Intelligent publishing system based on big data

Abstract

The invention provides an intelligent publishing system based on big data, and relates to the field of digital publishing. The intelligent publishing system based on the big data comprises a user construction module which is used for modeling user interests through a machine learning algorithm according to personal information input by a user and operation on a platform; the publication extraction module is used for extracting features of published products on the platform and making the published products correspond to user ranges; the personalized customization module is used for accurately recommending published products with corresponding characteristics to the user based on content circulation of big data; the intelligent printing and issuing module is used for comprehensively analyzing data such as historical sales conditions of publications of the same kind, publication warehouse reserves and related user data through big data to determine the printing and issuing quantity, and printing according to needs; and a big data auditing module. And through the use of the deep learning unit, the recommendation accuracy is improved, continuous changes of user interests can be adapted, and the timeliness and relevance of recommended contents are ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of digital publishing, and in particular to an intelligent publishing system based on big data. Background Art

[0002] With the prosperity of the Internet and digital publishing technology, self-publishing has become a phenomenon that cannot be ignored. Self-publishing in the current publishing context in my country mainly refers to self-publishing using online platforms as the medium, that is, online self-publishing; the connotation of "self" in self-publishing is further decomposed into "six selfs": self-created works, self-owned copyrights, self-raised funds, independent management, self-risk, and self-enjoyment of benefits; in this model, traditional publishing houses, platform operators, and copyright agencies only play a role as publishing service providers, and the entire self-publishing process is led by the author.

[0003] On the other hand, there is always a certain lag between publishers and market demand, which leads to the inability of publications to reasonably adapt to market demand, and the inability to carry out personalized digital publishing and information customization for readers and customer groups. The stickiness and interest focus of users will shift, resulting in the loss of readers and customers. The existing digital publishing technology lacks a personalized information recommendation and analysis system based on a data platform as the center, and is unable to carry out personalized digital publishing, information customization, and precise targeted publishing for readers and customer groups. The stickiness and interest focus of users will shift, resulting in the loss of readers and customers. Summary of the invention

[0004] 1. Technical issues to be solved

[0005] In view of the deficiencies of the prior art, the present invention provides an intelligent publishing system based on big data, which solves the problem that the existing publishing system is unable to perform personalized digital publishing, information customization, and precise targeted publishing for readers and customer groups.

[0006] (II) Technical solution

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: A big data-based intelligent publishing system, comprising:

[0008] User construction module: Models user interests through machine learning algorithms based on the personal information input by users and their operations on the platform, including a data collection unit, a data storage unit, a label assignment unit, and a label update unit. The data collection unit is used to collect user data, the data storage unit is used to store user operations on the platform, the label assignment unit is used to assign different identification labels to users to reduce the difficulty of subsequent personalized recommendations for users, and the label update unit updates the assigned labels according to user operations;

[0009] Publishing extraction module: extracts features from published products on the platform and matches published products with user ranges, including data preprocessing unit, feature word recognition unit, specific word filtering unit and feature classification unit. The data preprocessing unit is used to preprocess unit segmentation, part-of-speech tagging and data cleaning to prepare review corpus data; the feature word recognition unit identifies and selects product feature words through association rules, dependency syntax analysis and other methods, and classifies, removes duplicates and cleans the extracted feature words; the specific word filtering unit is used to filter out generalized nouns and verbs without obvious sentiment polarity to ensure the relevance and accuracy of the extracted feature words and opinion words; the feature classification unit uses the synonym forest and word vector method to expand feature seed words and screen and merge feature words based on large-scale review data combined with Chinese language characteristics;

[0010] Personalized customization module: Based on the content circulation of big data, it accurately recommends publishing products with corresponding characteristics to users, including deep learning units and content reorganization units. The deep learning unit is used to capture more complex relationships between user interest characteristics and video characteristics; the content reorganization unit is based on the structured content of fine-grained knowledge units and is customized and reorganized according to user needs to meet users' nonlinear and fragmented personalized needs;

[0011] Intelligent printing and distribution module: Through the comprehensive analysis of big data such as the historical sales of similar publications, the storage volume of publications, and relevant user data, the number of copies to be printed and distributed can be determined, and printing can be carried out on demand to reduce warehouse backlogs and the resulting storage costs;

[0012] Big data review module: The module reviews the content of published products through users, and conducts a secondary review of the review content proposed by users through intelligent means.

