A digital information push method and system based on big data

Abstract

This application relates to the technical field of information push, and in particular, to a digital information push method and system based on big data. The method includes: obtaining the browsing data of a user; analyzing the browsing data to determine whether there is singularity in the browsing; if so, obtaining the behavior data of the user, analyzing the behavior data to determine the dialectical ability of the user; determining a push plan and performing information push according to the dialectical ability. By analyzing the browsing behavior of the user, different types or viewpoints of information are pushed to help the user break the information cocoon and access more diverse content. Based on the analysis of the user behavior data and the dialectical ability, it is possible to more accurately predict which content the user is interested in, so as to push more relevant content, improve user satisfaction and participation. According to the dialectical ability of the user, personalized content that better conforms to the user's cognitive level and interest preference is provided, thereby enhancing the user experience.

Description

Technical Field

[0001] The present application relates to the technical field of information push, and in particular, to a digital information push method and system based on big data. Background Art

[0002] With the rapid development of Internet technology and the popularization of big data, information push technology, as an important part of personalized recommendation systems, has been widely applied in fields such as e-commerce, news reading, and social media; traditional information push methods mostly rely on explicit data such as users' browsing history and interest tags for recommendation.

[0003] However, users' browsing behaviors often have limitations. For example, they are long-term concentrated on content of a single type or view, resulting in the information cocoon effect and being unable to improve the diversity and depth of users' information reception. Therefore, how to provide users with diversified and personalized content that fits their cognitive level has become an important research direction in current information push technology. Summary of the Invention

[0004] The present application provides a digital information push method and system based on big data to solve the above problems.

[0005] In a first aspect, the present application provides a digital information push method based on big data, the method comprising:

[0006] Obtaining browsing data of a user; analyzing the browsing data to determine whether there is a singularity in the browsing;

[0007] If so, obtaining the behavior data of the user, analyzing the behavior data to determine the dialectical ability of the user;

[0008] Determining a push plan based on the dialectical ability and performing information push.

[0009] Through this solution, by analyzing the user's browsing behavior, if it is found that the user has long been concerned about content of a single type or view, information of different types or views will be pushed to help the user break the information cocoon and access more diversified content. Based on the analysis of the user's behavior data and dialectical ability, it is possible to more accurately predict which content the user is interested in, so as to push more relevant content, improve user satisfaction and engagement. Pushing content of different views encourages the user to think deeply and make comparisons, which helps to improve the user's dialectical thinking ability. According to the user's dialectical ability, personalized content that better matches the user's cognitive level and interest preferences can be provided, thereby enhancing the user experience. When the user receives relevant and valuable information, they are more likely to continue using the service, thus increasing user stickiness.

[0010] Optionally, the analyzing the browsing data to determine whether there is a singularity in the browsing includes:

[0011] Analyze the browsing data to determine the type of the browsed content;

[0012] Based on the type, determine the content similarity of each piece of browsed content;

[0013] According to the content similarity, determine the browsing attitude of the user towards each piece of browsed content;

[0014] According to the browsing attitude and the type, determine whether there is a singularity in the browsing.

[0015] Through this solution, by analyzing the user's browsing data, the user's interest preferences can be understood more accurately, so as to push more relevant content. Identify whether the user is long-term concentrated on content of a certain type or view, so as to push information of different types or views, help the user access diversified content, and avoid falling into an information cocoon.

[0016] Optionally, the behavior data includes click behavior, and the analyzing the behavior data to determine the dialectical ability of the user includes:

[0017] Analyze the click behavior to determine the clicked content;

[0018] Analyze the clicked content to determine the attitude held by the clicked object;

[0019] Compare the held attitude with the browsing attitude to determine whether the content view of the clicked content is consistent with that of the browsed content;

[0020] If they are consistent, based on the click behavior, determine the stay time of the clicked content;

[0021] According to the stay time, determine the dialectical ability of the user.

[0022] Through this solution, by analyzing the user's click behavior, the user's interest preferences can be understood more accurately, so as to push more relevant content. Provide personalized content recommendations according to the user's click behavior and attitude to improve the user experience. Push content of different views to encourage the user to think deeply and make comparisons, which helps to improve the user's dialectical thinking ability. Identify whether the user is long-term concentrated on content of a certain type or view, so as to push information of different types or views, help the user access diversified content, and avoid falling into an information cocoon.

[0023] Optionally, the determining the dialectical ability of the user according to the stay time includes:

[0024] During the stay time, obtain the stay operation of the user;

[0025] Analyze the stay operation to determine whether there is a situation of opposing views among the users;

[0026] If there is, analyze the situation of opposing views to determine the dialectical ability of the users;

[0027] If not, determine that the users lack dialectical ability.

[0028] Through this solution, by analyzing the stay time of users on content with opposing views, content that matches the users' interests and dialectical ability can be pushed more accurately, improving the accuracy of recommendations to achieve personalized information push based on user stay time data. Push diversified content that conforms to the users' cognitive level and dialectical ability. By pushing content with different views, help users access and understand diversified information.

[0029] Optionally, the behavior data includes search behavior, and the analyzing the behavior data to determine the dialectical ability of the users includes:

[0030] Analyze the search behavior to determine the search content of the users;

[0031] Analyze the search content to determine the similarity between the search content and the browsing content;

[0032] Determine the dialectical ability of the users according to the similarity.

[0033] Through this solution, by analyzing the search behavior of users, content that matches the users' interests and dialectical ability can be pushed more accurately, improving the accuracy of recommendations. Pushing diversified content that conforms to the users' cognitive level and dialectical ability can enhance the users' satisfaction and loyalty to the platform. Encouraging users to stay on content with opposing views can promote users to think deeply and conduct critical analysis, improving the users' dialectical thinking ability.

