A method and system for sharing and managing children's books

Abstract

The present invention provides a children's book sharing management method and system, which relates to the technical field of book resource management. The method includes: collecting historical reading record data and education stage data of multiple platform users, extracting group borrowing characteristics of each reference education stage from multiple groups of historical reading record data, and performing stage distribution analysis to generate a book borrowing evolution strategy for each reference education stage; obtaining target reading record data of a target user, performing group evolution offset detection on the target user to obtain the target user's group borrowing offset data; performing heat influence stripping on the target user, and extracting the target user's heat filtering preference data; fusing the target user's group borrowing offset data and heat filtering preference data to generate a book borrowing matching strategy for the target user in the target reading record data, which is used to optimize the children's book sharing platform's book sharing strategy for the target user. The present invention realizes the accurate recommendation of book resources for children.

Description

Technical Field

[0001] The present invention relates to the technical field of book resource management, and in particular to a children's book sharing management method and system. Background Art

[0002] A book-sharing platform is a digital system based on borrowing and sharing, connecting users with book resources and effectively improving the sharing efficiency of both print and digital books. For example, e-book lending platforms for public libraries and educational institutions support personalized learning for children through resource management and technology, helping parents and educators configure book resources based on learning needs.

[0003] Some platforms use information such as borrowing records and ratings to make book recommendations. However, because children's cognitive abilities, learning progress, and interests vary dynamically with age and grade level, failure to fully account for these changes can lead to recommendations that don't align with children's true needs. Furthermore, short-term popularity effects, such as those associated with short-term teaching activities, often impact borrowing frequency within a short period of time. However, these short-term trends don't reflect children's long-term interests, which can easily lead to deviations in interest and affect the accuracy of recommendations. Furthermore, even children of the same grade vary in their progress in acquiring knowledge and cognitive abilities due to various factors. For example, some children may be ahead in their studies while others may lag behind. Systems that rely solely on grade or age group recommendations cannot effectively address these individual differences, resulting in recommended books that may not meet each child's actual learning needs. Strategies for sharing book resources across children's groups need to be optimized to better tailor recommendations based on children's actual borrowing behavior. Summary of the Invention

[0004] In order to solve the above technical problems, the present invention proposes a children's book sharing management method and system for implementing personalized book resource recommendations to children.

[0005] A first aspect of the present invention provides a children's book sharing and management method, comprising:

[0006] Collect historical reading record data and education stage data of multiple platform users on the children's book sharing platform, and determine multiple reference education stages based on the education stage data;

[0007] Extract the borrowing characteristics of each reference education stage from multiple sets of historical reading record data, perform stage distribution analysis on multiple borrowing characteristics of each group, and generate the book borrowing evolution strategy for each reference education stage;

[0008] Obtain the target reading record data of the target user, determine the target education stage to which the target user currently belongs, perform group evolution deviation detection on the target user based on multiple book borrowing evolution strategies, and obtain the group borrowing deviation data of the target user;

[0009] Obtain borrowing popularity reference data of the target reading record data on the children's book sharing platform, and based on the borrowing popularity reference data and the target reading record data, separate the popularity influence of the target user and extract the target user's popularity filtering preference data;

[0010] By integrating the target user's group borrowing offset data and heat filtering preference data, a book borrowing matching strategy for the target user in the target reading record data is generated. The book borrowing matching strategy is used to optimize the children's book sharing platform's book sharing strategy for the target user.

[0011] Preferably, group borrowing characteristics of each reference education stage are extracted from multiple groups of historical reading record data, and stage distribution analysis is performed on multiple group borrowing characteristics to generate a book borrowing evolution strategy for each reference education stage, including:

[0012] Based on multiple groups of historical reading record data, multiple groups of group borrowing popularity data on different books in each reference education stage are determined, and the local borrowing features of each book in each group of group borrowing popularity data are extracted. The multiple local borrowing features of each book are fused based on the time series distribution of heat to generate group borrowing features for multiple books in the reference education stage. Based on multiple groups of historical reading record data, the time period distribution data on multiple books in the group borrowing features of each reference education stage are determined to generate the book borrowing evolution strategy for each reference education stage, including the group borrowing preference data in each local stage of the reference education stage.

