to and accessing a user database stored on a server to extract specific information from the user database; the first log data includes but is not limited to the user's click data, search data and reading time data, wherein the click data records the user's click behavior on the platform, such as the clicked book title, click timestamp, etc.; the search data records the user's search behavior, including search keywords, search time, etc.; the reading time data records the time the user spends reading each book, which helps to understand the user's reading habits and interest depth; the second log data is obtained by connecting to and accessing a temporary user database stored on a server to extract specific information from the user database. Obtained; The difference between the second log data and the first log data is that the first log data records the historical data of the user before the current visit, while the second log data records the latest data generated during the current visit, and the latest data is temporarily stored in a temporary user database on the server; the second log data includes but is not limited to the user's click data, search data and reading behavior data, among which the reading behavior data refers to various recordable data information generated by the user during the reading activity. These data can reflect the user's reading habits, interest preferences and reading effects, etc., specifically including: reading time and reading quantity, reading word count, number of book clicks and downloads, number of repeated listenings to books, type of books read, etc.

[0052] The book feature data is obtained by connecting to and accessing the book database stored on the server, and extracting specific information from the book database; the book feature data includes but is not limited to the book's identification code, author, category, publisher and label; among them, the book identification code is the unique identifier of each book, which is used to uniquely identify a book in the system; the author data is the author information of the book, including a single author or multiple authors (co-authored); the category data is the category or subcategory to which the book belongs, such as science fiction, literature, history, etc.; the publisher data is the publisher or publisher information of the book; the tag data is the tags or keywords related to the book, which are usually used to describe the content, theme or style of the book.

[0053] In an optional embodiment of the present invention, step 12 may include:

[0054] Step 121, weighting processing is performed on at least one of the click data, search data and reading time data in the first log data to generate user preference data.

[0055] In this embodiment, the first log data includes but is not limited to the user's click data, search data and reading time data; the weight of these data in the user preference data can be determined based on experience, business logic or data analysis results; for example, it can be considered that reading time data can better reflect the user's real preference than click data, so a higher weight is assigned to reading time data; similarly, for search data, its weight can be determined based on factors such as the relevance of search keywords and search frequency; the calculation expression of user preference data is:

[0056]

[0057] In the formula, P is the user preference data, p1~p3 are the user preference data corresponding to the click data, search data, and reading time data respectively, k1~k3 are the behavior type weights corresponding to the click data, search data, and reading time data respectively, t1~t3 are the time decay data corresponding to the click data, search data, and reading time data respectively, and c1~c3 are the number of click behaviors, search behaviors, and reading behaviors respectively.

[0058] After the weights are determined, the first log data is processed using weighted summation or other weighted calculation methods; for each user, a comprehensive score or preference index can be calculated based on their click data, search data, reading time data and corresponding weights, that is, user preference data is generated; this score or index will reflect the user's preference for different types of books or content.

[0059] In an optional embodiment of the present invention, step 13 may include:

[0060] Step 131, performing feature extraction processing on the user preference data to obtain user preference feature data;

[0061] Step 132, matching the user preference feature data with the book feature data to obtain first book matching data, wherein the first book matching data includes at least one of an identification code, a title, a publisher, and an author of the book;

[0062] Step 133, based on the first book matching data, search and process the book database stored in the server to obtain book push data, wherein the book push data includes at least one of the book identification code, book title, publisher, author, cover image, and content introduction.

[0063] In this embodiment, the user preference data is first analyzed and processed to extract features that can represent the user's reading preferences and interests; these features can be extracted from multiple dimensions such as the user's reading history, click behavior, search history, reading time, reading type (such as novels, popular science, history, etc.), and a set of refined and effective user preference feature data is generated, which will be used for subsequent book matching and push;

[0064] Compare and match the user preference feature data with the book feature data stored in the system to obtain first book matching data. The matching process includes: calculating the similarity or distance between the user preference feature data and the book feature data, sorting by the similarity or distance, and selecting the books that best match the user preference; these matching results will be used as the first book matching data, including the book identification code, book title, publisher, author and other information, which will be used for subsequent book retrieval;

[0065] The similarity formula is:

[0066]

[0067] In the formula, the cosine value cos(θ) is between -1 and 1. The larger the value, the more similar the two vectors are. A is the user preference feature data, B is the book feature data, and A and B are both word vectors.

