Informatization platform management method and system

Abstract

The invention relates to the technical field of informatization platform management, and particularly discloses an informatization platform management method and system, and the system comprises a front-end module and a rear-end management module which are in communication connection. Wherein the back-end management module comprises an information storage unit used for storing student information, a learning experience report, a conference record and a user message; the data backup unit is used for backing up data regularly; the intelligent semantic understanding unit is used for carrying out semantic understanding on the learning experience report to obtain an emotion analysis result; and the early warning unit is used for generating an early warning prompt in response to the fact that the sentiment analysis result is negative. According to the application, through a digital means, storage, backup, analysis and early warning are carried out on uploaded learning experience, released conference records and received user messages of students, so that existing problems and deficiencies can be found in time according to learning communication and feedback, and powerful basic data support is provided for construction of organizations, teams and the like; and thus, the improvement of work is facilitated.

Description

Technical Field

[0001] The present application relates to the technical field of information platform management, and more specifically, to an information platform management method and system. Background Art

[0002] With the development of information technology, digital transformation has become a trend in all walks of life, and the construction of organizations and teams is no exception. The construction of traditional organizations and teams relies on paper documents, manual statistical analysis, and decentralized management models. This method is inefficient and prone to errors, and it is difficult to adapt to the modern needs of efficient and fast information exchange. Especially in the context of new technologies such as big data, cloud computing, and artificial intelligence, how to use advanced information technology to improve the efficiency of organization and team construction, enhance the timeliness of communication and feedback, and provide basic support for the decision-making of organizations and teams has become an urgent problem to be solved. Therefore, an information platform management system is needed to specifically solve the above technical problems. Summary of the invention

[0003] The present application provides an information platform management method and system. The front end of the information platform management system supports students to upload learning experiences, publish meeting minutes, and receive user messages, etc., and uses the back end of the information platform management system to store, back up, analyze and warn these data through digital means, so as to timely discover existing problems and deficiencies based on these learning exchanges and feedback, provide strong basic data support for the decision-making of organizations, teams, etc., and thus help promote work improvements.

[0004] In a first aspect, an information platform management system is provided, comprising: A front-end module and a back-end management module connected by communication; The front-end module includes a learning experience submission unit, a meeting record publishing unit and a message unit; The backend management module includes an information storage unit, a data backup unit, an intelligent semantic understanding unit and an early warning unit; The intelligent semantic understanding unit comprises: A learning experience report extraction subunit, used to extract a set of learning experience reports from the information storage unit; A learning experience report text description extraction subunit, used for extracting a text portion from each learning experience report in the set of learning experience reports to obtain a set of learning experience report text descriptions; A random binary processing subunit, used for randomly binary-dividing the set of learning experience report text descriptions to obtain a first set of learning experience report text descriptions and a second set of learning experience report text descriptions; A text description semantic understanding subunit, used for performing semantic understanding on the first set of the learning experience report text descriptions and the second set of the learning experience report text descriptions to obtain a first set of learning experience report semantic coding feature vectors and a second set of learning experience report semantic coding feature vectors; The feature distribution field semantic dynamic aggregation subunit is used to perform feature distribution field semantic dynamic aggregation on the first set of semantic encoding feature vectors of the learning experience report and the second set of semantic encoding feature vectors of the learning experience report, respectively, to obtain the semantic aggregation representation vector of the first set of learning experience reports and the semantic aggregation representation vector of the second set of learning experience reports.

[0005] In a possible implementation, the learning experience submission unit is used to receive learning experience reports submitted by users, the meeting record publishing unit is used to publish meeting records, and the message unit is used to receive messages from users; The information storage unit is used to store student information, learning experience reports, meeting minutes and user messages, and the data backup unit is used to back up data regularly; the intelligent semantic understanding unit is used to perform semantic understanding of the learning experience report to obtain a sentiment analysis result; the early warning unit is used to generate a warning prompt in response to the sentiment analysis result being negative.

[0006] In a possible implementation, the intelligent semantic understanding unit further includes: An emotion adversarial semantic representation generating subunit, used for inputting the first set of learning information report semantic aggregation representation vectors and the second set of learning information report semantic aggregation representation vectors into an emotional theme adversarial network to obtain an emotion adversarial semantic representation of a learning experience report; The sentiment analysis result generating subunit is used to generate the sentiment analysis result based on the learning experience report sentiment adversarial semantic representation.

[0007] In a possible implementation, the text description semantic understanding subunit is used to: Performing semantic coding on each learning experience report text description in the first set of learning experience report text descriptions to obtain a first set of learning experience report semantic coding feature vectors; Semantic encoding is performed on each learning experience report text description in the second set of learning experience report text descriptions to obtain a second set of learning experience report semantic encoding feature vectors.

[0008] In a possible implementation, the feature distribution field semantic dynamic aggregation subunit includes: A static energy factor calculation secondary subunit is used to calculate the static energy factor of each learning experience report semantic coding feature vector in the first set of learning experience report semantic coding feature vectors to obtain a set of learning experience report semantic static energy factors; The feature significant aggregation processing secondary sub-unit is used to perform feature significant aggregation based on the imitation field on the first set of semantic encoding feature vectors of the learning experience report based on the set of semantic static energy factors of the learning experience report to obtain the semantic aggregation representation vector of the first set of learning experience reports.

[0009] In a possible implementation, the static energy factor calculation secondary subunit is used to: Calculating the mean and variance of the semantic encoding feature vector of the learning experience report to obtain the semantic mean of the learning experience report and the semantic variance of the learning experience report; Subtracting the learning experience report semantic coding feature vector from the learning experience report semantic mean by position, and calculating the fourth power of each position of the feature vector after the subtraction to obtain the learning experience report semantic difference modulation vector; Calculating the expected value of the semantic difference modulation vector of the learning experience report to obtain the semantic expected value of the learning experience report; The semantic expected value of the learning experience report is divided by the square of the semantic variance of the learning experience report, and the value obtained by the division is input into the sigmoid function to obtain the semantic static energy factor of the learning experience report.