[0013] Preferably, performing group evolution deviation detection on target users according to multiple book borrowing evolution strategies to obtain group borrowing deviation data of target users includes:

[0014] The individual borrowing preference data of the target user in each local stage is extracted from the target reading record data. According to the target education stage to which the target user currently belongs, multiple target borrowing evolution strategies associated with the target user in multiple book borrowing evolution strategies are determined. The multiple groups of individual borrowing preference data of the target user are matched with the multiple target borrowing evolution strategies, including performing group evolution offset detection on the target user based on the group borrowing preference data and individual borrowing preference data in each local stage, obtaining the group borrowing offset features of the target user in each local stage, and fusing the group borrowing offset features to obtain the group borrowing offset data of the target user.

[0015] Preferably, the target user's popularity influence is stripped off to extract the target user's popularity filtering preference data, including:

[0016] Determine multiple local book preference features of the target user in the target reading record data, extract multiple global book preference parameters of each local book preference feature from the borrowing heat reference data, calculate the heat deviation factor of each local book preference feature based on the multiple global book preference parameters, and perform heat influence stripping on the local book preference features of the target user based on the heat deviation factor to obtain the heat filtering preference data of the target user regarding the multiple local book preference features.

[0017] Preferably, the target user's group borrowing offset data and heat filtering preference data are integrated to generate a book borrowing matching strategy for the target user in the target reading record data, including:

[0018] Determine the target local stage to which the user belongs, combine the target education stage to which the target user belongs, and extract the local borrowing evolution strategy associated with the target local stage from the book borrowing evolution strategy of the target education stage;

[0019] Determine multiple strategy fusion weights for local borrowing evolution strategies based on group borrowing offset data, perform heat analysis on the local borrowing evolution strategies using the strategy fusion weights, and calculate first heat parameters for multiple books;

[0020] Multiple first popularity parameters are modified and processed through popularity filtering preference data, and the second popularity parameter corresponding to each first popularity parameter is calculated. Based on the second popularity parameters of multiple books, a book borrowing matching strategy for the target user in the target reading record data is generated.

[0021] Preferably, performing group evolution deviation detection on the target user to obtain the group borrowing deviation feature of the target user in each local stage includes:

[0022] The borrowing preference feature vectors corresponding to the group borrowing preference data and the individual borrowing preference data are extracted respectively, and the borrowing preference feature vectors corresponding to the group borrowing preference data and the individual borrowing preference data in each local stage are matched and analyzed, and the similarity parameters in each local stage are calculated. Based on multiple similarity parameters, the group borrowing offset characteristics of the target user in each local stage are generated.

[0023] A second aspect of the present invention provides a children's book sharing management system, which is used to implement the above-mentioned children's book sharing management method, including:

[0024] A reading data preprocessing module is used to collect historical reading record data and education stage data of multiple platform users on the children's book sharing platform, and determine multiple reference education stages based on the education stage data;

[0025] The book borrowing evolution analysis module is used to extract the group borrowing characteristics of each reference education stage from multiple sets of historical reading record data, perform stage distribution analysis on multiple group borrowing characteristics, and generate the book borrowing evolution strategy for each reference education stage;

[0026] The group borrowing deviation analysis module is used to obtain the target reading record data of the target user, determine the target education stage to which the target user currently belongs, and perform group evolution deviation detection on the target user based on multiple book borrowing evolution strategies to obtain the group borrowing deviation data of the target user;

[0027] The popularity impact stripping module is used to obtain the borrowing popularity reference data of the target reading record data on the children's book sharing platform, and based on the borrowing popularity reference data and the target reading record data, it strips the popularity impact of the target user and extracts the target user's popularity filtering preference data;

[0028] The book sharing strategy generation module is used to integrate the target user's group borrowing offset data and heat filtering preference data to generate the target user's book borrowing matching strategy in the target reading record data, and optimize the children's book sharing platform's book sharing strategy for the target user through the book borrowing matching strategy.