[0068] Based on the key information in the first book matching data, such as the book identification code, title, etc., the book database stored in the server is searched; the book database contains a large amount of book detailed information, including the book identification code, title, publisher, author, cover image, content introduction, etc.; through the retrieval process, the records matching the first book matching data are found in the book database, and the detailed information in these records are extracted as book push data; these push data will be used to generate personalized book recommendation lists or pages, which will be displayed to users to attract their attention and guide them to read.

[0069] In an optional embodiment of the present invention, step 12 may further include:

[0070] Step 122: performing weight reduction processing on the first log data and weighting processing on the second log data to generate user preference data.

[0071] In this embodiment, the main function is to merge the daily new data with the existing full user features, and at the same time attenuate the user's interest according to time to maintain dynamic updates; specifically, first collect or obtain the second log data according to a predetermined time interval, that is, the first preset time interval, and the second log data includes at least one of click data, search data and reading behavior data; these data reflect the user's behavior pattern when browsing book data; relative to the second log data, the first log data reflects the user's historical behavior pattern when browsing book data; because the first log data is outdated, inaccurate or no longer reflects the user's current preference, it is necessary to reduce the influence of the first log data in the subsequent analysis or recommendation algorithm, and the way to reduce the weight can be to directly reduce its value or adjust its weight coefficient; contrary to the first log data, since the second log data reflects the user's current interest and preference, it is necessary to increase the influence of the second log data in the subsequent analysis or recommendation algorithm, and the way of weighted processing can be to increase its value, increase its weight coefficient or place it at the core of the analysis; the formula for the above-mentioned process of generating user preference data can be expressed as:

[0072] P=aP1+bP2

[0073] In the formula, a is the weight of the first log data, b is the weight of the second log data, P1 is the user preference data corresponding to the first log data, and P2 is the user preference data corresponding to the second log data.

[0074] In an optional embodiment of the present invention, step 13 may further include:

[0075] Step 134, performing weighted processing on at least one of the click data, search data, and reading behavior data in the second log data to generate book preference data;

[0076] Step 135: Process the book preference data, the user preference data and the book feature data to obtain book push data.

[0077] In this embodiment, the collected second log data is first processed to extract the user's book preference data; specifically, different weights are assigned to different log data according to the type, importance and frequency of the data; for example, a higher weight can be given to books that are frequently clicked and read for a long time; while a lower weight is given to books that are only occasionally clicked or browsed; through weighted processing, the user's preference for different types of books, such as literature, science fiction, history, suspense, etc., can be calculated; based on the book preference data and the user preference data, matching books are searched in the book feature database; including analysis and comparison of the book feature data to find the books that are closest to the user's preferences; through matching processing, the system can generate a push data containing recommended books; this data will include key information such as the book's title, cover, introduction, rating, author, etc., so that users can quickly understand and make choices; the book push data is presented to the user through push notifications, message reminders or direct display on the client interface.

[0078] In an optional embodiment of the present invention, step 135 may include:

[0079] Step 1351, performing feature extraction processing on the book preference data and the user preference data to obtain book preference feature data;

[0080] Step 1352, matching the book preference feature data with the book feature data to obtain second book matching data, wherein the second book matching data includes at least one of an identification code, a title, a publisher, and an author of the book;

[0081] Step 1353: perform search processing on the book database stored in the server according to the second book matching data to obtain book push data.

[0082] In this embodiment, the collected user book preference data is first subjected to feature extraction; the purpose of feature extraction is to extract key information that can represent the user's book preference, i.e., features, from these raw data. These features can be the user's preferred book type (such as science fiction, literature, history, etc.), author preference, publisher preference, book theme, style preference (such as light, serious, suspense, etc.), etc.; after feature extraction processing, the raw book preference data is converted into structured book preference feature data;

[0083] Matching the book preference feature data with the book feature data in the book database; wherein the book feature data is the attribute description of the book itself, including the book identification code, title, publisher, author, publication date, book type, content summary, keywords, book reviews, etc.;

[0084] The matching process calculates the similarity or correlation between the user preference features and the book features to find the books that best match the user preferences, namely the second book matching data. This data includes key information such as the book's identification code, title, publisher, author, etc. This information will be used for subsequent book retrieval and push.

[0085] Searching the book database stored in the server according to the second book matching data; the search process includes searching for book records that match key information such as book identification code, book title, publisher, author, etc. in the second book matching data;

[0086] The result of the retrieval processing is book push data, which contains detailed information about the books that the user is interested in, such as title, author, publisher, cover image, introduction, price, etc.; this data will be used to generate a book recommendation list or push notification, which will be presented to the user to guide him or her to read further.

