Intelligent campus information issuing system

Abstract

The invention discloses a campus intelligent information issuing system, which belongs to the technical field of information pushing and comprises an information acquisition module, an information processing module, an information pushing module and a user terminal module. The information acquisition module is used for acquiring user information, and the user information at least comprises user basic data and user behavior data; the information processing module is used for constructing a user portrait by using a knowledge graph mining algorithm based on the user basic data and the user behavior data; the information pushing module is used for determining an information pushing object according to the type of to-be-pushed information and a user portrait, and pushing the to-be-pushed information to the information pushing object based on edge calculation; and the information pushing object receives the information to be pushed by using the user terminal module. The problem of poor pushing accuracy caused by lack of directional pushing functions for different user groups is solved. And the pushing efficiency is remarkably improved while the pushing accuracy and safety are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of information push, and in particular to an intelligent campus information publishing system. Background Art

[0002] With the rapid development of information technology, informatization has become the mainstream trend of social development. In order to ensure the rapid and effective dissemination of information, it is an inevitable trend to transmit information through informatization. The prior art, such as a campus information push method device and system with patent number CN108337329A, wherein the campus information push method includes: step S1: the user registers in the system; step S2: the administrator registers in the system; step S3: the administrator sends information to the registered user through the system; step S4: the system feeds back the message status to the administrator. The above scheme only needs to establish the big data of the integration of school, students and enterprises through simple registration by the user, and makes full use of the big data to realize the accurate push of messages through the system platform. However, when the above scheme sends the message to the system user, the corresponding message is not pushed to the corresponding system user in a targeted manner. Therefore, the above scheme lacks the directional push function for different user groups, and the push accuracy is low. Summary of the invention

[0003] In view of the problem that the prior art lacks the directional push function for different user groups, resulting in poor push accuracy, the present invention provides a campus intelligent information publishing system, which uses the association analysis capability of the knowledge graph mining algorithm to perform association analysis on user data to build a user portrait, and then determines the message push object according to the user portrait and the type of message to be pushed, thereby reducing the load pressure of edge computing to push data to a large number of users, and pushes the information to be pushed to the information push object through the privacy protection and secure transmission characteristics of edge computing, thereby reducing the exposure risk of user portraits and information to be pushed in network transmission, and solving the problem of poor push accuracy due to the lack of directional push function for different user groups. While significantly improving the push accuracy and security, it also significantly improves the push efficiency.

[0004] In order to solve the above technical problems, the present invention provides a campus intelligent information release system, including an information collection module, an information processing module, an information push module, and a user terminal module; The information collection module is used to collect user information, and the user information includes at least user basic data and user behavior data; the information processing module is used to construct a user portrait based on the user basic data and user behavior data using a knowledge graph mining algorithm; The information push module is used to determine the information push object according to the type of information to be pushed and the user portrait, and push the information to be pushed to the information push object based on edge computing; The information push object uses the user terminal module to receive the information to be pushed.

[0005] After adopting the above technical solution, the present invention has the following advantages: Considering that the existing technology does not perform feature analysis on users to obtain user portraits, it is difficult to achieve the targeted push function, and the knowledge graph mining algorithm has the ability to perform association analysis and can analyze user data. Therefore, the user portrait is obtained through the algorithm, and then the accurate push object is obtained, which reduces the load pressure of edge computing to push the information to be pushed to a large number of users. In the process of pushing, in order to avoid the outside world predicting the user portrait through the type of information to be pushed, thereby causing the risk of user information leakage, and edge computing has the characteristics of privacy protection and secure transmission, so the information to be pushed is pushed to the information push object through edge computing. Through the combination of knowledge graph mining algorithm and edge computing, the push accuracy and security are significantly improved while the push efficiency is significantly improved; The problem of poor push accuracy due to the lack of targeted push function for different user groups has been solved.