[0013] Preferably, the user's personal information includes but is not limited to age, gender, profession and interests, so as to facilitate accurate collection of the user's personal information.

[0014] Preferably, the user's operations on the platform include but are not limited to viewing, liking, commenting, sharing and searching, so as to facilitate real-time updating of the user's interests.

[0015] Preferably, the content circulation of the big data is based on the user's personal situation, consumption habits, reading preferences and other data.

[0016] Preferably, the personalized customization module also recommends publication products that are liked by other similar users to the user by comparing the viewing and interaction behaviors of similar users through collaborative filtering technology.

[0017] Preferably, the big data review module also includes a manual review function and an intelligent preliminary review function. The intelligent preliminary review function applies intelligent technology to the field of content risk control, relying on natural language processing, cross-modal audio and video recognition and other technologies to achieve intelligent review of content text, pictures and videos. The manual review function reviews user questions about intelligent review through manual review.

[0018] Preferably, the intelligent printing module also includes intelligent typesetting, intelligent construction of knowledge system and intelligent production of audio and video.

[0019] Preferably, the personalized customization module also assists editors in topic planning through big data and artificial intelligence technology, and explores potential readers and the content and topics that interest them.

[0020] (III) Beneficial effects

[0021] The present invention provides an intelligent publishing system based on big data, which has the following beneficial effects:

[0022] 1. The intelligent publishing system uses deep learning units to analyze the circulation of big data content, thereby achieving accurate recommendations based on user characteristics, and using deep learning technology to capture the complex relationship between user interests and publishing product characteristics. The application of deep learning technology enables the system to process and analyze large amounts of user data and publishing product characteristics, thereby providing highly personalized recommendations. This approach not only improves the accuracy of recommendations, but also adapts to the ever-changing interests of users, ensuring the timeliness and relevance of recommended content.

[0023] 2. This intelligent publishing system improves the production efficiency and quality of publications through automation technology. The application of intelligent typesetting and content production technology reduces manual intervention and ensures the consistency and professionalism of publications. The innovation of this approach is that it uses the latest technical means to simplify the publishing process and improve production efficiency.

[0024] 3. The personalized customization module in the intelligent publishing system uses collaborative filtering technology to compare the viewing and interaction behaviors of similar users and recommends publishing products that other similar users like. This innovation is that it uses the similarity of user behavior to optimize the recommendation algorithm, increasing the diversity and accuracy of recommendations. The application of collaborative filtering technology enables the system to discover new content that users may be interested in, thereby improving user experience and satisfaction. DETAILED DESCRIPTION

[0025] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0026] Example:

[0027] The embodiment of the present invention provides an intelligent publishing system based on big data, including:

[0028] User construction module: Modeling user interests through machine learning algorithms based on the personal information input by users and their operations on the platform. The user's personal information includes but is not limited to age, gender, profession and interest range, which facilitates accurate collection of user personal information. The user's operations on the platform include but are not limited to watching, liking, commenting, sharing and searching, which facilitates real-time updating of user interests. It includes a data collection unit, a data storage unit, a label assignment unit and a label update unit. The data collection unit is used to collect user data, the data storage unit is used to store user operations on the platform, the label assignment unit is used to assign different identification labels to users to reduce the difficulty of subsequent personalized recommendations for users, and the label update unit updates the assigned labels according to the user's operations;