[0034] Optionally, the determining the dialectical ability of the users according to the similarity includes:

[0035] Compare the similarity with a preset similarity threshold, and according to the comparison result, determine whether the search content is consistent with the browsing content;

[0036] If they are not consistent, analyze the search content to determine the actual description of the search content;

[0037] According to the actual description, determine whether there is an association between the search content and the browsing content;

[0038] If it is determined that there is an association, determine the association relationship between the search content and the browsing content according to the actual description;

[0039] Determine the dialectical ability of the user according to the association relationship.

[0040] Through this solution, by comparing the similarity between the search content and the browsing content, the user's interests and needs can be more accurately identified, so as to provide more relevant recommended content. Through similarity analysis, the user's attitudes and cognitive levels towards different information can be evaluated, and personalized content that better suits the user's cognitive ability can be provided. Adjusting the recommendation strategy according to the user's dialectical ability can enhance the user experience and make the user feel that the recommended content is more in line with their personal interests and needs. By identifying the inconsistencies between the search content and the browsing content, it is possible to avoid pushing overly single or one-sided content to the user and reduce recommendation biases.

[0041] Optionally, the determining the push plan according to the dialectical ability includes:

[0042] Determine the push type of the push content according to the dialectical ability;

[0043] Obtain the operation data and browsing data of the user within a preset time period;

[0044] Analyze the operation data to determine whether the user's holding attitude towards each piece of browsing data is consistent with the corresponding type of browsing attitude;

[0045] Compare the consistency with a preset accuracy threshold. If the consistency is higher than the preset accuracy threshold, determine the moment when the consistency is higher than the preset accuracy threshold as the push moment;

[0046] Determine the push plan according to the push type and the push moment.

[0047] Through this solution, the push content is more in line with the user's cognitive level and dialectical ability, provides more in-depth and comprehensive information, and meets the user's personalized needs. When the user receives relevant and valuable information, they are more likely to participate in interactions, such as commenting, sharing, or further exploring related content. By pushing different viewpoints and diverse content, it helps users break through the information cocoon and access and understand more diverse information. Providing content that matches the user's dialectical ability can promote the user's knowledge acquisition and the improvement of their cognitive ability. Adjusting the recommendation strategy according to the user's dialectical ability can enhance the user experience and make the user feel that the recommended content is more in line with their personal interests and needs.

[0048] Optionally, the determining the push type of the push content according to the dialectical ability includes:

[0049] Determine the acceptability of the user for different types of content according to the dialectical ability;

[0050] Determine the push type according to the acceptability.

[0051] Through this solution, the pushed content can stimulate users' thinking. Especially for users with a high level of critical thinking, they can exercise and improve their critical thinking ability by being exposed to diverse viewpoints and challenging content. For users with a low level of critical thinking, gradually introducing more diverse viewpoints and popular science content helps users broaden their knowledge and understand knowledge in different fields. In the process of being exposed to different viewpoints, users can learn how to evaluate the credibility of information and improve their information literacy. By pushing content with a deviation close to the viewpoints the user is interested in, it is possible to prevent users from being trapped in an information cocoon and promote users to be exposed to and accept different information and viewpoints. For users with a high level of critical thinking, by presenting opposing viewpoints, it can promote rational discussions among users and reduce unnecessary conflicts and misunderstandings. By pushing different viewpoints and challenging content, it can cultivate users' open-mindedness and make users more willing to accept and understand different opinions and viewpoints.

[0052] Optionally, determining the dialectical ability of the user according to the association relationship includes:

[0053] Analyze the search content and the browsing content to determine the search source and the browsing source;

[0054] Determine the content credibility according to the search source and the browsing source;

[0055] Obtain the user portrait; determine the cognitive level of the user according to the user portrait and the content credibility;

[0056] Determine the dialectical ability of the user according to the association relationship and the cognitive level.

[0057] Through this solution, by analyzing users' search and browsing behaviors, it is possible to more accurately understand users' interests and preferences. According to users' cognitive levels and dialectical abilities, pushing content with different depths and diversities helps users be exposed to and understand more information and avoid the information cocoon effect. Pushing diverse viewpoints and challenging content for users with a high level of critical thinking can promote users' in-depth thinking and critical analysis, thereby enhancing their thinking ability.

[0058] In a second aspect, the present application provides a big data-based digital information push system, and the system includes:

[0059] A browsing analysis module, configured to obtain users' browsing data; analyze the browsing data to determine whether there is monotony in browsing;

[0060] A behavior analysis module, configured to, if there is, obtain the user's behavior data, analyze the behavior data, and determine the dialectical ability of the user;

[0061] A solution determination module, configured to determine a push solution according to the dialectical ability and perform information push.

[0062] Optionally, when analyzing the browsing data to determine whether there is a singularity in browsing, the browsing analysis module is configured to:

[0063] Analyze the browsing data to determine the type of browsing content;

[0064] Based on the type, determine the content similarity of each piece of browsing content;

[0065] According to the content similarity, determine the browsing attitude of the user towards each piece of browsing content;

[0066] According to the browsing attitude and the type, determine whether there is a singularity in browsing.

[0067] Optionally, when the behavior data includes click behavior, when analyzing the behavior data to determine the dialectical ability of the user, the behavior analysis module is configured to:

[0068] Analyze the click behavior to determine the click content;

[0069] Analyze the click content to determine the attitude held by the click object;

[0070] Compare the held attitude with the browsing attitude to determine whether the click content is consistent with the content view of the browsing content;

[0071] If they are consistent, based on the click behavior, determine the stay time of the click content;

[0072] According to the stay time, determine the dialectical ability of the user.

[0073] Optionally, when determining the dialectical ability of the user according to the stay time, the behavior analysis module is configured to:

[0074] During the stay time, obtain the stay operation of the user;

[0075] Analyze the stay operation to determine whether there is a situation of opposing views of the user;

[0076] If there is, analyze the situation of opposing views to determine the dialectical ability of the user;

[0077] If not, determine that the user lacks dialectical ability.