[0006] Preferably, the method of constructing a user portrait based on the user basic data and the user behavior data using a knowledge graph mining algorithm includes: Integrate user basic data and user behavior data based on user identifier to obtain multi-dimensional data of the user; Performing data cleaning, outlier and missing value processing on the multidimensional data to obtain clean data of the user; Natural language processing is used to extract the relationships between entities from the clean data, and a user profile is constructed based on the relationships between entities.

[0007] In this solution, by integrating user data to obtain multidimensional data, and then forming a multidimensional data view of the user, it can more comprehensively reflect the user's characteristics and preferences, thereby ensuring the richness and accuracy of the user portrait. Natural language processing can identify and divide text, so through natural language processing, entities and relationships can be extracted from clean data. At the same time, extraction can be done directly from clean data. It only needs to divide the complex text in the clean data, and simple text can be directly extracted as entities, thereby improving the extraction efficiency.

[0008] Preferably, the extracting the relationships between entities from the clean data using natural language processing includes: Preset a first relational comparison set based on the mutual relationship between the attributes of each page data in various campus systems, use natural language processing to divide the page data that exceeds the preset conditions in each page data, and obtain a second relational comparison set; Performing a union operation on the first relation comparison set and the second relation comparison set to obtain a process relation comparison set; Filter invalid relationship items in the process relationship comparison set based on business logic, obtain semantic similarity of relationship items in the process relationship comparison set according to a semantic similarity algorithm, and merge the relationship items according to the semantic similarity to obtain a final relationship comparison set; The relationships between entities are extracted from the clean data based on the final relationship comparison set.

[0009] In this solution, by presetting the first relationship comparison set of the relationships between the attributes of the page data of various systems in the campus, the comprehensiveness of the first relationship comparison set is ensured, and then the comprehensiveness of the relationships between the extracted entities is ensured. The page data exceeding the preset conditions is divided through natural language processing to avoid the loss of the relationships between entities in the complex data in the page data, and further ensure the comprehensiveness of the relationships between the extracted entities. By removing invalid relationship items, union operations and obtaining semantic similarity, the redundancy of the extraction process and the extracted entities and their relationships is reduced, the extraction efficiency is further improved, and at the same time, the quality of the final relationship comparison set is also improved.

[0010] Preferably, the constructing of a user portrait based on the relationship between entities includes: An initial user portrait is constructed based on the categories to which the entities belong and the relationships between the entities, the occurrence frequencies corresponding to the relationships between the entities are obtained based on the clean data, and the initial user portrait is optimized based on the occurrence frequencies to obtain a user portrait. By optimizing the initial user portrait in combination with the occurrence frequencies, the real needs and characteristics of the user can be better reflected, further improving the accuracy and comprehensiveness of the user portrait.

[0011] Preferably, determining the information push object according to the type of information to be pushed and the user portrait includes: Several types of information to be pushed are obtained from multiple dimensions, and the several types of user portraits are compared with the several types of information to be pushed. Among the several types of user portraits, there are a preset number of types that are successfully compared with the several types of information to be pushed. The user corresponding to the user portrait is the information push object.

[0012] Preferably, the step of pushing the information to be pushed to the information push object based on edge computing includes: The data packet carrying the information to be pushed and the IP address of the information push object is encrypted using an encryption algorithm to obtain encrypted data, and the encrypted data is transmitted to the information push object through a wireless network.

[0013] Preferably, it also includes a feedback analysis module, which is used to collect statistics on the reading situation of the information push object and adjust the user portrait according to the reading situation.

[0014] Preferably, the information push module includes a content review module; The content review module is used to review the information to be pushed, and when the review fails, a reminder message is generated and sent to the management personnel.

[0015] Preferably, the content review module is also used to prompt the target confirmation module in the information push module to determine the information push object according to the type of information to be pushed and the user portrait when the review is passed, and push the information to be pushed to the information push object based on edge computing.

[0016] Preferably, the user terminal module includes an attendance module and a display module; The attendance module is used to record the user's training attendance in real time, and the display module is used to display the user's training attendance.