[0010] In a possible implementation, the feature significant aggregation processing secondary subunit includes: The clustering initial center vector selects a three-level sub-unit, which is used to select the learning experience report semantic encoding feature vector corresponding to the largest learning experience report semantic static energy factor in the set of learning experience report semantic static energy factors as the learning experience report semantic clustering initial center vector; The third-level sub-unit for dynamic aggregation energy factor calculation is used to calculate the dynamic aggregation energy factor of each learning experience report semantic encoding feature vector in the set of the learning experience report semantic encoding feature vectors based on the spatial span between each learning experience report semantic encoding feature vector in the first set of the learning experience report semantic encoding feature vectors and the initial center vector of the learning experience report semantic clustering, as well as the static energy factor of each learning experience report semantic encoding feature vector and the static energy factor of the initial center vector of the learning experience report semantic clustering, so as to obtain a set of learning experience report semantic dynamic aggregation energy factors; The third-level subunit for calculating the dynamic aggregation weight factor is used to input the set of the learning experience report semantic dynamic aggregation energy factors into the gated mask unit to obtain the set of learning experience report semantic dynamic aggregation weight factors; The third-level sub-unit for set semantic aggregation processing is used to use the set of semantic dynamic aggregation weight factors of the learning experience report as weighted weights, calculate the weighted sum of the first set of semantic encoding feature vectors of the learning experience report to obtain the first set semantic aggregation representation vector of the learning experience report.

[0011] In a possible implementation, the dynamic aggregate energy factor calculation three-level subunit is used to: Multiplying the static energy factor of the learning experience report semantic encoding feature vector and the static energy factor of the learning experience report semantic clustering initial center vector by the first weighting parameter to obtain the first learning experience report semantic dynamic aggregation energy factor; Multiplying the square of the spatial span between the learning experience report semantic encoding feature vector and the learning experience report semantic clustering initial center vector by a second weighting parameter to obtain a second learning experience report semantic dynamic aggregation energy factor; The first learning experience report semantic dynamic aggregation energy factor is divided by the second learning experience report semantic dynamic aggregation energy factor to obtain the learning experience report semantic dynamic aggregation energy factor.

[0012] In a possible implementation, the emotion confrontation semantic representation generating subunit is used to: calculate the position difference between the first set semantic aggregation representation vector of the learning information report and the second set semantic aggregation representation vector of the learning information report to obtain the learning experience report emotion confrontation semantic representation vector as the learning experience report emotion confrontation semantic representation; The sentiment analysis result generating subunit is used to: input the learning experience report sentiment adversarial semantic representation vector into a classifier-based sentiment identifier to obtain the sentiment analysis result.

[0013] In a second aspect, an information platform management method is provided, comprising: Extracting a collection of learning experience reports from the information storage unit; Extracting a text portion from each learning experience report in the set of learning experience reports to obtain a set of learning experience report text descriptions; Randomly dividing the set of learning experience report text descriptions into two sets to obtain a first set of learning experience report text descriptions and a second set of learning experience report text descriptions; Performing semantic understanding on the first set of the learning experience report text descriptions and the second set of the learning experience report text descriptions to obtain a first set of learning experience report semantic coding feature vectors and a second set of learning experience report semantic coding feature vectors; Performing feature distribution field semantic dynamic aggregation on the first set of semantically encoded feature vectors of the learning experience report and the second set of semantically encoded feature vectors of the learning experience report to obtain a first set of semantically aggregated representation vectors of the learning experience report and a second set of semantically aggregated representation vectors of the learning experience report; Inputting the first set of semantic aggregation representation vectors of the learning information report and the second set of semantic aggregation representation vectors of the learning information report into an emotional theme adversarial network to obtain emotional adversarial semantic representations of learning experience reports; The sentiment analysis result is generated based on the emotion adversarial semantic representation of the learning experience report.

[0014] The present application has at least the following technical effects: the present application randomly splits a plurality of learning experience report text descriptions into two parts, each of which contains a specific number of learning experience report text descriptions, and then introduces data processing and semantic understanding algorithms based on artificial intelligence and natural language processing technology in the back end to respectively perform semantic dynamic aggregation analysis on the sets of two learning experience report text descriptions, so as to respectively capture the semantic aggregation representations of the two learning experience report sets, and use the characteristic differences between the semantic aggregation representations of the two learning experience reports with emotions as the theme to analyze the emotional tendency of the learning experience report, so as to identify positive, negative or neutral feedback in the learning experience report. The emotional tendency in the learning experience report can be automatically identified by game learning. If the emotional tendency is found to be negative, the corresponding early warning prompt can be triggered by the early warning unit to remind the managers of the organization, team, etc. to pay attention and take measures to solve the problem. This intelligent information platform management system and instant feedback mechanism can greatly shorten the time interval from problem discovery to solution, thereby providing strong basic data support for the decision-making of organizations, teams, etc., and thus helping to promote the improvement of work. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings of the embodiments of the present application are briefly introduced below. Obviously, the drawings described below only relate to some embodiments of the present application, and are not intended to limit the present application.

[0016] Figure 1 This is a schematic block diagram of an information platform management system according to an embodiment of the present application.

[0017] Figure 2 This is a schematic block diagram of an intelligent semantic understanding unit in an information platform management system according to an embodiment of the present application.

[0018] Figure 3 This is a schematic diagram of the data flow of the intelligent semantic understanding unit in the information platform management system of an embodiment of the present application.

[0019] Figure 4 This is a schematic flowchart of the information platform management method according to an embodiment of the present application. DETAILED DESCRIPTION

[0020] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work also fall within the scope of protection of the present application.

[0021] It should be noted that the acquisition and processing of all information or data in this application are carried out in compliance with the relevant national data protection laws and policies and with the authorization given by the owner of the corresponding device.

[0022] With the advancement of science and technology, digital transformation has become a key trend to promote the development of all walks of life. Organization and team building activities are also facing the need for this change. Traditional organization and team building work usually relies on the delivery of paper documents, manual data statistics and analysis, and relatively decentralized management methods. This model is not only inefficient, but also prone to errors. It can no longer meet the needs of today's society for efficient and real-time information exchange. Especially in the current era of widespread application of cutting-edge technologies such as big data, cloud computing, and artificial intelligence, exploring how to use these advanced information technology tools to optimize the process of organization and team building work, speed up the exchange of information, and provide more accurate and timely data support to assist in the construction decision-making of organizations and teams has become an urgent problem to be solved.