[0078] Optionally, when the behavior data includes search behavior, when analyzing the behavior data to determine the dialectical ability of the user, the behavior analysis module is configured to:

[0079] Analyze the search behavior to determine the search content of the user;

[0080] Analyze the search content to determine the similarity between the search content and the browsing content;

[0081] Determine the dialectical ability of the user according to the similarity.

[0082] Optionally, when determining the dialectical ability of the user according to the similarity, the behavior analysis module is used to:

[0083] Compare the similarity with a preset similarity threshold, and determine whether the search content is consistent with the browsing content according to the comparison result;

[0084] If they are inconsistent, analyze the search content to determine the actual description of the search content;

[0085] Determine whether there is an association between the search content and the browsing content according to the actual description;

[0086] If it is determined that there is an association, determine the association relationship between the search content and the browsing content according to the actual description;

[0087] Determine the dialectical ability of the user according to the association relationship.

[0088] Optionally, when determining the push scheme according to the dialectical ability, the scheme determination module is used to:

[0089] Determine the push type of the push content according to the dialectical ability;

[0090] Obtain the operation data and browsing data of the user within a preset time period;

[0091] Analyze the operation data to determine whether the holding attitude of the user towards each browsing data is consistent with the corresponding type of browsing attitude;

[0092] Compare the consistency with a preset accuracy threshold. If the consistency is higher than the preset accuracy threshold, determine the moment when the consistency is higher than the preset accuracy threshold as the push moment;

[0093] Determine the push scheme according to the push type and the push moment.

[0094] Optionally, when determining the push type of the push content according to the dialectical ability, the scheme determination module is used to:

[0095] Determine the acceptability of the user for different types of content according to the dialectical ability;

[0096] Determine the push type according to the acceptability.

[0097] Optionally, when determining the dialectical ability of the user according to the association relationship, it is used for:

[0098] Analyze the search content and the browsing content to determine the search source and the browsing source;

[0099] Determine the content credibility according to the search source and the browsing source;

[0100] Obtain the user portrait; determine the cognitive level of the user according to the user portrait and the content credibility;

[0101] Determine the dialectical ability of the user according to the association relationship and the cognitive level. Description of the Drawings

[0102] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained according to these drawings without creative efforts.

[0103] Figure 1 A schematic diagram of an application scenario provided by an embodiment of the present application;

[0104] Figure 2 A flowchart of a method for pushing digital information based on big data provided by an embodiment of the present application;

[0105] Figure 3 A schematic diagram of the structure of a system for pushing digital information based on big data provided by an embodiment of the present application. Detailed Embodiments

[0106] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some, but not all, of the embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the protection scope of the present application.

[0107] In addition, the term "and / or" in this text is merely a description of the association relationship between associated objects, indicating that there can be three relationships. For example, A and / or B can represent three situations: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this text generally represents an "or" relationship between the front and back associated objects unless otherwise specified.

[0108] The embodiments of the present application will be further described in detail below with reference to the accompanying drawings of the specification.

[0109] Users' browsing behaviors often have limitations. For example, they may focus on content of a single type or view for a long time, resulting in the information cocoon effect and being unable to improve the diversity and depth of users' information reception. Therefore, how to provide users with diversified and personalized content that matches their cognitive level has become an important research direction in current information push technology.

[0110] Based on this, the present application provides a big data-based digital information push method and system, which obtains users' browsing data; analyzes the browsing data to determine whether there is monotony in browsing; if so, obtains the users' behavior data, analyzes the behavior data to determine the users' dialectical ability; determines a push plan based on the dialectical ability and performs information push. By analyzing users' browsing behaviors, if it is found that users have long been concerned about content of a single type or view, information of different types or views will be pushed to help users break out of the information cocoon and access more diversified content. Based on the analysis of users' behavior data and dialectical ability, it is possible to more accurately predict which content users are interested in, so as to push more relevant content, improve user satisfaction and engagement. Pushing content with different views encourages users to think deeply and make comparisons, which helps to improve users' dialectical thinking ability. According to users' dialectical ability, personalized content that better matches users' cognitive level and interest preferences can be provided, thereby enhancing the user experience. When users receive relevant and valuable information, they are more likely to continue using the service, thus increasing user stickiness.

[0111] Figure 1A schematic diagram of an application scenario provided for this application. When in the scenario, the method provided for this application is applied. Specifically, the method provided for this application is applied to any server. The server interacts with the client application, obtains the browsing data authorized by the user on the client application, comprehensively understands the user's interests and preferences, and provides a basis for subsequent personalized recommendations. Analyze the browsing data to determine whether there is singularity in the browsing; it can identify whether the user is trapped in an information cocoon and provide a basis for adjusting the push strategy. Obtain the behavior data authorized by the user on the client to enhance the accuracy of user behavior analysis. Analyze the behavior data to determine the user's dialectical ability and provide a basis for pushing content with different viewpoints. According to the dialectical ability, determine the push plan, customize the push plan based on the user's dialectical ability, improve the accuracy of the pushed content and user satisfaction. Perform information push on the client to achieve accurate push of personalized content, improve the user experience and information reception efficiency. The specific implementation method can refer to the following embodiments.

[0112] Figure 2 A flowchart of a digital information push method based on big data provided for an embodiment of this application. The method of this embodiment can be applied to the server in the above scenario. As Figure 2 shown, the method includes:

[0113] S201. Obtain the browsing data of the user; analyze the browsing data to determine whether there is singularity in the browsing.

[0114] The browsing data can be all relevant data generated when the user browses web pages, applications or platforms on the Internet, such as the URL of the accessed web page, the content of the browsed page, the page stay time, the page scrolling behavior, the browsing frequency, etc.