[0017] Beneficial effects of this program: Considering that the existing technology does not perform feature analysis on users to obtain user portraits, it is difficult to achieve the targeted push function, and the knowledge graph mining algorithm has the ability to perform association analysis and can analyze user data. Therefore, the user portrait is obtained through the algorithm, and then the accurate push object is obtained, which reduces the load pressure of edge computing to push the information to be pushed to a large number of users. In the process of pushing, in order to avoid the outside world predicting the user portrait through the type of information to be pushed, thereby causing the risk of user information leakage, and edge computing has the characteristics of privacy protection and secure transmission, so the information to be pushed is pushed to the information push object through edge computing. Through the combination of knowledge graph mining algorithm and edge computing, the push accuracy and security are significantly improved while the push efficiency is significantly improved; By presetting the first relationship comparison set of the mutual relationship between the attributes of the page data of various systems in the campus, the comprehensiveness of the first relationship comparison set is ensured, and then the comprehensiveness of the mutual relationship between the extracted entities is ensured. The page data exceeding the preset conditions is divided by natural language processing, which avoids the loss of the mutual relationship between entities in the complex data in the page data, and further ensures the comprehensiveness of the mutual relationship between the extracted entities. By removing invalid relationship items, union operations and obtaining semantic similarity, the redundancy of the extraction process and the extracted entities and mutual relationships is reduced, and the extraction efficiency is further improved. At the same time, the quality of the final relationship comparison set is also improved, thereby improving the accuracy and comprehensiveness of the user portrait, and further improving the push accuracy; The problem of poor push accuracy due to the lack of targeted push function for different user groups has been solved. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Other features, objects and advantages of the present invention will become more apparent by reading the detailed description of non-limiting embodiments made with reference to the following drawings. The drawings are only for the purpose of illustrating preferred embodiments and are not to be considered as limiting the present invention. Also, the same reference symbols are used throughout the drawings to represent the same parts.

[0019] Figure 1 This is a schematic diagram of the structure of a campus intelligent information publishing system of the present invention; Figure 2 The present invention is a flowchart of information push in a campus intelligent information publishing system. DETAILED DESCRIPTION

[0020] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific implementation method described herein is only an optimal embodiment of the present invention, which is only used to explain the present invention and does not limit the scope of protection of the present invention. All other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0021] Before discussing the exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flow charts. Although the flow charts describe the operations (or steps) as sequential processes, many of the operations (or steps) therein can be implemented in parallel, concurrently, or simultaneously. In addition, the order of the operations can be rearranged. The process can be terminated when its operation is completed, but can also have additional steps not included in the drawings; the process can correspond to a method, function, procedure, subroutine, subprogram, etc.

[0022] Embodiment 1: like Figure 1 As shown, a campus intelligent information release system includes an information collection module, an information processing module, an information push module, and a user terminal module; The information collection module is used to collect user information, and the user information at least includes user basic data and user behavior data.

[0023] In this embodiment, the basic data of the user, i.e., identity information, contact information and basic attribute information, is registered in the publishing system. When the information collection module of the publishing system obtains the user behavior data, the basic data of the user is first extracted from the publishing system, and according to the user unique identifier in the basic data of the user, the data related to the user unique identifier in each system of the campus is obtained through multiple interfaces in the information collection module. This part of data is the user's behavior data. By collecting user behavior data from multiple systems, the integrity of the user data is ensured, and the user's preferences, needs and habits can be more accurately reflected, which improves the accuracy of the user portrait constructed subsequently.

[0024] The information processing module is used to construct a user portrait based on user basic data and user behavior data using a knowledge graph mining algorithm.

[0025] The method of constructing a user profile based on basic user data and user behavior data using a knowledge graph mining algorithm includes: Integrate user basic data and user behavior data based on user identifier to obtain multi-dimensional data of the user; Performing data cleaning, outlier and missing value processing on the multidimensional data to obtain clean data of the user; Natural language processing is used to extract the relationships between entities from the clean data, and a user profile is constructed based on the relationships between entities.