[0023] In response to the above technical problems, in the technical solution of this application, an information platform management system is proposed. Figure 1It is a schematic block diagram of the information platform management system of the embodiment of the present application. The information platform management system 10 includes: a front-end module 100 and a back-end management module 200 connected in communication; the front-end module 100 includes a learning experience submission unit 110, a meeting record publishing unit 120 and a message unit 130, wherein the learning experience submission unit is used to receive a learning experience report submitted by a user, the meeting record publishing unit is used to publish meeting records, and the message unit is used to receive user messages; the back-end management module 20 includes: an information storage unit 210, a data backup unit 220, an intelligent semantic understanding unit 230 and an early warning unit 240, wherein the information storage unit is used to store student information, learning experience reports, meeting records and user messages, and the data backup unit is used to back up data regularly; the intelligent semantic understanding unit is used to perform semantic understanding on the learning experience report to obtain a sentiment analysis result; the early warning unit is used to generate an early warning prompt in response to the sentiment analysis result being negative. It should be understood that the front-end module is mainly responsible for data acquisition and page display, while the back-end management module is responsible for data storage and processing. The front end of the information platform management system supports students to upload learning experiences, publish meeting minutes, and receive user messages, etc., and uses the back end of the information platform management system to store, back up, analyze and warn these data through digital means, so as to timely discover existing problems and deficiencies based on these learning exchanges and feedback, and provide strong basic data support for the construction decisions of organizations, teams, etc., which will help promote work improvements.

[0024] In particular, in the above-mentioned information platform management system, it is crucial to conduct semantic understanding and sentiment analysis of students' learning experience, which can quickly identify whether the students' reactions to activities or learning materials are positive, negative or neutral. This helps to discover potential problems in a timely manner. For example, some students may be confused or dissatisfied with specific learning content. If this negative emotion is not handled in a timely manner, it may further affect the students' enthusiasm and the construction of organizations and teams. Not only that, the sentiment analysis results obtained by semantic understanding of the learning experience report can serve as an important basis for improving work. If the overall learning experience semantics shows a negative emotional tendency, then it is necessary to consider adjusting the existing learning materials or methods to better stimulate students' enthusiasm and initiative in learning.

[0025] Based on this, in the process of semantic understanding of the learning experience report to obtain the result of sentiment analysis, the technical concept of the present application is to randomly split a plurality of learning experience report text descriptions into two parts, each of which contains a specific number of learning experience report text descriptions, and then introduce data processing and semantic understanding algorithms based on artificial intelligence and natural language processing technology in the back end to respectively perform semantic dynamic aggregation analysis on the sets of two learning experience report text descriptions, so as to respectively capture the semantic aggregation representations of the two learning experience report sets, and use the characteristic differences between the semantic aggregation representations of the two learning experience reports with emotions as the theme to analyze the emotional tendency of the learning experience report, so as to identify the positive, negative or neutral feedback in the learning experience report. The emotional tendency in the learning experience report can be automatically identified by the game learning method. If the emotional tendency is found to be negative, the corresponding early warning prompt can be triggered by the early warning unit to remind the management personnel to pay attention and take measures to solve the problem. This intelligent information platform management system and instant feedback mechanism can greatly shorten the time interval from problem discovery to solution, thereby providing strong basic data support for the construction decision-making of organizations, teams, etc., and thus helping to promote the improvement of work.

[0026] Figure 2 This is a schematic block diagram of an intelligent semantic understanding unit in an information platform management system according to an embodiment of the present application. Figure 3 Schematic diagram of data flow of intelligent semantic understanding unit in the information platform management system of the present application embodiment. Figure 2 and Figure 3As shown, the intelligent semantic understanding unit 230 includes: a learning experience report extraction subunit 231, which is used to extract a set of learning experience reports from the information storage unit; a learning experience report text description extraction subunit 232, which is used to extract text parts from each learning experience report in the set of learning experience reports to obtain a set of learning experience report text descriptions; a random binary processing subunit 233, which is used to randomly binary the set of learning experience report text descriptions to obtain a first set of learning experience report text descriptions and a second set of learning experience report text descriptions; a text description semantic understanding subunit 234, which is used to perform semantic understanding on the first set of learning experience report text descriptions and the second set of learning experience report text descriptions to obtain a first set of learning experience report semantic encoding feature vectors and a second set of semantically encoded feature vectors of learning experience reports; a feature distribution field domain semantic dynamic aggregation subunit 235, used to perform feature distribution field domain semantic dynamic aggregation on the first set of semantically encoded feature vectors of learning experience reports and the second set of semantically encoded feature vectors of learning experience reports, respectively, to obtain a first set of semantically aggregated representation vectors of learning experience reports and a second set of semantically aggregated representation vectors of learning experience reports; an emotion adversarial semantic representation generation subunit 236, used to input the first set of semantically aggregated representation vectors of learning information reports and the second set of semantically aggregated representation vectors of learning information reports into an emotion theme adversarial network to obtain an emotion adversarial semantic representation of learning experience reports; an emotion analysis result generation subunit 237, used to generate the emotion analysis result based on the emotion adversarial semantic representation of learning experience reports.

[0027] In the above-mentioned information platform management system, the learning experience report extraction subunit 231 is used to extract a collection of learning experience reports from the information storage unit. It should be understood that the learning experience report contains the student's personal understanding, feelings and possible problems of the learning materials. By extracting these reports, the students' learning attitudes, understanding and acceptance can be analyzed so as to promptly discover any negative emotions or other problems, thereby helping managers to adjust strategies in a timely manner and improve the learning experience. In addition, extracting a collection of learning experience reports is also a basic step in sentiment analysis. Only when enough data is obtained can semantic understanding and sentiment analysis be further performed, and finally sentiment analysis results are generated to provide a reference for construction decisions of organizations, teams, etc.

[0028] In the above-mentioned information platform management system, the learning experience report text description extraction subunit 232 is used to extract text parts from each learning experience report in the set of learning experience reports to obtain a set of learning experience report text descriptions. It should be understood that by extracting the text part, the system can focus on the text content in the learning experience report submitted by the trainee, rather than other elements in the report such as format, signature or other non-text information. Doing so helps to more accurately understand the trainee's reaction to the activities or learning materials, thereby quickly identifying the trainee's positive, negative or neutral attitude towards these activities or materials.