[0029] The present invention has the following beneficial effects:

[0030] The present invention extracts group characteristics and analyzes the stage distribution of the borrowing behavior of children's groups, constructs a book borrowing evolution strategy for different educational stages, performs group evolution offset detection on target users based on the target reading record data of target users, analyzes the matching properties of target users and group reading behavior, further considers the interference effect of short-term hot spots on individual interests, and quantitatively analyzes the impact of short-term hot spots on individuals from dimensions such as behavior attenuation and local interest matching. Finally, the offset characteristics of group preferences are integrated with individual reading preferences after filtering out the influence of short-term hot spots, and a book borrowing matching strategy for target users based on target reading record data is generated, which effectively distinguishes between stable needs driven by cognitive development and temporary preferences induced by the external environment, improves the accuracy of children's book recommendations in terms of knowledge connection and interest matching, and realizes accurate and personalized book recommendations for different individuals. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 The figure is a flowchart of an exemplary children's book sharing and management method according to an embodiment of the present invention.

[0032] Figure 2 This is a structural diagram of an exemplary children's book sharing management system in an embodiment of the present invention. DETAILED DESCRIPTION

[0033] In order to enable those skilled in the art to better understand the technical solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0034] Figure 1 This is a flowchart of an exemplary children's book sharing management method in an embodiment of the present invention. Figure 1 , a children's book sharing management method, comprising:

[0035] Step S10: Collect historical reading record data and education stage data of multiple platform users on the children's book sharing platform, and determine multiple reference education stages based on the education stage data.

[0036] In this embodiment, historical reading record data and educational stage data of users on different platforms are collected from a children's book sharing platform. The historical reading records include behavioral data such as the types of books read or borrowed by platform users, borrowing frequency, and borrowing duration. It should be noted that different book sharing platforms may simultaneously involve relevant records of online and offline books. Those skilled in the art will recognize that this application is not limited to analyzing relevant record data of one category of online or offline book resources. Educational stage data covers information such as the user's academic year, grade, or age. By analyzing this data, multiple reference educational stages are determined. For example, the reference educational stage can be divided into first grade, second grade, etc. of elementary school, or further divided into the upper and lower semesters of different grades, as well as different periods within each semester. Under school education, children's cognitive abilities, learning needs, and reading interests within each reference educational stage will have certain commonalities. At the same time, they are influenced by factors such as family background and individual differences. There will also be individual needs that are somewhat different from the child's grade or age.

[0037] Step S20: extracting group borrowing characteristics of each reference education stage from multiple groups of historical reading record data, performing stage distribution analysis on multiple group borrowing characteristics, and generating a book borrowing evolution strategy for each reference education stage.

[0038] In this embodiment, group characteristic information mining is performed on the historical reading records of users on multiple platforms. This includes extracting group borrowing characteristics for each reference education stage. This is used to reflect the common interests of children in borrowing books at different reference education stages. For example, younger elementary school students may tend to read picture books and children's literature, and as they get older, they become more inclined to read more specialized books such as science and history. Furthermore, in response to these changes in group interests, the detailed preference evolution within the reference education stage is further mined to generate corresponding book borrowing evolution strategies. This identifies the group behavior characteristics of different education stages and reflects the changes in children's interests and the evolution of their reading habits at different education stages.

[0039] In some implementation processes, the extraction of group borrowing characteristics and the generation of book borrowing evolution strategies based on the reference education stage specifically include:

[0040] Step S201: Determine multiple groups of group borrowing popularity data for different books in each reference education stage based on multiple groups of historical reading record data, extract local borrowing features of each book in each group of group borrowing popularity data, perform feature fusion based on the time series distribution of heat on multiple local borrowing features of each book, and generate group borrowing features for multiple books in the reference education stage.

[0041] Specifically, the group borrowing popularity data for each reference education stage is determined through multiple groups of historical reading record data, including the overall popularity data of different book types in the group under the reference education stage. For example, popularity is evaluated through borrowing frequency, borrowing duration, etc., and the popularity data of different book types is obtained by quantifying the distribution of overall borrowing frequency or borrowing duration. For each reference education stage, there will be a group of group borrowing popularity data corresponding to the same education stage in different time periods. The borrowing-related features of each book in each group of group borrowing popularity data are extracted to obtain the local borrowing features of each book in each group of group borrowing popularity data. Specifically, it can be the corresponding popularity features of different types of books in the group borrowing popularity data obtained after overall normalization.