[0026] The extracting the relationship between entities from the clean data using natural language processing includes: presetting a first relationship comparison set based on the relationship between the attributes of each page data in various campus systems, dividing the page data exceeding the preset condition in each page data using natural language processing, and obtaining a second relationship comparison set; Performing a union operation on the first relation comparison set and the second relation comparison set to obtain a process relation comparison set; Filter invalid relationship items in the process relationship comparison set based on business logic, obtain semantic similarity of relationship items in the process relationship comparison set according to a semantic similarity algorithm, and merge the relationship items according to the semantic similarity to obtain a final relationship comparison set; The relationships between entities are extracted from the clean data based on the final relationship comparison set.

[0027] The constructing of a user profile based on the relationship between entities includes: An initial user portrait is constructed based on the categories to which the entities belong and the relationships between the entities, the occurrence frequencies corresponding to the relationships between the entities are obtained based on the clean data, and the initial user portrait is optimized based on the occurrence frequencies to obtain a user portrait.

[0028] In this embodiment, various campus systems include course management systems, faculty information systems, student information systems, and library management systems. The specific relationship between the attributes of the data on each page is as follows: if the attributes of the content recorded on the user record page in the course management system are: name, course, and course score, if the name is XX, the course is computer network, and the course score is 90, it can be represented as XX studying computer network, and XX studying computer network The course score is 90, then the relationship between the entity name and the entity course is "learning", and the relationship between the entity computer network and the entity score 90 is "course score". A set similar to the above-mentioned relationships is the first relationship comparison set. At the same time, the correspondence between the above-mentioned entities and relationships is only part of the case in this embodiment. The entities and relationships The corresponding situation can be adjusted accordingly according to the actual situation. By obtaining the mutual relationship of attributes in each page data in each system and presetting the first relationship comparison set, the complexity of obtaining the first relationship comparison set through natural language processing is reduced. At the same time, the page data exceeding the preset conditions, that is, the text data exceeding a certain size, is divided through natural language processing to obtain the second relationship comparison set, which avoids the loss of the mutual relationship between entities in the large text data in the page data, ensures the comprehensiveness of the mutual relationship between the extracted entities, and uses natural language processing to divide the page data exceeding the preset conditions in each page data, and obtains the second relationship comparison set. Specifically, the data in the large text data is divided according to the part of speech, and the part of speech is noun, verb, adjective, etc., according to the above.

[0029] In this embodiment, the first relationship comparison set and the second relationship comparison set are combined to obtain the process relationship comparison set: if both the first relationship comparison set and the second relationship comparison set contain the course score of XX studying computer network with a score of 90, then the course score of XX studying computer network with a score of 90 is deleted from either the first relationship comparison set or the second relationship comparison set. Invalid relationship items are relationship items that are not established in business logic, such as XX works in computer network. At this time, according to business logic, the relationship between XX and computer network cannot be "works in", so this type of relationship item is classified as invalid and filtered out from the process relationship comparison set. Since there is basically no context information in the process relationship comparison set, the semantic dictionary method is used to calculate the semantic similarity of the relationship items. The similarity threshold is preset according to user needs. When the similarity between two relationship items is greater than the preset similarity threshold, the two relationship items are merged into one relationship item. At the same time, if there are specific quantifiers such as frequency and score in the two relationship items, the frequency and score are merged. For example, XX's course score for computer network is 90, and XX's course score for computer network is 80. After merging, XX's course score for computer network is 170 or XX's course score for computer network is 170. By removing invalid relationship items, performing union operations and obtaining semantic similarity, the redundancy of the extraction process and the extracted entities and relationships is reduced, and the extraction efficiency is further improved. At the same time, the quality of the final relationship comparison set is also improved, thereby improving the accuracy and comprehensiveness of the user portrait and further improving the push accuracy.