[0087] A specific embodiment of the method for pushing electronic book data provided by the embodiment of the present invention is as follows:

[0088] Step 1, obtaining log data and book feature data;

[0089] Specifically, the user behavior logs are used to read information such as clicks, searches, and reading time to obtain user feature data and user behavior data. The user feature data includes information such as the user's age, gender, birthday, geographic location, and industry. The user behavior data includes information such as the user's book clicks, searches, reading records, and reading time. The book feature data includes information such as the book's ID, author, category, publisher, and tag, which is used for matching during preference calculation and regular maintenance of new book information.

[0090] Step 2, processing the log data to obtain user preference data;

[0091] Specifically, based on the user's click, search and reading behaviors, combined with the e-book metadata, the user's book, category, and tag preferences are calculated to profile the user. The so-called user profile is to label the user with various labels, such as age, gender, interests, browsing history, etc. These labels can be divided into fact labels, model labels, prediction labels, etc. On the other hand, they can be divided into primary labels, secondary labels, tertiary labels, etc. according to the scope they represent; through the above classification, a label system is obtained, such as Figure 2 As shown;

[0092] At the same time, the user's recent behavior records are recorded, and the most recent 100 click, search and reading records are retained; the daily new data is integrated with the existing full user features, and the user's interest is decayed according to time to maintain dynamic updates; the updated user features are saved as JSON format files for use by the recommendation system;

[0093] The preference score of each book is calculated based on the user's click, search and reading behaviors; book preferences are aggregated by category and tag to generate first-level category preferences, second-level category preferences and tag preferences; the newly calculated preferences are merged with the existing user profile, and a decay factor is applied to keep the user characteristics dynamically updated; the proportion of historically interacted books in the user's interest weight is gradually reduced to ensure that the system can better capture changes in user interests; avoid recommending content that is too narrow, and by decaying old behaviors, ensure that the distribution of recommended content changes rapidly with user behavior, rather than being locked by previously interacted content.

[0094] Step 3, processing the user preference data and the book feature data to obtain book push data;

[0095] Specifically, feature extraction is performed on the user preference data to obtain user preference feature data, that is, by processing the user's e-book reading, clicking, and searching behavior logs, the user's personalized features, including book preferences, category preferences, tag preferences, etc., are constructed; these features are updated and maintained by daily incremental calculation, and the newly added log data is merged with the existing full user feature archive every day, so as to realize dynamic tracking and attenuation processing of user preferences; the preference feature data and the book feature data are matched to obtain first book matching data, wherein the first book matching data includes at least one of the book's identification code, book title, publisher, and author; according to the first book matching data, a book database stored in a server is searched to obtain book push data, wherein the book push data includes at least one of the book's identification code, book title, publisher, author, cover image, and content introduction;

[0096] Processing the book feature data also includes extracting book feature data, using the user's behavior log (clicks, reading time, etc.) within a preset time range to make statistics in various dimensions, mainly including: book title, author, category, tag and other features; calculating the click, reading, and conversion rates of each stage for each book, and reducing data noise through confidence processing to generate a more detailed portrait, that is, generating a behavioral portrait of each institution and each industry for the revised book, and saving the updated user features as a JSON format file for use by the recommendation system;

[0097] like Figure 3As shown, the process of generating book push data includes:

[0098] (1) Recall: Select a portion of candidate items from a huge book database; use multiple recall channels, each channel selects a certain number of books according to a specific weight strategy; recall tens of thousands or even thousands of candidate books from hundreds of thousands of books;

[0099] (2) Rough sorting: further filter out the books that users may be interested in from the recalled books; use a smaller-scale machine learning model to score the recalled books; and filter out the hundreds of books with the highest scores based on the model scores;

[0100] (3) Fine ranking: Accurately assess the user's interest in the remaining books; carefully score the books after rough ranking; and obtain a fine ranking score that reflects the user's interest;

[0101] (4) Rearrangement: Optimize the book list that is finally displayed to the user, perform random sampling based on the refined ranking score and diversity score, and adjust the order of the books; produce the ten books that are finally displayed to the user; the model that generates the book push data is a Word2Vec model trained based on user behavior data (Session); the Word2Vec model is used to generate the vector representation (embedding) of each e-book through the session data formed by the user's reading behavior within a day, and the books are recommended by calculating the similarity; the whole process includes reading the user behavior session, training the Word2Vec model, generating the book embedding vector, and calculating the Top K similar books for each book. The vector learned by the model can express the semantic similarity between books, thereby improving the recommendation accuracy and personalization of the recommendation system. The current implementation uses the Skip-gram training method of the Word2Vec model, which predicts the relationship between words (in this case, books) from the context; the specific implementation is based on the gensim library, and the model effect is controlled by parameters such as window size and embedding size; through training, the model will generate a vector representation of each book, and can calculate the Top K similar books for each book based on these vectors. This similarity can be applied to the recommendation system to recommend related or similar e-books to users;