[0115] Singularity can be the phenomenon that whether there is a long-term concentration on the content of a specific type or viewpoint in the user's browsing behavior. If the user always views a certain type of content or viewpoint, then it can be considered that their browsing behavior has singularity. In this case, the user may be trapped in an information cocoon and it is difficult to access diversified information.

[0116] Specifically, in an environment of information explosion, users' browsing behaviors often have limitations. For example, they may focus on content of a single type or view for a long time, resulting in the information cocoon effect. To effectively break this limitation and enhance the diversity and depth of users' information reception, it is necessary to collect users' browsing data on network platforms through user authorization, including the URLs of visited web pages, page content, access time, etc. Clean the obtained data to remove duplicate and incorrect data to ensure data quality. Store the cleaned browsing data in a database. Use natural language processing technology to classify the browsing content to determine the content areas that users mainly focus on. Calculate the similarity between the content browsed by users. Text similarity algorithms such as cosine similarity can be used. Analyze users' browsing attitudes towards this content based on the content they browse. For example, through the time users stay on the page, page scrolling behavior, etc. Based on the above analysis results, judge whether there is a singularity in users' browsing, that is, whether they mainly focus on content of a certain specific type or view.

[0117] S202. If so, obtain the user's behavior data, analyze the behavior data, and determine the user's dialectical ability.

[0118] Behavior data can be interactive behavior records generated by users on the Internet, such as click behaviors (such as clicking on links, buttons), search behaviors (such as entering keywords, viewing search results), interactive behaviors (such as liking, commenting, sharing), etc.

[0119] Dialectical ability can be the user's ability to accept and judge different views, which can reflect whether the user can accept and think about information different from existing views, and whether the user can form his own opinions based on comparing and evaluating different views.

[0120] Specifically, if there is a singularity in the user's browsing data, collect the user's click behavior data on the Internet, including the clicked links, click frequencies, etc. Analyze the content clicked by the user to determine the attitude held by the clicked object. For example, through content analysis or user interactive behaviors. Compare the user's click attitude with the browsing attitude to judge whether there is consistency. For the clicked content with consistent attitudes, analyze the user's stay time as the basis for evaluating dialectical ability. Combine the stay time and user operations to evaluate the user's dialectical ability, such as whether there are situations of opposing views.

[0121] S203. Determine the push plan according to the dialectical ability and perform information push.

[0122] The push plan can be an information push plan customized for users based on the user's behavior data and dialectical ability, such as determining the type of push content, push time, push frequency, etc.

[0123] Information pushing can be a process of actively sending customized information content to users through an Internet platform.

[0124] Specifically, according to the user's dialectical ability, determine the type of pushed content, such as content with multiple viewpoints, in-depth reports, etc. Analyze the user's behavioral data to determine the active time period of the user as the pushing opportunity. Combine the user's dialectical ability and active time period to formulate a personalized pushing strategy. According to the pushing strategy, select content suitable for the user to push. During the user's active time period, send the content to the user through the pushing platform.

[0125] Through this solution, by analyzing the user's browsing behavior, if it is found that the user has been long-term focusing on content of a single type or viewpoint, information of different types or viewpoints will be pushed to help the user break out of the information cocoon and access more diverse content. Based on the analysis of the user's behavioral data and dialectical ability, it is possible to more accurately predict which content the user is interested in, so as to push more relevant content, improve user satisfaction and engagement. Pushing content with different viewpoints encourages users to think deeply and make comparisons, which helps to enhance the user's dialectical thinking ability. According to the user's dialectical ability, personalized content more in line with the user's cognitive level and interest preferences can be provided, thus enhancing the user experience. When users receive relevant and valuable information, they are more likely to continue using the service, thereby increasing user stickiness.

[0126] In some embodiments, analyze the browsing data to determine the type of browsing content; based on the type, determine the content similarity of each piece of browsing content; according to the content similarity, determine the user's browsing attitude towards each piece of browsing content; according to the browsing attitude and type, determine whether there is a singularity in the browsing.

[0127] The browsing content can be information viewed by the user on the Internet, such as web articles, videos, pictures, social media posts, etc.

[0128] The type can be a category or label for classifying the browsing content, such as news, entertainment, technology, education, etc.

[0129] The content similarity can be the degree of similarity between any two or more pieces of browsing content browsed by the user in terms of theme, style, viewpoint, etc.

[0130] The browsing attitude can be the user's interest, preference and evaluation of the browsing content.

[0131] Specifically, data such as the user's browsing history, page dwell time, and page scroll position can be obtained through methods such as log records and third-party data providers. Classify the content in the browsing data for subsequent analysis. Use natural language processing techniques (such as text classification algorithms) to classify the browsing content and identify the field to which the content belongs (such as technology, entertainment, news, etc.). Evaluate the similarity between the user's browsing contents. Use text similarity calculation methods (such as cosine similarity, Jaccard similarity, etc.) to calculate the similarity between the browsing contents. Determine the user's browsing attitude towards each type of browsing content. Analyze the user's interest and attitude towards the content through the user's behavior on the page (such as dwell time, scroll behavior, likes, comments, etc.). Judge whether the user's browsing behavior is monotonous. Analyze the type distribution of the user's browsing content. If the user focuses on browsing a certain type of content for a long time, there may be monotony. Compare the similarity of the user's browsing contents. If the similarity is high, there may be monotony. Combine the user's browsing attitude towards different types of content. If the user shows extremely high interest in a certain type of content and low interest in other types of content, there may be monotony.

[0132] Through this solution, by analyzing the user's browsing data, we can more accurately understand the user's interest preferences, and thus push more relevant content. Identify whether the user has been concentrating on a certain type or view of content for a long time, and thus push information of different types or views to help the user access diversified content and avoid falling into an information cocoon.