[0030] In this embodiment, the category to which the entity belongs is flexibly set according to user needs. If the entity corresponding to the user is computer network, operating system, advanced mathematics, probability theory and circuit design, then the category to which the entity computer and operating system belong is the computer category, the category to which the entity advanced mathematics and probability theory belong is the mathematics category, and the category to which the entity circuit design belongs is the electrical category. Then the user belongs to the computer category, the mathematics category and the electrical category. At the same time, if the user's frequency of learning computer network is 2 and the frequency of learning operating system is 3, the frequency of the computer category is 5. Similarly, the frequencies of the mathematics category and the electrical category can be obtained. The initial user portrait is optimized through the frequency, so that the user's habits and characteristics are clearly represented through the user portrait, which makes it easier to determine the information push object based on the user portrait, and at the same time improves the comprehensiveness of the user portrait.

[0031] The information push module is used to determine the information push object according to the type of information to be pushed and the user portrait, and push the information to be pushed to the information push object based on edge computing.

[0032] Determining the information push object according to the type of information to be pushed and the user portrait includes: Several types of information to be pushed are obtained from multiple dimensions, and the several types of user portraits are compared with the several types of information to be pushed. Among the several types of user portraits, at least two types are successfully compared with the several types of information to be pushed. The user corresponding to the user portrait is the object of information push.

[0033] The pushing of the information to be pushed to the information push object based on edge computing includes: The data packet carrying the information to be pushed and the IP address of the information push object is encrypted using an encryption algorithm to obtain encrypted data, and the encrypted data is transmitted to the information push object through a wireless network.

[0034] The information push module includes a content review module; The content review module is used to review the information to be pushed, and when the review fails, a reminder message is generated and sent to the management personnel.

[0035] The content review module is also used to prompt the target confirmation module in the information push module to determine the information push object according to the type of information to be pushed and the user portrait when the review is passed, and push the information to be pushed to the information push object based on edge computing.

[0036] In this embodiment, if Figure 2 As shown, the manager edits the information to be pushed through the information collection module. After the editing is completed, the edited information is stored in the publishing system so that the information to be pushed can be pushed in time through the publishing system. At the same time, in addition to pushing the information to be pushed through the user portrait, the object to be pushed can also be directly preset in the publishing system to meet the special needs of information push, thereby improving the flexibility of the publishing system. The major category of information type corresponds to the category to which the entity belongs. There is a corresponding subcategory under the corresponding major category, and the subcategory is the corresponding entity. The type of user portrait is consistent with the type of information. The preset number is set according to user needs. If the push needs to be more accurate, the preset number is set accordingly. By flexibly setting the preset number, the flexibility of information push is improved.

[0037] As an implementation method, the preset number can also be preset according to the frequency of occurrence corresponding to the relationship between entities. For example, if the user's category is computer category, and the frequency of computer category is 20, and the information category is computer category, mathematics category and electrical category, then because the frequency of computer category is relatively large, the user is also the information push object of the information, thereby further improving the flexibility and accuracy of information push.

[0038] The information push object uses the user terminal module to receive the information to be pushed.

[0039] The user terminal module includes an attendance module and a display module; The attendance module is used to record the user's training attendance in real time, and the display module is used to display the user's training attendance.

[0040] In this embodiment, if Figure 2 As shown in the figure, the user terminal module is specifically an electronic class card. The attendance module in the user terminal module can record the user's training attendance in real time, avoiding errors and omissions that may occur in manual recording. At the same time, the display module can intuitively display these attendance data to users and managers, improving the transparency of attendance. Automated and transparent attendance management helps to improve training efficiency. At the same time, users can clearly understand their training attendance through the display module, which is convenient for improving user experience and participation.

[0041] It also includes a feedback analysis module, which is used to count the reading situation of the information push object and adjust the user portrait according to the reading situation.

[0042] In this embodiment, the user portrait is adjusted according to the reading situation. Specifically, after the information to be pushed is pushed to the corresponding push object, the frequency of the user's category is modified according to whether the object reads the information to be pushed. The preset number mentioned above can also be preset according to the occurrence frequency corresponding to the relationship between entities. Therefore, the frequency is modified to improve the accuracy and adaptability of information push. At the same time, the feedback analysis module can also be used to timely adjust the content of the information to be pushed according to the reading situation, thereby improving user satisfaction.