[0042] For each book's multiple local borrowing characteristics, that is, the local borrowing characteristics of a certain type of book in different classes, we consider temporal evolution. For example, for a certain grade, there may be differences in the popularity of a certain type of book between the previous class and the current class. We extract the characteristic parameters of the time-series distribution of popularity from the multiple local borrowing characteristics of the book to serve as the group borrowing characteristics of different types of books. This can be represented by calculating the product of the mean and standard deviation of the popularity of a certain type of book in multiple classes. A larger mean value indicates that the overall popularity of this type of book is higher, while a larger standard deviation indicates that the overall popularity of this type of book is more evenly distributed over time. This characterizes the overall borrowing popularity of different types of books in a specific reference education stage.

[0043] Step S202: Determine the group borrowing characteristics of each reference education stage based on multiple groups of historical reading record data, including the time period distribution data of multiple books, and generate a book borrowing evolution strategy for each reference education stage.

[0044] Specifically, the group borrowing characteristics of the reference education stage include the overall popularity of different types of books in a specific education stage, such as the popularity of different types of books in the group in the first semester of the second grade. In this case, we further analyze the time distribution characteristics of each type of book in this educational stage based on multiple sets of historical reading record data. Specifically, we extract the overall distribution performance of the popularity of each type of book in a specific educational stage from multiple sets of historical reading record data. For example, it gradually increases or decreases from the beginning of the semester to the end of the semester, or the overall popularity is relatively average throughout the semester, so as to formulate a book borrowing evolution strategy for each reference educational stage. For example, books with relatively average popularity distribution and high overall popularity can be recommended at different times throughout the semester. Books with high overall popularity but only concentrated at the end of the semester can have their recommendation priority lowered in the early and middle stages of the semester. In this way, we obtain a book borrowing evolution strategy for multiple books in each reference educational stage, which can specifically include group borrowing preference data in each local stage of the reference educational stage. For example, in a certain semester, group borrowing preference data on different book types from an overall perspective in the early, middle and late stages of the semester, which is expressed as the borrowing reference popularity of different types of books. In actual scenarios, at the beginning of the semester, students may prefer relatively pleasant fairy tales or fables. However, as the semester progresses, as students' cognitive abilities gradually improve, they may be more suitable to read relatively complex popular science books, etc., to reflect the dynamic changes in the popularity of different types of books during the educational stage.

[0045] Step S30: Obtain target reading record data of the target user, determine the target education stage to which the target user currently belongs, perform group evolution deviation detection on the target user according to multiple book borrowing evolution strategies, and obtain group borrowing deviation data of the target user.

[0046] In this embodiment, the target reading record data for the target user, the individual to be analyzed, includes the target user's individual reading record data for different types of books during a specific educational stage. For example, the target user's relevant reading record data for the previous three months includes the user's overall book reading preferences during the current corresponding educational stage, i.e., the educational stage to which the target reading record data belongs. The target reading record data can be used to mine the deviation characteristics of the target user's preference evolution from the group's patterns, and analyze whether the target user differs from the group due to differences in knowledge mastery or cognitive ability, which manifests as deviations in book reading habits.

[0047] In some implementations, detecting group evolution deviation of target users specifically includes:

[0048] Extract the target user's individual borrowing preference data for each local stage from the target reading record data. For example, the book borrowing data corresponding to the beginning, middle, and end of a semester reflects the changes in individual reading habits at different times within the semester. Based on the target user's current target education stage, determine the target user's associated multiple book borrowing evolution strategies.

[0049] Specifically, the multiple target borrowing evolution strategies associated with the target user can be the current target education stage, as well as the target borrowing evolution strategies corresponding to several education stages before and after the target education stage. Then, multiple groups of individual borrowing preference data of the target user are matched with the multiple target borrowing evolution strategies to realize group evolution deviation detection of the target user and obtain the group borrowing deviation characteristics of the target user in each local stage.