[0133] In some embodiments, analyze the click behavior to determine the clicked content; analyze the clicked content to determine the attitude held by the clicked object; compare the held attitude with the browsing attitude to determine whether the content view of the clicked content is consistent with that of the browsing content; if they are consistent, based on the click behavior, determine the dwell time of the clicked content; and determine the user's dialectical ability according to the dwell time.

[0134] The clicked content can be any content that the user clicks on the Internet, such as web links, buttons, pictures, videos, articles, comments of other users, etc.

[0135] The clicked object can be the specific element that the user clicks, such as the like or dislike button, or the content of the comment, etc.

[0136] The held attitude can refer to the emotional tendency, view position, etc. expressed by the clicked object.

[0137] The browsing attitude can be the user's interest, preference, and evaluation of the browsing content.

[0138] The content view can be the view or position expressed by the browsing content.

[0139] The dwell time can be the time a user stays on a content page, such as reading time, interaction time, etc.

[0140] Specifically, log the interactive elements such as links, buttons, and pictures clicked by the user through the website or application. Use natural language processing technology to analyze the content clicked by the user, extract keywords and topic information, so as to determine the type and topic of the click object. Through content analysis, identify the emotional tendency, viewpoint stance, etc. in the clicked content. For example, use sentiment analysis tools to determine whether the content is positive, negative, or neutral, so as to determine the attitude or viewpoint held by the click object. Analyze the user's attitude when browsing previous content, for example, judge by behaviors such as liking, commenting, and sharing. Compare the attitude of the click object with the user's previous browsing attitude. Through comparative analysis, judge whether the attitude of the clicked content is consistent with the user's browsing content. For example, whether they are both positive evaluations or both negative evaluations. For the clicked content with consistent attitudes, analyze the user's dwell time, that is, the time the user stays on the clicked content page, including reading time, interaction time, etc. According to the dwell time, evaluate the user's dialectical ability. Specifically: Analyze the dwell operations of the user on the clicked content page, such as whether to conduct in-depth reading, whether to view relevant comments, etc. If the user stays on content with opposing viewpoints for a long time and shows behaviors of exploring different viewpoints, it can be considered that the user has good dialectical ability. If the user only stays on content that is consistent with their own viewpoints and rarely explores other viewpoints, it may indicate that the user has weak dialectical ability.

[0141] Through this solution, by analyzing the user's click behavior, the user's interest preferences can be understood more accurately, so as to push more relevant content. Provide personalized content recommendations according to the user's click behavior and attitude to improve the user experience. Push content with different viewpoints to encourage users to think deeply and make comparisons, which helps to improve the user's dialectical thinking ability. Identify whether the user has been concentrating on content of a certain type or viewpoint for a long time, so as to push information of different types or viewpoints to help users access diversified content and avoid being trapped in an information cocoon.

[0142] In some embodiments, within the dwell time, obtain the user's dwell operations; analyze the dwell operations to determine whether there is a situation of opposing viewpoints for the user; if so, analyze the situation of opposing viewpoints to determine the user's dialectical ability; if not, determine that the user lacks dialectical ability.

[0143] The dwell operations can be various interactive behaviors of the user within the time the user stays on a certain page or content.

[0144] The situation of opposing viewpoints can be that in the content browsed or interacted with by the user, there are different viewpoints or stances, and these viewpoints may be contradictory or controversial.

[0145] Specifically, by logging the user's stay time on different pages through a website or application, identify the operations performed by the user during the stay time, such as the user's clicks, scrolls, likes, comments, shares, etc.; as well as the frequency and duration of these operations. Use natural language processing technology to analyze the content involved in the user's stay operations, determine the type and theme of the content, and extract keywords, themes, and sentiment tendencies. Through content analysis, identify whether the user has read articles with different viewpoints under the same theme, such as the viewpoints of the positive and negative sides. Analyze whether the user has conducted in-depth reading and comparison of the content with opposing viewpoints. Observe whether the user interacts with the content with opposing viewpoints, such as posting comments, participating in discussions, etc. If the user shows behaviors of exploring, understanding, and analyzing different viewpoints on the content with opposing viewpoints, it can be considered that the user has good dialectical ability. If the user only stays on the content that is consistent with their own viewpoints and rarely explores other viewpoints, it may indicate that the user has relatively weak dialectical ability.

[0146] Through this solution, by analyzing the user's stay time on the content with opposing viewpoints, more accurate content that matches the user's interests and dialectical ability can be pushed, improving the accuracy of recommendations to achieve personalized information push based on the user's stay time data. Push diversified content that meets the user's cognitive level and dialectical ability. By pushing content with different viewpoints, help the user access and understand diversified information.

[0147] In some embodiments, analyze the search behavior to determine the user's search content; analyze the search content to determine the similarity between the search content and the browsing content; and determine the user's dialectical ability based on the similarity.

[0148] The search content can be the query keywords, phrases, or sentences entered by the user in the search engine.

[0149] The similarity can be the degree of similarity between two or more pieces of content.

[0150] Specifically, data such as the search keywords entered by the user, the search time, and the search frequency are recorded through the search engine logs. The natural language processing technology is used to analyze the search keywords to extract the keywords, themes, and user intentions corresponding to the search content. The natural language processing technology is used to analyze the text of the browsing content to extract the keywords, themes, and opinion information corresponding to the browsing content. Text similarity calculation methods (such as cosine similarity, Jaccard similarity, etc.) are used to compare the keywords, themes, and opinions corresponding to the search content with those corresponding to the browsing content. If the similarity between the user's search content and the browsing content is high, it indicates that the user may be delving deeper into a known topic, which may be a manifestation of dialectical ability. If the similarity between the user's search content and the browsing content is low, it indicates that the user may be seeking different opinions or new information, which may also be a manifestation of dialectical ability. If the user's search content is highly inconsistent with the browsing content, it indicates that the user may be avoiding exposure to different opinions, which may suggest a relatively weak dialectical ability.