[0029] Optionally, in one embodiment of the present application, extracting text parts from each learning experience report in the set of learning experience reports to obtain a set of learning experience report text descriptions includes: removing non-text elements such as HTML tags, special characters, numbers, punctuation marks, etc. in the text, and retaining the plain text content. Use regular expressions to match and extract text paragraphs in the learning experience report, excluding non-text parts such as titles and author information.

[0030] In the above-mentioned information platform management system, the random binary processing subunit 233 is used to randomly divide the set of the learning experience report text description into two sets to obtain a first set of learning experience report text description and a second set of learning experience report text description. It should be understood that, considering that when the emotional tendency in the learning experience report text description is actually analyzed and identified, since only a part of the emotional tendency in the student's learning experience report may be negative, and it is more subtle in the text description, it is difficult to detect and discover through the traditional semantic feature extraction method. Therefore, in the technical solution of the present application, the set of the learning experience report text description is further randomly divided into two sets to obtain a first set of learning experience report text description and a second set of learning experience report text description. It should be understood that random segmentation can ensure that the data distribution in the two sets is similar, thereby avoiding analysis bias caused by data skewness. This method helps to improve the reliability and fairness of subsequent analysis results. The two sets of learning experience report text descriptions after random binary division can be used to build a game learning environment. Semantic aggregation feature information is extracted from different learning experience report text description sets after binary division by joint game learning, and the semantic aggregation features of the two are compared based on emotional themes to find the commonalities and subtle emotional differences between them. This information is very important for identifying emotional tendencies in learning experience reports, especially when the emotional tendencies are not very obvious, comparative analysis can help the model distinguish different emotional categories more accurately.

[0031] In the above-mentioned information platform management system, the text description semantic understanding subunit 234 is used to perform semantic understanding on the first set of the learning experience report text description and the second set of the learning experience report text description to obtain a first set of learning experience report semantic coding feature vectors and a second set of learning experience report semantic coding feature vectors. It should be understood that the individual learning experience report text descriptions in the first set of the learning experience report text descriptions are semantically coded to extract the semantic feature information of the individual learning experience report text descriptions in the first set, thereby obtaining a first set of learning experience report semantic coding feature vectors. In addition, the individual learning experience report text descriptions in the second set of the learning experience report text descriptions are semantically coded to extract the semantic feature information of the individual learning experience report text descriptions in the second set, thereby obtaining a second set of learning experience report semantic coding feature vectors.

[0032] Optionally, in one embodiment of the present application, the text description semantic understanding subunit 234 is used to: semantically encode each learning experience report text description in a first set of the learning experience report text description to obtain a first set of semantic encoding feature vectors of the learning experience report; and semantically encode each learning experience report text description in a second set of the learning experience report text description to obtain a second set of semantic encoding feature vectors of the learning experience report.

[0033] Optionally, in a specific embodiment of the present application, a converter-based context encoder is used to semantically encode each learning experience report text description in a first set of learning experience report text descriptions to obtain a first set of semantic encoding feature vectors of the learning experience report, and the converter-based context encoder is used to semantically encode each learning experience report text description in a second set of learning experience report text descriptions to obtain a second set of semantic encoding feature vectors of the learning experience report.

[0034] In the above-mentioned information platform management system, the feature distribution field semantic dynamic aggregation subunit 235 is used to perform feature distribution field semantic dynamic aggregation on the first set of the learning experience report semantic coding feature vectors and the second set of the learning experience report semantic coding feature vectors to obtain the first set of learning experience report semantic aggregation representation vectors and the second set of learning experience report semantic aggregation representation vectors. It should be understood that each learning experience report semantic coding feature vector in the first set of learning experience report semantic coding feature vectors respectively contains different learning experience report semantic feature information in the first set, and each learning experience report semantic coding feature vector in the second set of learning experience report semantic coding feature vectors respectively contains different learning experience report semantic feature information in the second set. Here, each learning experience report semantic feature is regarded as a semantic particle. When participating in the subsequent semantic feature aggregation process to obtain a more sufficient and accurate learning experience report semantic aggregation representation, these different learning experience report semantic particles will have different importance and contribution to the subsequent sentiment analysis task. Based on this, in the technical solution of the present application, the first set of the learning experience report semantic encoding feature vectors and the second set of the learning experience report semantic encoding feature vectors are further subjected to feature distribution field semantic dynamic aggregation to obtain the first set of learning experience report semantic aggregation representation vectors and the second set of learning experience report semantic aggregation representation vectors. In particular, here, the process of feature distribution field semantic dynamic aggregation is modeled based on the interaction between the semantic particles of each learning experience report through a concept similar to a force field. This force field not only takes into account the spatial relationship between the semantics of each learning experience report (such as Euclidean distance), but also combines the properties of each learning experience report semantics itself (such as static energy factor) and the overall semantic structure of the entire learning experience report semantic set. In this way, the importance of each learning experience report semantic feature in the semantic aggregation process can be dynamically adjusted, and the information can be integrated into a more compact and representative learning experience report semantic aggregation representation form, i.e., a learning experience report semantic aggregation representation vector, according to a specific aggregation strategy.

[0035] Optionally, in one embodiment of the present application, the feature distribution field semantic dynamic aggregation subunit 235 includes: a static energy factor calculation secondary subunit, used to calculate the static energy factor of each learning experience report semantic encoding feature vector in the first set of the learning experience report semantic encoding feature vectors to obtain a set of learning experience report semantic static energy factors; a feature significance aggregation processing secondary subunit, used to perform feature significance aggregation based on the imitation force field on the first set of the learning experience report semantic encoding feature vectors based on the set of learning experience report semantic static energy factors to obtain the first set of learning experience report semantic aggregation representation vectors.