[0102] DeepWalk is a graph embedding algorithm based on random walks. It learns low-dimensional representations (embeddings) of nodes in a graph by performing random walks on the graph, collecting node sequences, and using these sequences as training data. These representations can be used for many graph analysis tasks, such as node classification, clustering, link prediction, etc. Based on the DeepWalk algorithm, this e-book data push method supports the use of alias sampling to accelerate the random walk process, adds the option of sampling using Softmax, and introduces a label walk mechanism for book content to enhance the diversity and content relevance of graph embedding representations.

[0103] Step 4: Send the book push data to the client.

[0104] When the electronic book data push method of the present invention is executed for the first time, due to the lack of sufficient first log data, recommendations can be made through preset sample data or a random recommendation method (this process can be called a cold start); when sufficient first log data has been accumulated, the electronic book data push method of steps 1 to 4 can be used to recommend data.

[0105] The above-mentioned embodiment of the present invention forms log data by counting the reading behavior of users in a day, generates a vector representation of each electronic book based on the log data, and recommends books by calculating similarity; the whole process includes reading user behavior log data, training a book recommendation model, generating an embedding vector of the book, and calculating similar books for each book; the vector learned by the model can express the semantic similarity between books, thereby improving the recommendation accuracy and personalization of the recommendation system.

[0106] like Figure 4 As shown, the embodiment of the present invention further provides a device 40 for pushing electronic book data, comprising:

[0107] An acquisition module 41 is used to acquire log data and book feature data;

[0108] The processing module 42 is used to process the log data to obtain user preference data; process the user preference data and the book feature data to obtain book push data; and send the book push data to the client.

[0109] Optionally, obtain log data and book feature data, including:

[0110] Acquire first log data and second log data from a user database stored in a server, wherein the first log data includes at least one of click data, search data, and reading time data; and the second log data includes at least one of click data, search data, and reading behavior data;

[0111] The book characteristic data is obtained from a book database stored in the server.

[0112] Optionally, the log data is processed to obtain user preference data, including:

[0113] At least one of the click data, search data and reading time data in the first log data is weighted to generate user preference data.

[0114] Optionally, the user preference data and the book feature data are processed to obtain book push data, including:

[0115] Performing feature extraction processing on the user preference data to obtain user preference feature data;

[0116] Performing matching processing on the user preference feature data and the book feature data to obtain first book matching data, wherein the first book matching data includes at least one of an identification code, a title, a publisher, and an author of the book;

[0117] According to the first book matching data, a book database stored in a server is searched and processed to obtain book push data, wherein the book push data includes at least one of a book identification code, a book title, a publisher, an author, a cover image, and a content summary.

[0118] Optionally, the log data is processed to obtain user preference data, including:

[0119] The first log data is subjected to weight reduction processing, and the second log data is subjected to weight addition processing to generate user preference data.

[0120] Optionally, the user preference data and the book feature data are processed to obtain book push data, including:

[0121] Performing weighted processing on at least one of the click data, search data, and reading behavior data in the second log data to generate book preference data;

[0122] The book preference data, the user preference data and the book feature data are processed to obtain book push data.

[0123] Optionally, the book preference data, the user preference data and the book feature data are processed to obtain book push data, including:

[0124] Performing feature extraction processing on the book preference data and the user preference data to obtain book preference feature data;

[0125] Matching the book preference feature data with the book feature data to obtain second book matching data, wherein the second book matching data includes at least one of an identification code, a title, a publisher, and an author of the book;

[0126] According to the second book matching data, a book database stored in the server is searched and processed to obtain book push data.

[0127] It should be noted that the device is a device corresponding to the above method, and all implementation methods in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0128] like Figure 5 As shown, the embodiment of the present invention further provides a computing device 50, including a processor 51, a memory 52, and a program or instruction stored in the memory 52 and executable on the processor 51. When the program or instruction is executed by the processor 51, each process of the above-mentioned electronic book data push method embodiment is implemented, and the same technical effect can be achieved. To avoid repetition, it is not repeated here. It should be noted that the computing device in the embodiment of the present invention includes the above-mentioned mobile electronic device and non-mobile electronic device.