[0151] Through this solution, by analyzing the user's search behavior, content that matches the user's interests and dialectical ability can be more accurately pushed, improving the accuracy of recommendations. Pushing diversified content that matches the user's cognitive level and dialectical ability can enhance the user's satisfaction and loyalty to the platform. Encouraging users to stay on content with opposing viewpoints can promote in-depth thinking and critical analysis by users, improving the user's dialectical thinking ability.

[0152] In some embodiments, the similarity is compared with a preset similarity threshold. According to the comparison result, it is determined whether the search content is consistent with the browsing content; if not, the search content is analyzed to determine the actual description of the search content; according to the actual description, it is determined whether there is an association between the search content and the browsing content; if an association is determined to exist, the association relationship between the search content and the browsing content is determined according to the actual description; and the user's dialectical ability is determined according to the association relationship.

[0153] The preset similarity threshold can be a standard value preset in the similarity analysis, used to determine whether the similarity between two contents is high enough to determine whether they belong to the same category or theme, and is stored in a preset database.

[0154] The comparison result can be a conclusion obtained after comparing the preset similarity threshold with the actually calculated similarity.

[0155] The actual description can be keywords, phrases, or sentences extracted through natural language processing technology when analyzing the search content, and these descriptions reflect the actual intentions and concerns of the user's search.

[0156] The association relationship can be the connection established between the search content and the browsing content, such as direct association, indirect association, or no association.

[0157] Specifically, text similarity calculation methods (such as cosine similarity, Jaccard similarity, etc.) are used to compare the keywords, themes, and viewpoints corresponding to the search content with those corresponding to the browsing content. Based on historical data analysis or expert experience, a similarity threshold is set to determine the consistency between the search content and the browsing content. The similarity is compared with the preset similarity threshold to determine whether the search content is consistent with the browsing content. If the similarity is higher than the threshold, it is considered that the search content is consistent with the browsing content; if the similarity is lower than the threshold, it is considered inconsistent. For inconsistent search content, its actual description is analyzed through natural language processing technology, and the keywords and phrases in the search content are extracted to understand the user's actual search intention. The keywords, themes, and viewpoints corresponding to the search content are analyzed and compared with those corresponding to the browsing content to determine whether there is a correlation. If there is a strong correlation between the user's search content and the browsing content, it may indicate that the user has good dialectical ability. If the correlation between the user's search content and the browsing content is weak or there is no correlation, it may indicate that the user has weak dialectical ability.

[0158] Through this solution, by comparing the similarity between the search content and the browsing content, the user's interests and needs can be more accurately identified, so as to provide more relevant recommended content. Through similarity analysis, the user's attitudes and cognitive levels towards different information can be evaluated, and personalized content more in line with the user's cognitive ability can be provided. Adjusting the recommendation strategy according to the user's dialectical ability can improve the user experience and make the user feel that the recommended content is more in line with their personal interests and needs. By identifying the inconsistency between the search content and the browsing content, it is possible to avoid pushing overly single or one-sided content to the user and reduce recommendation bias.

[0159] In some embodiments, according to the dialectical ability, the push type of the pushed content is determined; the operation data and browsing data of the user within a preset time period are obtained; the operation data is analyzed to determine whether the user's holding attitude towards each browsing data is consistent with the corresponding type of browsing attitude; the consistency is compared with the preset accuracy threshold. If the consistency is higher than the preset accuracy threshold, the moment when the consistency is higher than the preset accuracy threshold is determined as the push moment; according to the push type and the push moment, the push scheme is determined.

[0160] The push type can be the directional type of the information push content.

[0161] The pushed content can be the information content to be pushed.

[0162] The preset time period can be the time range for data collection and analysis within a specific time period, stored in a preset database.

[0163] The operation data can be the interaction behavior data of users on the platform, such as click data, scroll data, dwell time, sharing data, etc.

[0164] The holding attitude can be the subjective view or stance of the user towards the browsed content.

[0165] The preset accuracy threshold can be a standard value set in data analysis for judging whether the consistency of user behavior data is high enough and is stored in a preset database.

[0166] The push moment can be a determined information push timing.

[0167] Specifically, for users with stronger dialectical abilities, push content containing different viewpoints and in-depth analyses to promote deeper thinking and discussion. For users with weaker dialectical abilities, push more basic and explanatory content to help users build a more comprehensive understanding. Through the user behavior logs on the platform, collect the operation data of users' clicks, scrolls, likes, comments, etc. within a specific time period, as well as the browsing data of the pages, articles, videos, etc. browsed. Analyze the user's interaction behaviors (such as likes, comments, shares, etc.) to infer the user's interest and attitude towards the browsed content. Compare the user's interaction behaviors with the preset browsing attitude model to judge whether the user's behaviors conform to the expected attitude pattern. Determine a suitable accuracy threshold based on historical data analysis or expert experience for judging whether the consistency between the user's behaviors and the preset attitude is high enough. If the consistency of the user's behaviors is higher than the accuracy threshold, then it is considered an appropriate timing for pushing content. Select appropriate content and push timing according to the user's dialectical ability and behavior data. Design the pushed content to ensure that the content matches the user's interests and cognitive level. Select appropriate push channels and forms.

[0168] Through this solution, the pushed content is more in line with the user's cognitive level and dialectical ability, provides more in-depth and extensive information, and meets the user's personalized needs. When users receive relevant and valuable information, they are more likely to participate in interactions, such as commenting, sharing, or further exploring relevant content. By pushing different viewpoints and diverse content, it helps users break through the information cocoon and access and understand more diverse information. Providing content that matches the user's dialectical ability can promote the user's knowledge acquisition and the improvement of cognitive ability. Adjusting the recommendation strategy according to the user's dialectical ability can enhance the user experience and make the user feel that the recommended content is more in line with their personal interests and needs.

[0169] In some embodiments, according to the dialectical ability, determine the acceptability of the user for different types of content; according to the acceptability, determine the push type.