[0036] Optionally, in one embodiment of the present application, the static energy factor calculation secondary subunit is used to: calculate the mean and variance of the semantic coding feature vector of the learning experience report to obtain the semantic mean and semantic variance of the learning experience report; subtract the semantic coding feature vector of the learning experience report from the semantic mean of the learning experience report by position, and calculate the fourth power of each position of the feature vector after subtraction to obtain the semantic difference modulation vector of the learning experience report; calculate the expected value of the semantic difference modulation vector of the learning experience report to obtain the semantic expectation value of the learning experience report; divide the semantic expectation value of the learning experience report by the square of the semantic variance of the learning experience report, and input the value obtained by division into the sigmoid function to obtain the semantic static energy factor of the learning experience report.

[0037] Optionally, in one embodiment of the present application, the feature significant aggregation processing secondary sub-unit includes: a cluster initial center vector selection tertiary sub-unit for selecting the learning experience report semantic encoding feature vector corresponding to the largest learning experience report semantic static energy factor in the set of learning experience report semantic static energy factors as the learning experience report semantic cluster initial center vector; a dynamic aggregation energy factor calculation tertiary sub-unit for calculating the learning experience report semantic encoding feature vector based on the spatial span between each learning experience report semantic encoding feature vector in the first set of the learning experience report semantic encoding feature vector and the learning experience report semantic cluster initial center vector, as well as the static energy factor of each learning experience report semantic encoding feature vector and the learning experience report semantic cluster initial center vector. The static energy factor of the initial center vector is used to calculate the dynamic aggregation energy factor of each learning experience report semantic coding feature vector in the set of the learning experience report semantic coding feature vector to obtain a set of learning experience report semantic dynamic aggregation energy factors; the dynamic aggregation weight factor calculation tertiary subunit is used to input the set of learning experience report semantic dynamic aggregation energy factors into the gated mask unit to obtain a set of learning experience report semantic dynamic aggregation weight factors; the set semantic aggregation processing tertiary subunit is used to use the set of learning experience report semantic dynamic aggregation weight factors as weighted weights to calculate the weighted sum of the first set of learning experience report semantic coding feature vectors to obtain the first set semantic aggregation representation vector of the learning experience report.

[0038] Optionally, in one embodiment of the present application, the set of learning experience report semantic dynamic aggregation energy factors is input into a gated mask unit to obtain a set of learning experience report semantic dynamic aggregation weight factors, including: normalizing the set of learning experience report semantic dynamic aggregation energy factors to obtain a set of normalized learning experience report semantic dynamic aggregation energy factors; masking the set of normalized learning experience report semantic dynamic aggregation energy factors to obtain the set of learning experience report semantic dynamic aggregation weight factors.

[0039] Optionally, in one embodiment of the present application, the set of learning experience report semantic dynamic aggregation energy factors is normalized to obtain a set of normalized learning experience report semantic dynamic aggregation energy factors, including: calculating an exponential function value with the natural constant e as the base and the negative of each learning experience report semantic dynamic aggregation energy factor in the set of learning experience report semantic dynamic aggregation energy factors as the exponent to obtain a set of learning experience report semantic dynamic aggregation energy exponential factors; adding the set of learning experience report semantic dynamic aggregation energy exponential factors to the constant one by position, and then calculating the reciprocal of each learning experience report semantic dynamic aggregation energy exponential modulation factor in the obtained set of learning experience report semantic dynamic aggregation energy exponential modulation factors to obtain the set of normalized learning experience report semantic dynamic aggregation energy factors.

[0040] Optionally, in one embodiment of the present application, the set of normalized learning experience report semantic dynamic aggregation energy factors is masked to obtain the set of learning experience report semantic dynamic aggregation weight factors, including: in response to each normalized learning experience report semantic dynamic aggregation energy factor in the set of normalized learning experience report semantic dynamic aggregation energy factors being greater than a predetermined threshold, setting the normalized learning experience report semantic dynamic aggregation energy factors greater than the predetermined threshold to the original value, and setting the rest to zero, so as to obtain the set of learning experience report semantic dynamic aggregation weight factors.

[0041] Optionally, in one embodiment of the present application, the dynamic aggregation energy factor calculation three-level sub-unit is used to: multiply the static energy factor of the learning experience report semantic encoding feature vector and the static energy factor of the learning experience report semantic clustering initial center vector with a first weighting parameter to obtain a first learning experience report semantic dynamic aggregation energy factor; multiply the square of the spatial span between the learning experience report semantic encoding feature vector and the learning experience report semantic clustering initial center vector with a second weighting parameter to obtain a second learning experience report semantic dynamic aggregation energy factor; divide the first learning experience report semantic dynamic aggregation energy factor by the second learning experience report semantic dynamic aggregation energy factor to obtain the learning experience report semantic dynamic aggregation energy factor.

[0042] In summary, in the embodiment of the present application, the first set of the semantic encoding feature vectors of the learning experience report is subjected to feature distribution field semantic dynamic aggregation using the following semantic dynamic aggregation formula to obtain the first set of learning experience report semantic aggregation representation vectors; The semantic dynamic aggregation formula is: in, A first set of semantic encoding feature vectors for the learning experience report, are the first, second, ..., and third in the first set of semantic encoding feature vectors of the learning experience report. ,..., The semantic encoding feature vector of the learning experience report, It is The feature values ​​of each position in the semantic encoding feature vector of the learning experience report, To calculate the expected value, and They are The mean and variance of yes function, yes The corresponding learning experience report semantic static energy factor, To select the maximum value corresponding to value, is the maximum matching value, is the initial center vector of the semantic clustering of learning experience reports, yes The corresponding learning experience report semantic static energy factor, express and The space span between and is the weighting parameter, for The corresponding learning experience report semantic dynamic aggregation energy factor, yes The corresponding normalized learning experience report semantic dynamic aggregation energy factor, For masking, is a predetermined threshold, yes The corresponding learning experience report semantic dynamic aggregation weight factor, The number of vectors in the first set of semantic encoding feature vectors for the learning experience report, It is the semantic aggregation representation vector of the first set of learning experience reports.

[0043] Specifically, taking the first set of the learning experience report semantic coding feature vectors as an example, in the process of performing feature distribution field semantic dynamic aggregation on the first set of the learning experience report semantic coding feature vectors to obtain the first set of learning experience report semantic aggregation representation vectors, first, for each vector in the first set of input learning experience report semantic coding feature vectors, its stability or importance in the feature space is evaluated based on a predefined metric function, and a numerical representation is assigned, namely, the learning experience report semantic static energy factor. In this way, the status and importance of each learning experience report semantic feature in its data distribution and the overall learning experience report semantics can be quantified, providing a basis for the subsequent selection of the initial clustering center. At the same time, by identifying those learning experience report semantic feature points that are relatively isolated or have a high local influence, it is helpful to discover potential data cluster cores. In this way, the uniqueness and influence of each learning experience report semantics in the feature space can be quantified, providing a basis for the subsequent aggregation process.