[0129] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0130] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0131] In the embodiments provided by the present invention, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0132] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0133] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0134] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods of various embodiments of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, ROM, RAM, magnetic disks, or optical disks.

[0135] In addition, it should be noted that in the apparatus and method of the present invention, it is obvious that each component or each step can be decomposed and / or recombined. These decompositions and / or recombinations should be regarded as equivalent schemes of the present invention. Moreover, the steps of performing the above-mentioned series of processing can naturally be performed in chronological order according to the order of description, but it is not necessary to perform them in chronological order, and some steps can be performed in parallel or independently of each other. For those of ordinary skill in the art, it is understood that all or any steps or components of the method and apparatus of the present invention can be implemented in any computing device (including processors, storage media, etc.) or a network of computing devices in hardware, firmware, software or a combination thereof, which can be achieved by those of ordinary skill in the art using their basic programming skills after reading the description of the present invention.

[0136] Therefore, the purpose of the present invention can also be achieved by running a program or a group of programs on any computing device. The computing device can be a well-known general-purpose device. Therefore, the purpose of the present invention can also be achieved by simply providing a program product containing a program code for implementing a method or device. That is to say, such a program product also constitutes the present invention, and a storage medium storing such a program product also constitutes the present invention. Obviously, the storage medium can be any well-known storage medium or any storage medium developed in the future. It should also be pointed out that in the device and method of the present invention, it is obvious that each component or each step can be decomposed and / or recombined. These decompositions and / or recombinations should be regarded as equivalent schemes of the present invention. In addition, the steps of performing the above-mentioned series of processing can naturally be performed in chronological order according to the order of description, but it is not necessary to perform them in chronological order. Some steps can be performed in parallel or independently of each other.

[0137] The above are preferred embodiments of the present invention. It should be pointed out that, for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A method for pushing electronic book data, characterized in that: include: Get log data and book feature data; Processing the log data to obtain user preference data; Processing the user preference data and the book feature data to obtain book push data; The book push data is sent to the client.

2. The method for pushing electronic book data according to claim 1, characterized in that: The obtaining of log data and book feature data includes: Acquire first log data and second log data from a user database stored in a server, wherein the first log data includes at least one of click data, search data, and reading time data; and the second log data includes at least one of click data, search data, and reading behavior data; The book characteristic data is obtained from a book database stored in the server.

3. The method for pushing electronic book data according to claim 2, characterized in that: The log data is processed to obtain user preference data, including: At least one of the click data, search data and reading time data in the first log data is weighted to generate user preference data.

4. The method for pushing electronic book data according to claim 3, characterized in that: Processing the user preference data and the book feature data to obtain book push data includes: Performing feature extraction processing on the user preference data to obtain user preference feature data; Matching the user preference feature data with the book feature data to obtain first book matching data, wherein the first book matching data includes at least one of an identification code, a title, a publisher, and an author of the book; According to the first book matching data, a book database stored in a server is searched and processed to obtain book push data, wherein the book push data includes at least one of a book identification code, a book title, a publisher, an author, a cover image, and a content summary.

5. The method for pushing electronic book data according to claim 2, characterized in that: The log data is processed to obtain user preference data, including: The first log data is subjected to weight reduction processing, and the second log data is subjected to weight addition processing to generate user preference data.

6. The method for pushing electronic book data according to claim 5, characterized in that: Processing the user preference data and the book feature data to obtain book push data includes: Performing weighted processing on at least one of the click data, search data, and reading behavior data in the second log data to generate book preference data; The book preference data, the user preference data and the book feature data are processed to obtain book push data.

7. The method for pushing electronic book data according to claim 6, characterized in that: Processing the book preference data, the user preference data and the book feature data to obtain book push data includes: Performing feature extraction processing on the book preference data and the user preference data to obtain book preference feature data; Matching the book preference feature data with the book feature data to obtain second book matching data, wherein the second book matching data includes at least one of an identification code, a title, a publisher, and an author of the book; According to the second book matching data, a book database stored in the server is searched and processed to obtain book push data.

8. A device for pushing electronic book data, characterized in that: include: The acquisition module is used to obtain log data and book feature data; The processing module is used to process the log data to obtain user preference data; process the user preference data and the book feature data to obtain book push data; and send the book push data to the client.

9. A computing device, characterized in that include: A processor and a memory storing a computer program, wherein when the computer program is executed by the processor, the method according to any one of claims 1 to 7 is performed.

10. A computer-readable storage medium, characterized in that: Instructions are stored, and when the instructions are executed on a computer, the computer is caused to execute the method according to any one of claims 1 to 7.

Images