[0029] Publishing extraction module: extracts features from published products on the platform and matches published products with user ranges, including data preprocessing unit, feature word recognition unit, specific word filtering unit and feature classification unit. The data preprocessing unit is used to preprocess unit segmentation, part-of-speech tagging and data cleaning to prepare review corpus data; the feature word recognition unit identifies and selects product feature words through association rules, dependency syntax analysis and other methods, and classifies, removes duplicates and cleans the extracted feature words; the specific word filtering unit is used to filter out generalized nouns and verbs without obvious sentiment polarity to ensure the relevance and accuracy of the extracted feature words and opinion words; the feature classification unit uses the synonym forest and word vector method to expand feature seed words and screen and merge feature words based on large-scale review data combined with Chinese language characteristics;

[0030] Personalized customization module: Based on the content circulation of big data, it accurately recommends publishing products with corresponding characteristics to users, including deep learning units and content reorganization units. The content circulation of big data is based on the user's personal situation, consumption habits, reading preferences and other data. The personalized customization module also uses collaborative filtering technology to compare the viewing and interaction behaviors of similar users to recommend other similar users' favorite publishing products. The personalized customization module also uses big data and artificial intelligence technology to assist editors in topic planning, explore potential readers and the content and topics they are interested in. The deep learning unit is used to capture more complex relationships between user interest characteristics and video characteristics; the content reorganization unit is based on the structured content of fine-grained knowledge units and is customized and reorganized according to user needs to meet the user's nonlinear and fragmented personalized needs;

[0031] Intelligent printing and distribution module: Through big data, comprehensive analysis of historical sales of similar publications, storage volume of publications, relevant user data and other data is used to determine the number of prints and distribution, and print on demand to reduce warehouse backlogs and the resulting storage costs. The intelligent printing module also includes intelligent typesetting, intelligent construction of knowledge systems, and intelligent production of audio and video.

[0032] Big data review module: The module reviews the content of published products through users, and conducts a secondary review of the review content proposed by users through intelligent means. The big data review module also includes manual review function and intelligent preliminary review function. The intelligent preliminary review function applies intelligent technology to the field of content risk control, relying on natural language processing, cross-modal audio and video recognition and other technologies to achieve intelligent review of content text, pictures and videos. The manual review function reviews user questions about intelligent review through manual review.

[0033] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent publishing system based on big data, characterized in that: include: User building module: Modeling user interests through machine learning algorithms based on the personal information entered by users and their operations on the platform; Publishing extraction module: extracts features of publishing products on the platform and matches publishing products with user ranges; Personalized customization module: Based on the content circulation of big data, it accurately recommends publishing products with corresponding characteristics to users; Intelligent printing and distribution module: through the comprehensive analysis of big data such as the historical sales of similar publications, the storage volume of publications, and relevant user data, the number of copies to be printed is determined and printed on demand; Big data review module: The module reviews the content of published products through users, and conducts a secondary review of the review content proposed by users through intelligent means.

2. The intelligent publishing system based on big data according to claim 1, characterized in that: The user's personal information includes but is not limited to age, gender, profession and interests.

3. The intelligent publishing system based on big data according to claim 1, characterized in that: The user's operations on the platform include but are not limited to watching, liking, commenting, sharing and searching.

4. The intelligent publishing system based on big data according to claim 1, characterized in that: The content circulation of the big data is based on the user's personal situation, consumption habits, reading preferences and other data.

5. The intelligent publishing system based on big data according to claim 1, characterized in that: The personalized customization module also recommends publications that are liked by other similar users through collaborative filtering technology by comparing the viewing and interaction behaviors of similar users.

6. The intelligent publishing system based on big data according to claim 1, characterized in that: The big data review module also includes a manual review function and an intelligent preliminary review function. The intelligent preliminary review function applies intelligent technology to the field of content risk control, relying on natural language processing, cross-modal audio and video recognition and other technologies to achieve intelligent review of content texts, pictures and videos. The manual review function reviews user questions about intelligent review through manual review.

7. The intelligent publishing system based on big data according to claim 1, characterized in that: The intelligent printing module also includes intelligent typesetting, intelligent construction of knowledge system and intelligent production of audio and video.

8. The intelligent publishing system based on big data according to claim 1, characterized in that: The personalized customization module also assists editors in topic planning through big data and artificial intelligence technology, and explores potential readers and the content and topics that interest them.