[0044] Next, the learning experience report semantic encoding feature vector corresponding to the largest learning experience report semantic static energy factor in the set of learning experience report semantic static energy factors is selected as the initial center vector of learning experience report semantic clustering. That is, a reliable starting point is determined for iterative clustering operations. This point should represent the most representative part of the learning experience report semantic set, which can serve as the basis for subsequent clustering to ensure the rationality of the clustering results. This enhances the clustering process's ability to capture the true semantic set structure and reduces the adverse effects caused by random initialization.

[0045] Then, based on the spatial span between the semantic encoding feature vectors of each learning experience report and the initial center vector of the semantic clustering of the learning experience report, as well as the static energy factor of the semantic encoding feature vectors of each learning experience report and the static energy factor of the initial center vector of the semantic clustering of the learning experience report, the dynamic aggregation energy factor of the semantic encoding feature vectors of each learning experience report is further derived. In particular, the dynamic aggregation energy factor not only takes into account the semantic proximity of the learning experience report in terms of position, but also includes the importance difference of the semantic intrinsic attributes of the learning experience report presented at each point, aiming to more accurately capture the semantic interaction between learning experience reports, so as to measure the semantic importance and contribution of each learning experience report semantic in the overall collection and subsequent feature aggregation process.

[0046] Furthermore, the set of learning experience report semantic dynamic aggregation energy factors is input into the gated mask unit, which automatically adjusts the weight of each learning experience report semantic dynamic aggregation energy factor according to the learned knowledge. In this way, by nonlinearly transforming and screening the original energy factors, the model's attention to the key learning experience report text semantic information is enhanced, while the irrelevant parts are suppressed, so that the learning experience report semantics related to the subsequent tasks are given a higher weight in the aggregation process.

[0047] Finally, the obtained set of learning experience report semantic dynamic aggregation weight factors is used as weighting coefficients to calculate the weighted sum of the first set of learning experience report semantic encoding feature vectors, so as to combine all learning experience report semantic feature information, and at the same time, appropriately emphasize or weaken the feature expression according to the semantic importance of each learning experience report, so as to generate a more concise and representative learning experience report first set semantic aggregation representation vector, which contains the main semantic information of the original learning experience report and removes redundant or irrelevant parts. In this way, not only can more accurate and reliable sentiment analysis results be obtained, but also a solid data foundation can be provided for subsequent decision support and work.

[0048] In the above-mentioned information platform management system, the emotion confrontation semantic representation generation subunit 236 is used to input the first set of learning information report semantic aggregation representation vectors and the second set of learning information report semantic aggregation representation vectors into the emotional theme confrontation network to obtain the learning experience report emotion confrontation semantic representation. It should be understood that since the first set of learning information report semantic aggregation representation vectors and the second set of learning information report semantic aggregation representation vectors respectively contain the first set of learning information report semantic aggregation feature representations and the second set of learning information report semantic aggregation feature representations, the semantic aggregation representations of the two respectively reflect the different learning experience report text description sets obtained by random binary division in the learning information report. Therefore, in order to be able to compare the semantic aggregation features of the two based on emotional themes through game learning, to discover the commonalities and subtle emotional differences between them, so as to perform emotional analysis and recognition, and provide a data basis for the construction work of organizations, teams, etc., in the technical solution of the present application, the first set of learning information report semantic aggregation representation vectors and the second set of learning information report semantic aggregation representation vectors are further input into the emotional theme confrontation network to obtain the learning experience report emotional confrontation semantic representation vectors.

[0049] Optionally, in one embodiment of the present application, the emotion confrontation semantic representation generating subunit 236 is used to: calculate the position difference between the semantic aggregation representation vector of the first set of learning information reports and the semantic aggregation representation vector of the second set of learning information reports to obtain the learning experience report emotion confrontation semantic representation vector as the learning experience report emotion confrontation semantic representation. It should be understood that when the set of learning experience report text descriptions is randomly divided into two sets to form two different sets, each set will obtain its own semantic aggregation representation vector through the semantic understanding process. These two vectors reflect the semantic features of different sets respectively. By calculating the position difference between these two vectors, an information vector containing the semantic difference between the two sets can be obtained, that is, the learning experience report emotion confrontation semantic representation vector. This difference vector is actually the expression of different emotional tendencies in the learning experience report, which highlights the differences in semantic features in different sets, so that the system can more keenly capture subtle changes in emotions. The use of the position difference method can ensure that even in the case of subtle emotional differences, it can be revealed by comparing the semantic representations between different sets. This method is particularly suitable for dealing with situations where the emotional tendency is not very obvious. Comparative analysis can help the model distinguish different emotional categories more accurately. In addition, this method can also strengthen the focus on key semantic information while suppressing irrelevant parts, thereby giving higher weight to emotional features related to subsequent tasks during the aggregation process.

[0050] In the above-mentioned information platform management system, the sentiment analysis result generation subunit 237 is used to generate the sentiment analysis result based on the sentiment confrontation semantic representation of the learning experience report. Optionally, in one embodiment of the present application, the sentiment analysis result generation subunit is used to: input the sentiment confrontation semantic representation vector of the learning experience report into a classifier-based sentiment identifier to obtain the sentiment analysis result. That is to say, the sentiment confrontation semantic representation of the learning experience report obtained through joint game learning is used for classification processing to analyze the emotional tendency of the learning experience report to identify positive, negative or neutral feedback in the learning experience report. The emotional tendency in the learning experience report can be automatically identified by game learning. If the emotional tendency is found to be negative, the corresponding early warning prompt can be triggered by the early warning unit to remind the management personnel to pay attention and take measures to solve the problem. This intelligent information platform management system and instant feedback mechanism can greatly shorten the time interval from problem discovery to solution, thereby providing strong basic data support for decision-making, and then help promote the improvement of the construction work of organizations, teams, etc.