[0050] In this process, the target user is subjected to group evolution deviation detection based on the group borrowing preference data and individual borrowing preference data in each local stage. Specifically, the difference between the user's borrowing preference and the group's borrowing preference in a specific local stage can be analyzed. For example, borrowing preference feature vectors representing the preference characteristics of different types of books are extracted from the group borrowing preference data and the individual borrowing preference data. Each element in the vector represents the quantitative value of the preference for a certain type of book. By analyzing and matching the borrowing preference feature vector of the user's individual in each local stage with the borrowing preference feature vector of the group in the local stage and the local stages before and after, The corresponding matching values, such as similarity parameters, are characterized by Euclidean distance, to determine whether the user's borrowing preferences at the current local stage are generally consistent, or appear to be ahead, behind, or consistent with group rules, and different phenomena are quantified and recorded to obtain the target user's group borrowing deviation characteristics at each local stage. Finally, multiple group borrowing deviation characteristics of the target user are integrated to calculate the overall proportion of different deviation states. For example, analysis shows that 60% of the borrowing preferences in multiple local stages are consistent with the group's borrowing preferences at the same local stage, and 40% are highly matched with the borrowing preferences of the next local stage. In this way, the target user's group borrowing deviation data is generated.

[0051] Step S40: Obtain borrowing popularity reference data of the target reading record data from the children's book sharing platform, perform heat influence stripping on the target user based on the borrowing popularity reference data and the target reading record data, and extract heat filtering preference data of the target user.

[0052] In this embodiment, the key goal of stripping the popularity influence of the target user is to identify the impact of the popularity effect on the target user's borrowing behavior, strip away these external interference factors, and ultimately obtain the user's true book preferences. In this process, the borrowing popularity reference data of the children's book sharing platform for the target reading record data is first obtained, that is, the overall borrowing popularity related data of different types of books in the platform during the time period of the target reading record data. Then, further analysis is conducted to determine whether some of the user's preferred book types are caused by short-term popularity mutations in the platform during the current time period. This can reduce the impact of short-term popularity mutation events on the platform on the target user's current preferences and obtain heat-filtered preference data that can more realistically represent the user's current preferences.

[0053] In some implementation processes, the heat influence stripping of target users specifically includes:

[0054] Determine multiple local book preference features of the target user in the target reading record data, extract multiple global book preference parameters of each local book preference feature from the borrowing popularity reference data, and calculate the popularity deviation factor of each local book preference feature based on the multiple global book preference parameters.

[0055] Specifically, the local book preference features can be some book types that exceed the preset popularity threshold in the overall popularity of different book types in the target reading record data. For these local book preference features, the corresponding popularity parameters can be obtained based on the borrowing frequency and other quantitative factors, and the borrowing popularity reference data can be further analyzed to calculate multiple global book preference parameters for each local book preference feature.

[0056] In this embodiment, the local heat intensity parameter, behavior decay parameter, and local interest matching parameter are used to characterize whether the user's book type preference in the target reading record data is affected by the local heat anomaly event in the platform. The local heat intensity parameter is characterized by the ratio between the borrowing frequency of a certain type of book and all types of books in the borrowing heat reference data. The local interest matching parameter is characterized by the ratio between the borrowing frequency of a certain type of book and all types of books in the target reading record data.

[0057] The behavior decay parameter is characterized by the ratio between the borrowing frequencies of a certain type of book in the first target window period and the second target window period in the target reading record data of the target user. The window size of the target window period can be set according to the time length involved in the target reading record data. For example, a quarter is set as the target sliding window, and then the borrowing popularity reference data is traversed through the target sliding window. The window with the highest total borrowing frequency of a specific type of book among multiple sliding windows is recorded as the second target window period, and the next event window of the second target window period is recorded as the first target window period. For example, if the total borrowing frequency of a certain type of book in the second week of the borrowing popularity reference data is the highest, the second week is recorded as the second target window period, and the time window corresponding to the third week is recorded as the first target window period. After locating the target window period through the borrowing popularity reference data, the target reading record data is analyzed through the target window period to measure whether the user's short-term preference is affected by the popularity event.