[0170] The acceptability can be the degree of acceptance of the user for a specific type of content or viewpoint.

[0171] Specifically, by analyzing the user's search behavior, browsing history, interaction data, etc., natural language processing and machine learning algorithms are used to evaluate the user's acceptance and judgment ability of different viewpoints. For users with stronger dialectical ability, they may be more likely to accept different viewpoints and challenging content. For users with weaker dialectical ability, they may be more inclined to accept content with higher consistency and familiarity. For users with a high level of critical thinking, they may show more rich, opposing, and intense viewpoints to promote in-depth thinking and discussion. For users with a low level of critical thinking, they may show more viewpoints that are closer to their interested viewpoints, as well as viewpoints for popular science or basic knowledge popularization, to help users establish a more comprehensive understanding. For users with a high level of critical thinking, content containing multiple viewpoints, in-depth analysis, and challenging topics is selected. For users with a low level of critical thinking, content related to their interests and easy to understand is selected, and more diverse viewpoints are gradually introduced.

[0172] Through this solution, the pushed content can stimulate users' thinking. Especially for users with a high level of critical thinking, their critical thinking ability can be exercised and improved by exposing them to multiple viewpoints and challenging content. For users with a low level of critical thinking, gradually introducing more diverse viewpoints and popular science content helps users broaden their knowledge and understand knowledge in different fields. In the process of users being exposed to different viewpoints, they can learn how to evaluate the credibility of information and improve their information literacy. By pushing content that is closer to the viewpoints deviating from the user's interests, it can prevent users from being trapped in an information cocoon and promote users to contact and accept different information and viewpoints. For users with a high level of critical thinking, by presenting opposing viewpoints, it can promote rational discussions among users and reduce unnecessary conflicts and misunderstandings. By pushing different viewpoints and challenging content, it can cultivate users' open-mindedness and make users more willing to accept and understand different opinions and viewpoints.

[0173] In some embodiments, the search content and browsing content are analyzed to determine the search source and browsing source; based on the search source and browsing source, the content credibility is determined; the user profile is obtained; based on the user profile and content credibility, the user's cognitive level is determined; based on the association relationship and cognitive level, the user's dialectical ability is determined.

[0174] The search source can be the starting point or platform where the user conducts a search query.

[0175] The browsing source can be the starting point or platform where the user accesses the content.

[0176] The content credibility can be the reliability of the content source and the accuracy of the information.

[0177] The user profile can be a detailed description of the user's characteristics.

[0178] The cognitive level can be the user's ability to understand and process information.

[0179] Specifically, use natural language processing technology to analyze search keywords and the themes of browsed pages, and identify the hot topics and fields that users are interested in. Record the website domains, social media accounts, news sources, etc. that users visit, and analyze the characteristics and credibility of these sources. Use a pre-established credibility scoring system or a machine learning model to evaluate the credibility of content based on factors such as the historical performance of the source, user evaluations, and professional certifications. Combine user registration information and behavioral data to create a user profile in order to more accurately understand user characteristics. Analyze the user's preference for highly credible content and the degree of understanding of complex concepts and in-depth analysis content to infer the user's cognitive level. Determine the user's dialectical ability based on the way the user processes different viewpoints and evidence, and whether the user is willing to accept and consider new information.

[0180] Through this solution, by analyzing the user's search and browsing behaviors, the user's interests and preferences can be more accurately understood. According to the user's cognitive level and dialectical ability, push content with different depths and diversities, which helps users access and understand more information and avoid the information cocoon effect. Push multiple viewpoints and challenging content to users with a high critical thinking dimension, which can promote users' in-depth thinking and critical analysis, thereby enhancing their thinking ability.

[0181] Figure 3 The structural schematic diagram of a digital information push system based on big data provided by an embodiment of the present application is as Figure 3 shown. The digital information push system 300 based on big data in this embodiment includes: a browsing analysis module 301, a behavior analysis module 302, and a solution determination module 303.

[0182] The browsing analysis module 301 is used to obtain the browsing data of the user; analyze the browsing data to determine whether there is monotony in the browsing.

[0183] The behavior analysis module 302 is used to, if there is monotony, obtain the behavior data of the user, analyze the behavior data, and determine the dialectical ability of the user.

[0184] The solution determination module 303 is used to determine a push solution based on the dialectical ability and perform information push.

[0185] Optionally, when the browsing analysis module 301 analyzes the browsing data to determine whether there is monotony in the browsing, it is used to:

[0186] Analyze the browsing data to determine the type of browsing content.

[0187] Based on the type, determine the content similarity of each piece of browsing content.

[0188] Determine the browsing attitude of the user towards each type of browsed content according to the content similarity.

[0189] Determine whether there is a singularity in browsing according to the browsing attitude and the type.

[0190] Optionally, when the behavior data includes click behavior and the behavior analysis module 302 analyzes the behavior data to determine the dialectical ability of the user, it is used for:

[0191] Analyze the click behavior to determine the clicked content.

[0192] Analyze the clicked content to determine the attitude of the clicked object.

[0193] Compare the attitude with the browsing attitude to determine whether the content view of the clicked content is consistent with that of the browsed content.

[0194] If they are consistent, determine the stay time of the clicked content based on the click behavior.

[0195] Determine the dialectical ability of the user according to the stay time.

[0196] Optionally, when the behavior analysis module 302 determines the dialectical ability of the user according to the stay time, it is used for:

[0197] Obtain the stay operation of the user during the stay time.

[0198] Analyze the stay operation to determine whether there is a situation of opposing views of the user.

[0199] If there is, analyze the situation of opposing views to determine the dialectical ability of the user.

[0200] If not, determine that the user lacks dialectical ability.

[0201] Optionally, when the behavior data includes search behavior and the behavior analysis module 302 analyzes the behavior data to determine the dialectical ability of the user, it is used for:

[0202] Analyze the search behavior to determine the search content of the user.