[0051] When the semantic aggregation representation vector of the first set of learning information reports and the semantic aggregation representation vector of the second set of learning information reports respectively represent the report semantic understanding clustering features of the first set of text descriptions of the learning experience reports and the report semantic understanding clustering features of the second set of text descriptions of the learning experience reports, when they are input into the emotional theme adversarial network, the obtained learning experience report emotional adversarial semantic representation vector will also have semantic representation diversity due to the differences in semantic distribution position and semantic distribution granularity in the semantic aggregation representation vector of the first set of learning information reports and the semantic aggregation representation vector of the second set of learning information reports, thereby affecting the iterative consistency of the classification mapping and reducing the accuracy of the classification results.

[0052] Preferably, the emotional adversarial semantic representation vector of the learning experience report is passed through a classifier-based emotion identifier to obtain an emotional analysis result, including: Performing statistical feature analysis on the emotion adversarial semantic representation vector of the learning experience report to obtain the learning experience report semantic aggregation principal component amplitude, the learning experience report semantic aggregation statistical mean and the learning experience report semantic aggregation dispersion index; Based on the learning experience report semantic aggregation principal component amplitude, the learning experience report semantic aggregation statistical mean and the learning experience report semantic aggregation dispersion index, the learning experience report semantic aggregation primary dynamic adaptation range value and the learning experience report semantic aggregation secondary dynamic adaptation range value are generated, which are expressed as: in, represents the amplitude of the semantic aggregation principal component of the learning experience report, represents the semantic aggregation statistical mean of the learning experience report, represents the semantic aggregation dispersion index of the learning experience report, Indicates the learning experience report semantic aggregation primary dynamic adaptation range value, Indicates the learning experience report semantic aggregation secondary dynamic adaptation range value; After performing inverse operation on the learning experience report emotion confrontation semantic representation vector, the learning experience report semantic aggregation primary dynamic adaptation range value and the learning experience report semantic aggregation secondary dynamic adaptation range value are used to perform feature multiplier modulation on the results respectively, and then the original fluctuation of the learning experience report emotion confrontation semantic representation vector is eliminated by differential operation to obtain the learning experience report semantic aggregation dynamic adaptation probability vector, which is expressed as: in, represents the learning experience report emotion adversarial semantic representation vector, It means point multiplication by position. It means subtracting by position. Represents the learning experience report semantic aggregation dynamic adaptation probability vector; After the natural exponential basis transformation is performed on the learning experience report semantic aggregation dynamic adaptation probability vector, the learning experience report semantic aggregation primary dynamic adaptation range value and the learning experience report semantic aggregation secondary dynamic adaptation range value are used to perform translation modulation on the learning experience report semantic aggregation to obtain the learning experience report semantic aggregation prediction deduction vector, which is expressed as: in, represents the natural exponential function, represents the weight hyperparameter, It means adding by position. Represents the semantic aggregation prediction and deduction vector of learning experience report; Information deduction is performed on the learning experience report semantic aggregation prediction deduction vector to obtain an optimized learning experience report emotion adversarial semantic representation vector, which is expressed as: in, Representation optimized learning experience report sentiment adversarial semantic representation vector; The optimized learning experience report emotion adversarial semantic representation vector is passed through the classifier-based emotion identifier to obtain the emotion analysis result.

[0053] Accordingly, in this preferred embodiment, the learning experience report emotion confrontation semantic representation vector is subjected to a standardized analysis without distribution assumptions, a probability inference range of the global representation quantity based on the feature vector is constructed, and the correlation attribution test between the components is implemented by parameter-based dynamic probability adaptation, thereby verifying the feasibility of the quasi-standardized representation mode to the composite optimization benchmark. Finally, a dynamic balance between end-to-end fidelity and information integrity coefficient is achieved in the construction of the deduction model, ensuring that the prediction system of the learning experience report emotion confrontation semantic representation vector has an operational optimization path to improve the execution iteration consistency of each local feature distribution in the mapping task, so as to improve the feature classification convergence effect. In this way, the accuracy of the sentiment analysis results obtained by the classifier-based sentiment identifier of the learning experience report emotion confrontation semantic representation vector is improved. It is possible to automatically identify the emotional tendency in the overall learning experience report. If the emotional tendency is found to be negative, the corresponding warning prompt can be triggered through the warning unit to remind the management personnel to pay attention and take measures to solve the problem.

[0054] In summary, according to the information platform management system of the embodiment of the present application, the front end of the information platform management system supports students to upload learning experiences, publish meeting minutes, and receive user messages, etc., and uses the back end of the information platform management system to store, back up, analyze and warn these data through digital means, so as to timely discover existing problems and deficiencies based on these learning exchanges and feedback, provide strong basic data support for decision-making, and thus help promote work improvements.

[0055] Figure 4 Schematic flow chart of the information platform management method of the embodiment of the present application. Figure 4As shown, the information platform management method includes: S1, extracting a set of learning experience reports from an information storage unit; S2, extracting text parts from each learning experience report in the set of learning experience reports to obtain a set of learning experience report text descriptions; S3, randomly dividing the set of learning experience report text descriptions into two sets to obtain a first set of learning experience report text descriptions and a second set of learning experience report text descriptions; S4, performing semantic understanding on the first set of learning experience report text descriptions and the second set of learning experience report text descriptions to obtain a first set of learning experience report semantic coding feature vectors and a learning experience report text description. A second set of semantic coding feature vectors; S5, performing feature distribution field semantic dynamic aggregation on the first set of semantic coding feature vectors of the learning experience report and the second set of semantic coding feature vectors of the learning experience report, respectively, to obtain the first set of semantic aggregation representation vectors of the learning experience report and the second set of semantic aggregation representation vectors of the learning experience report; S6, inputting the first set of semantic aggregation representation vectors of the learning information report and the second set of semantic aggregation representation vectors of the learning information report into the emotion theme adversarial network to obtain the emotion adversarial semantic representation of the learning experience report; S7, generating the emotion analysis result based on the emotion adversarial semantic representation of the learning experience report.

[0056] The specific operations of each step in the above information platform management method have been referenced above. Figures 1 to 3 It has been introduced in detail in the description of the information platform management system, and therefore, its repeated description will be omitted.