[0058] Finally, the heat deviation factor of any local book preference feature is calculated as follows: Where, The heat deviation factor representing the local book preference feature, The local heat intensity parameter representing the local book preference feature, The behavioral decay parameter representing the local book preference characteristics, Local interest matching parameters that represent local book preference characteristics. When the platform's popularity for a certain type of book is relatively high in the short term, and after borrowing this type of book, the user's subsequent interest in it decays rapidly, and at the same time, the user's overall preference for this type of book is relatively weak, this indicates that the user's interest in this type of book deviates significantly from their own habits. The larger the popularity deviation factor, the more likely the user has been affected by short-term popularity events. On the other hand, if a certain type of book is relatively popular on the platform in the short term, but the user's interest in this type of book is relatively stable over a certain period of time and continues to be good, that is, there is no severe decay phenomenon, then it can be more accurately regarded as the user's interest preference.

[0059] Finally, the heat influence of the local book preference features of the target user is stripped according to the heat deviation factor. In this process, for multiple local book preference features, the larger the corresponding heat deviation factor, the lower the reference value. The deviation influence weight obtained by normalizing multiple heat deviation factors can be used. The ratio between the heat parameter corresponding to the local book preference feature and the deviation influence weight obtained after normalization is used as the final heat parameter of each local book preference feature, thereby obtaining the heat filtering preference data of the target user on multiple local book preference features. In this way, the borrowing data affected by short-term popular factors are stripped off to ensure that the remaining data can better reflect the user's overall interest preferences in a specific period.

[0060] Step S50: The target user's group borrowing offset data and heat filtering preference data are integrated to generate a book borrowing matching strategy for the target user in the target reading record data, and the children's book sharing platform's book sharing strategy for the target user is optimized through the book borrowing matching strategy.

[0061] In this embodiment, for the target user's group borrowing offset data and popularity filtering preference data, the group borrowing offset data can reflect the differences in the user's reading behavior relative to the group at the same educational stage, such as being ahead, lagging behind, or in line with expectations. Although the popularity influencing factors are not filtered, it can reveal the objective state of the user's knowledge mastery progress and can also more objectively reflect the common laws of the educational stage. Even if certain books are widely borrowed by the group due to their popularity, the cognitive needs involved in the educational stage are consistent. For example, a math picture book becomes a short-term hot spot due to teacher recommendation. This type of popularity is reasonable and should be retained. Or, due to the influence of a certain teaching activity, students in a certain grade recommend that they concentrate on reading a book on natural observation in class. Although these are short-term hot spots, they are reasonable and necessary to retain because they are in line with teaching activities, which can more truly reflect the dynamic characteristics of the educational stage. The popularity filtering preference data can more essentially reflect the individual's long-term preference laws by stripping away the interference of short-term hot spots.

[0062] Therefore, the fusion process of group borrowing offset data and popularity filtering preference data can first determine the user's current target local stage and target educational stage, such as the midterm of the second semester of third grade. Based on the group borrowing offset data, the local borrowing evolution strategy associated with the target local stage is determined, including information extracted from the book borrowing evolution strategy for the target educational stage. Specifically, the local strategy information corresponding to the target local stage and the book borrowing evolution strategy for the target educational stage in the preceding or following local stage adjacent to the target local stage, determined based on the lead or lag in the group borrowing offset, is used. Furthermore, based on the group borrowing offset data, multiple strategy fusion weights for the local borrowing evolution strategy are determined. For example, if the group borrowing offset data shows 70% conformance to the group pattern and 30% lead, the strategy fusion weights for multiple book types in the local borrowing evolution strategy for the target local stage and the following local stage adjacent to the target local stage are 0.7 and 0.3, respectively. The strategy fusion weights are then used to weight the borrowing reference heat of multiple book types in the local borrowing evolution strategy to obtain the first heat parameter for each book type. And further based on the heat filtering preference data, the first heat parameter of each type of book is modified to obtain the second heat parameter of each type of book. Finally, according to the second heat parameter of the type of book, a book borrowing matching strategy for the target user regarding the target reading record data is generated. In this process, the real-time heat parameters of the target user regarding different types of books in the target local stage can be determined first, and then the difference between the second heat parameter and the real-time heat parameter of each type of book can be analyzed. If the second heat parameter is higher than the real-time heat parameter, the recommendation priority of the corresponding type of book is increased, otherwise the recommendation priority is reduced. Finally, the book borrowing matching strategy of the children's book sharing platform for the target user is optimized through the book borrowing matching strategy.