[0043] The specific implementation described above is a preferred implementation of a campus intelligent information publishing system of the present invention, and is not intended to limit the specific implementation scope of the present invention. The scope of the present invention includes but is not limited to this specific implementation. All equivalent changes made in accordance with the shape and structure of the present invention are within the protection scope of the present invention.

Claims

1. A campus intelligent information publishing system, characterized in that: It includes information collection module, information processing module, information push module and user terminal module; The information collection module is used to collect user information, and the user information includes at least user basic data and user behavior data; the information processing module is used to construct a user portrait based on the user basic data and user behavior data using a knowledge graph mining algorithm; The information push module is used to determine the information push object according to the type of information to be pushed and the user portrait, and push the information to be pushed to the information push object based on edge computing; The information push object uses the user terminal module to receive the information to be pushed.

2. According to claim 1, a campus intelligent information release system is characterized in that: The method of constructing a user profile based on basic user data and user behavior data using a knowledge graph mining algorithm includes: Integrate user basic data and user behavior data based on user identifier to obtain multi-dimensional data of the user; Performing data cleaning, outlier and missing value processing on the multidimensional data to obtain clean data of the user; Natural language processing is used to extract the relationships between entities from the clean data, and a user profile is constructed based on the relationships between entities.

3. A campus intelligent information release system according to claim 2, characterized in that: The extracting the relationship between entities from the clean data using natural language processing includes: Preset a first relational comparison set based on the mutual relationship between the attributes of each page data in various campus systems, use natural language processing to divide the page data that exceeds the preset conditions in each page data, and obtain a second relational comparison set; Performing a union operation on the first relation comparison set and the second relation comparison set to obtain a process relation comparison set; Filter invalid relationship items in the process relationship comparison set based on business logic, obtain semantic similarity of relationship items in the process relationship comparison set according to a semantic similarity algorithm, and merge the relationship items according to the semantic similarity to obtain a final relationship comparison set; The relationships between entities are extracted from the clean data based on the final relationship comparison set.

4. A campus intelligent information release system according to claim 1 or 3, characterized in that: The constructing of a user profile based on the relationship between entities includes: An initial user portrait is constructed based on the categories to which the entities belong and the relationships between the entities, the occurrence frequencies corresponding to the relationships between the entities are obtained based on the clean data, and the initial user portrait is optimized based on the occurrence frequencies to obtain a user portrait.

5. According to claim 1, a campus intelligent information release system is characterized in that: Determining the information push object according to the type of information to be pushed and the user portrait includes: Several types of information to be pushed are obtained from multiple dimensions, and the several types of user portraits are compared with the several types of information to be pushed. Among the several types of user portraits, there are a preset number of types that are successfully compared with the several types of information to be pushed. The user corresponding to the user portrait is the information push object.

6. A campus intelligent information release system according to claim 1, characterized in that: The pushing of the information to be pushed to the information push object based on edge computing includes: The data packet carrying the information to be pushed and the IP address of the information push object is encrypted using an encryption algorithm to obtain encrypted data, and the encrypted data is transmitted to the information push object through a wireless network.

7. A campus intelligent information release system according to claim 1, characterized in that: It also includes a feedback analysis module, which is used to count the reading situation of the information push object and adjust the user portrait according to the reading situation.

8. The campus intelligent information release system according to claim 1 is characterized in that: The information push module includes a content review module; The content review module is used to review the information to be pushed, and when the review fails, a reminder message is generated and sent to the management personnel.

9. A campus intelligent information release system according to claim 8, characterized in that: The content review module is also used to prompt the target confirmation module in the information push module to determine the information push object according to the type of information to be pushed and the user portrait when the review is passed, and push the information to be pushed to the information push object based on edge computing.

10. The campus intelligent information release system according to claim 1, characterized in that: The user terminal module includes an attendance module and a display module; The attendance module is used to record the user's training attendance in real time, and the display module is used to display the user's training attendance.

Images