[0203] Analyze the search content to determine the similarity between the search content and the browsed content.

[0204] Determine the dialectical ability of the user according to the similarity.

[0205] Optionally, when the behavior analysis module 302 determines the dialectical ability of the user according to the similarity, it is used for:

[0206] Compare the similarity with a preset similarity threshold, and determine whether the search content is consistent with the browsing content according to the comparison result;

[0207] If they are inconsistent, analyze the search content to determine the actual description of the search content;

[0208] Determine whether there is an association between the search content and the browsing content according to the actual description;

[0209] If it is determined that there is an association, determine the association relationship between the search content and the browsing content according to the actual description;

[0210] Determine the dialectical ability of the user according to the association relationship.

[0211] Optionally, when the scheme determination module 303 determines the push scheme according to the dialectical ability, it is used for:

[0212] Determine the push type of the push content according to the dialectical ability;

[0213] Obtain the operation data and browsing data of the user within a preset time period;

[0214] Analyze the operation data to determine whether the holding attitude of the user towards each browsing data is consistent with the corresponding type of browsing attitude;

[0215] Compare the consistency with a preset accuracy threshold. If the consistency is higher than the preset accuracy threshold, determine the moment when the consistency is higher than the preset accuracy threshold as the push moment;

[0216] Determine the push scheme according to the push type and the push moment.

[0217] Optionally, when the scheme determination module 303 determines the push type of the push content according to the dialectical ability, it is used for:

[0218] Determine the acceptability of the user for different types of content according to the dialectical ability;

[0219] Determine the push type according to the acceptability.

[0220] Optionally, when determining the dialectical ability of the user according to the association relationship, it is used for:

[0221] Analyze the search content and the browsing content to determine the search source and the browsing source;

[0222] Determine the content credibility according to the search source and the browsing source;

[0223] Obtain a user profile; determine the cognitive level of the user according to the user profile and the content credibility;

[0224] Determine the dialectical ability of the user according to the association relationship and the cognitive level.

[0225] The system of this embodiment can be used to execute the method of any of the above embodiments, and its implementation principle and technical effects are similar, which will not be elaborated here.

Claims

1. A digital information push method based on big data, characterized in that, It includes: Obtain the browsing data of the user; Analyze the browsing data to determine whether the browsing is single; If it exists, obtain the behavior data of the user, analyze the behavior data, and determine the dialectical ability of the user; Determine the push scheme according to the dialectical ability and perform information push; The analyzing the browsing data to determine whether the browsing is single includes: Analyze the browsing data to determine the type of browsing content; Based on the type, determine the content similarity of each browsing content; According to the content similarity, determine the browsing attitude of the user towards each browsing content; According to the browsing attitude and the type, determine whether the browsing is single; The behavior data includes click behavior. The analyzing the behavior data to determine the dialectical ability of the user includes: Analyze the click behavior to determine the clicked content; Analyze the clicked content to determine the attitude held by the clicked object; Compare the held attitude with the browsing attitude to determine whether the content view of the clicked content is consistent with the browsing content; If they are consistent, based on the click behavior, determine the stay time of the clicked content; According to the stay time, determine the dialectical ability of the user; The determining the push scheme according to the dialectical ability includes: According to the dialectical ability, determine the push type of the push content; Obtain the operation data and browsing data of the user within a preset time period; Analyze the operation data to determine whether the attitude held by the user towards each browsing data is consistent with the browsing attitude of the corresponding type; Compare the consistency with a preset accuracy threshold. If the consistency is higher than the preset accuracy threshold, determine the moment when the consistency is higher than the preset accuracy threshold as the push moment; According to the push type and the push moment, determine the push scheme.

2. The method according to claim 1, characterized in that, The determining the dialectical ability of the user according to the stay time includes: Within the stay time, obtain the stay operation of the user; Analyze the stay operation to determine whether there is a situation of opposing views of the user; If it exists, analyze the situation of opposing views to determine the dialectical ability of the user; If it does not exist, determine that the user lacks dialectical ability.

3. The method according to claim 1, characterized in that, The behavior data includes search behavior. The analyzing the behavior data to determine the dialectical ability of the user includes: Analyze the search behavior to determine the search content of the user; Analyze the search content to determine the similarity between the search content and the browsing content; According to the similarity, determine the dialectical ability of the user.

4. The method according to claim 3, wherein The determining the dialectical ability of the user according to the similarity includes: Compare the similarity with a preset similarity threshold. According to the comparison result, determine whether the search content is consistent with the browsing content; If they are not consistent, analyze the search content to determine the actual description of the search content; According to the actual description, determine whether there is an association between the search content and the browsing content; If it is determined that there is an association, according to the actual description, determine the association relationship between the search content and the browsing content; Determine the dialectical ability of the user according to the association relationship.

5. The method according to claim 1, characterized in that Determine the push type of the push content according to the dialectical ability, including: Determine the acceptability of the user for different types of content according to the dialectical ability; Determine the push type according to the acceptability.

6. The method according to claim 4, wherein Determine the dialectical ability of the user according to the association relationship, including: Analyze the search content and the browsing content to determine the search source and the browsing source; Determine the content credibility according to the search source and the browsing source; Obtain the user portrait; determine the cognitive level of the user according to the user portrait and the content credibility; Determine the dialectical ability of the user according to the association relationship and the cognitive level.

7. A digital information push system based on big data, characterized in that, Applied to the method according to any one of claims 1-6, including: A browsing analysis module, configured to obtain the browsing data of the user; analyze the browsing data to determine whether there is singularity in the browsing; A behavior analysis module, configured to, if so, obtain the behavior data of the user, analyze the behavior data, and determine the dialectical ability of the user; A solution determination module, configured to determine a push solution and perform information push according to the dialectical ability.

Images