[0063] It is worth noting that due to the differences between children and adults in cognitive ability, learning progress, and interest preferences, for example, children's interests and learning needs are more likely to change significantly with age and grade. Some platforms recommend shared book resources through borrowing records, rating systems, etc. Due to the constant changes in children's cognitive development, they are more likely to have different preferences for the subject types of books at different times. If the dynamic nature of children's interest changes is not fully considered, it is likely that some recommended books will not fully meet the real needs of children at different learning stages. At the same time, factors such as hot events that occur in the short term and the individual differences in the mastery of teaching knowledge among different children will all affect children's needs. The present invention comprehensively considers the impact of the above factors on children's real preferences and formulates book sharing strategies for different users, so that the platform can achieve accurate personalized recommendations for different individuals in a wide range of children's groups, improve the efficiency of book use and sharing effects, and consider interest-driven flexible coordination under the overall teaching needs or teaching direction to enhance the reading experience of child users.

[0064] Figure 2 This is a structural diagram of an exemplary children's book sharing management system in an embodiment of the present invention. Figure 2 Based on the concept of the above-mentioned children's book sharing management method, a children's book sharing management system includes:

[0065] A reading data preprocessing module is used to collect historical reading record data and education stage data of multiple platform users on the children's book sharing platform, and determine multiple reference education stages based on the education stage data;

[0066] The book borrowing evolution analysis module is used to extract the group borrowing characteristics of each reference education stage from multiple sets of historical reading record data, perform stage distribution analysis on multiple group borrowing characteristics, and generate the book borrowing evolution strategy for each reference education stage;

[0067] The group borrowing deviation analysis module is used to obtain the target reading record data of the target user, determine the target education stage to which the target user currently belongs, and perform group evolution deviation detection on the target user based on multiple book borrowing evolution strategies to obtain the group borrowing deviation data of the target user;

[0068] The popularity impact stripping module is used to obtain the borrowing popularity reference data of the target reading record data on the children's book sharing platform, and based on the borrowing popularity reference data and the target reading record data, it strips the popularity impact of the target user and extracts the target user's popularity filtering preference data;

[0069] The book sharing strategy generation module is used to integrate the target user's group borrowing offset data and heat filtering preference data to generate the target user's book borrowing matching strategy in the target reading record data, and optimize the children's book sharing platform's book sharing strategy for the target user through the book borrowing matching strategy.

[0070] The foregoing description is merely a detailed description of the present invention, which is intended to enable those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art. Portions not described in detail in this specification are well known to those skilled in the art.

Claims

1. A children's book sharing management method, characterized in that: include: Collect historical reading record data and education stage data of multiple platform users on the children's book sharing platform, and determine multiple reference education stages based on the education stage data; Extracting group borrowing characteristics of each reference education stage from multiple sets of historical reading record data, performing stage distribution analysis on multiple group borrowing characteristics, and generating a book borrowing evolution strategy for each reference education stage, including determining multiple sets of group borrowing popularity data on different books in each reference education stage based on multiple sets of historical reading record data, extracting local borrowing characteristics of each book in each set of group borrowing popularity data, including normalized popularity characteristics corresponding to different types of books in group borrowing popularity data, performing feature fusion based on the time series distribution of popularity on multiple local borrowing characteristics of each book, generating group borrowing characteristics for multiple books in the reference education stage, determining time period distribution data for multiple books in the group borrowing characteristics of each reference education stage based on multiple sets of historical reading record data, and generating a book borrowing evolution strategy for each reference education stage, including group borrowing preference data in each local stage of the reference education stage; Obtain target reading record data of a target user, determine the target education stage to which the target user currently belongs, perform group evolution offset detection on the target user according to multiple book borrowing evolution strategies, and obtain group borrowing offset data of the target user, including extracting individual borrowing preference data of the target user in each local stage from the target reading record data, determining multiple target borrowing evolution strategies associated with the target user in multiple book borrowing evolution strategies according to the target education stage to which the target user currently belongs, matching multiple groups of individual borrowing preference data of the target user with the multiple target borrowing evolution strategies, including performing group evolution offset detection on the target user according to the group borrowing preference data and individual borrowing preference data in each local stage, obtaining group borrowing offset features of the target user in each local stage, and fusing the group borrowing offset features to obtain the group borrowing offset data of the target user; Obtain borrowing popularity reference data of the target reading record data on the children's book sharing platform, and based on the borrowing popularity reference data and the target reading record data, separate the popularity influence of the target user and extract the target user's popularity filtering preference data; By integrating the target user's group borrowing offset data and heat filtering preference data, a book borrowing matching strategy for the target user in the target reading record data is generated. The book borrowing matching strategy is used to optimize the children's book sharing platform's book sharing strategy for the target user.

2. A children's book sharing management method according to claim 1, characterized in that: Remove the target user's popularity influence and extract the target user's popularity filtering preference data, including: Determine multiple local book preference features of the target user in the target reading record data, where the local book preference features are multiple book types whose popularity is greater than a preset popularity threshold in the target reading record data, extract multiple global book preference parameters of each local book preference feature from the borrowing popularity reference data, calculate a popularity deviation factor of each local book preference feature based on the multiple global book preference parameters, and perform heat influence stripping on the local book preference features of the target user based on the heat deviation factor to obtain heat filtered preference data of the target user regarding the multiple local book preference features.

3. A children's book sharing management method according to claim 2, characterized in that: By integrating the target user's group borrowing offset data and popularity filtering preference data, a book borrowing matching strategy is generated for the target user in the target reading record data, including: Determine the target local stage to which the user belongs, combine the target education stage to which the target user belongs, and extract the local borrowing evolution strategy associated with the target local stage from the book borrowing evolution strategy of the target education stage; Determine multiple strategy fusion weights for local borrowing evolution strategies based on group borrowing offset data, perform heat analysis on the local borrowing evolution strategies using the strategy fusion weights, and calculate first heat parameters for multiple books; Multiple first popularity parameters are modified and processed through popularity filtering preference data, and the second popularity parameter corresponding to each first popularity parameter is calculated. Based on the second popularity parameters of multiple books, a book borrowing matching strategy for the target user in the target reading record data is generated.

4. A children's book sharing management method according to claim 2, characterized in that: Perform group evolution deviation detection on the target user to obtain the group borrowing deviation characteristics of the target user in each local stage, including: The borrowing preference feature vectors corresponding to the group borrowing preference data and the individual borrowing preference data are extracted respectively, and the borrowing preference feature vectors corresponding to the group borrowing preference data and the individual borrowing preference data in each local stage are matched and analyzed, and the similarity parameters in each local stage are calculated. Based on multiple similarity parameters, the group borrowing offset characteristics of the target user in each local stage are generated.

5. A children's book sharing management system, characterized in that: The system is used to implement the children's book sharing management method according to any one of claims 1 to 4, comprising: A reading data preprocessing module is used to collect historical reading record data and education stage data of multiple platform users on the children's book sharing platform, and determine multiple reference education stages based on the education stage data; The book borrowing evolution analysis module is used to extract the group borrowing characteristics of each reference education stage from multiple sets of historical reading record data, perform stage distribution analysis on multiple group borrowing characteristics, and generate the book borrowing evolution strategy for each reference education stage; The group borrowing deviation analysis module is used to obtain the target reading record data of the target user, determine the target education stage to which the target user currently belongs, and perform group evolution deviation detection on the target user based on multiple book borrowing evolution strategies to obtain the group borrowing deviation data of the target user; The popularity impact stripping module is used to obtain the borrowing popularity reference data of the target reading record data on the children's book sharing platform, and based on the borrowing popularity reference data and the target reading record data, it strips the popularity impact of the target user and extracts the target user's popularity filtering preference data; The book sharing strategy generation module is used to integrate the target user's group borrowing offset data and heat filtering preference data to generate the target user's book borrowing matching strategy in the target reading record data, and optimize the children's book sharing platform's book sharing strategy for the target user through the book borrowing matching strategy.

Images