[0057] The basic principle of the present invention is described above in conjunction with specific embodiments. However, it should be pointed out that the advantages, strengths, effects, etc. mentioned in the present invention are only examples and not limitations, and it cannot be considered that these advantages, strengths, effects, etc. must be possessed by each embodiment of the present invention. In addition, the specific details of the above embodiments are only for the purpose of illustration and facilitation of understanding, rather than limitation, and the above details do not limit the present invention to being implemented by adopting the above specific details.

[0058] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, please refer to the relevant description of other embodiments. In the several embodiments provided by the present invention, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the module division is only a logical function division, and there may be other division methods in actual implementation. The modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment.

[0059] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0060] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference to a figure in a claim should not be considered as limiting the claim to which it relates.

[0061] In addition, it is obvious that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units stated in the device claim can also be implemented by one unit through software or hardware.

[0062] Finally, it should be noted that the above description has been given for the purpose of illustration and description. In addition, the above embodiments are only used to illustrate the technical solution of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. An information platform management system, characterized in that: include: A front-end module and a back-end management module connected by communication; The front-end module includes a learning experience submission unit, a meeting record publishing unit and a message unit; The backend management module includes an information storage unit, a data backup unit, an intelligent semantic understanding unit and an early warning unit; The intelligent semantic understanding unit comprises: A learning experience report extraction subunit, used to extract a set of learning experience reports from the information storage unit; A learning experience report text description extraction subunit, used for extracting a text portion from each learning experience report in the set of learning experience reports to obtain a set of learning experience report text descriptions; A random binary processing subunit, used for randomly binary-dividing the set of learning experience report text descriptions to obtain a first set of learning experience report text descriptions and a second set of learning experience report text descriptions; A text description semantic understanding subunit, used for performing semantic understanding on the first set of the learning experience report text descriptions and the second set of the learning experience report text descriptions to obtain a first set of learning experience report semantic coding feature vectors and a second set of learning experience report semantic coding feature vectors; The feature distribution field semantic dynamic aggregation subunit is used to perform feature distribution field semantic dynamic aggregation on the first set of semantic encoding feature vectors of the learning experience report and the second set of semantic encoding feature vectors of the learning experience report, respectively, to obtain the semantic aggregation representation vector of the first set of learning experience reports and the semantic aggregation representation vector of the second set of learning experience reports.

2. The information platform management system according to claim 1, characterized in that: The learning experience submission unit is used to receive learning experience reports submitted by users, the meeting record publishing unit is used to publish meeting records, and the message unit is used to receive user messages; The information storage unit is used to store student information, learning experience reports, meeting minutes and user messages, and the data backup unit is used to back up data regularly; the intelligent semantic understanding unit is used to perform semantic understanding of the learning experience report to obtain a sentiment analysis result; the early warning unit is used to generate a warning prompt in response to the sentiment analysis result being negative.

3. The information platform management system according to claim 2 is characterized in that: The intelligent semantic understanding unit further includes: An emotion adversarial semantic representation generating subunit, used for inputting the first set of learning information report semantic aggregation representation vectors and the second set of learning information report semantic aggregation representation vectors into an emotional theme adversarial network to obtain an emotion adversarial semantic representation of a learning experience report; The sentiment analysis result generating subunit is used to generate the sentiment analysis result based on the learning experience report sentiment adversarial semantic representation.

4. The information platform management system according to claim 3 is characterized in that: The text description semantic understanding subunit is used to: Performing semantic coding on each learning experience report text description in the first set of learning experience report text descriptions to obtain a first set of learning experience report semantic coding feature vectors; Semantic encoding is performed on each learning experience report text description in the second set of learning experience report text descriptions to obtain a second set of learning experience report semantic encoding feature vectors.

5. The information platform management system according to claim 4, characterized in that: The feature distribution field semantic dynamic aggregation subunit includes: A static energy factor calculation secondary subunit is used to calculate the static energy factor of each learning experience report semantic coding feature vector in the first set of learning experience report semantic coding feature vectors to obtain a set of learning experience report semantic static energy factors; The feature significant aggregation processing secondary sub-unit is used to perform feature significant aggregation based on the imitation field on the first set of semantic encoding feature vectors of the learning experience report based on the set of semantic static energy factors of the learning experience report to obtain the semantic aggregation representation vector of the first set of learning experience reports.

6. The information platform management system according to claim 5, characterized in that: The static energy factor calculation secondary subunit is used for: Calculating the mean and variance of the semantic encoding feature vector of the learning experience report to obtain the semantic mean of the learning experience report and the semantic variance of the learning experience report; Subtracting the learning experience report semantic coding feature vector from the learning experience report semantic mean by position, and calculating the fourth power of each position of the feature vector after the subtraction to obtain the learning experience report semantic difference modulation vector; Calculating the expected value of the semantic difference modulation vector of the learning experience report to obtain the semantic expected value of the learning experience report; The semantic expected value of the learning experience report is divided by the square of the semantic variance of the learning experience report, and the value obtained by the division is input into the sigmoid function to obtain the semantic static energy factor of the learning experience report.

7. The information platform management system according to claim 6, characterized in that: The characteristic significant aggregation processing secondary subunit includes: The clustering initial center vector selects a three-level sub-unit, which is used to select the learning experience report semantic encoding feature vector corresponding to the largest learning experience report semantic static energy factor in the set of learning experience report semantic static energy factors as the learning experience report semantic clustering initial center vector; The third-level sub-unit for dynamic aggregation energy factor calculation is used to calculate the dynamic aggregation energy factor of each learning experience report semantic encoding feature vector in the set of the learning experience report semantic encoding feature vectors based on the spatial span between each learning experience report semantic encoding feature vector in the first set of the learning experience report semantic encoding feature vectors and the initial center vector of the learning experience report semantic clustering, as well as the static energy factor of each learning experience report semantic encoding feature vector and the static energy factor of the initial center vector of the learning experience report semantic clustering, so as to obtain a set of learning experience report semantic dynamic aggregation energy factors; The third-level subunit for calculating the dynamic aggregation weight factor is used to input the set of the learning experience report semantic dynamic aggregation energy factors into the gated mask unit to obtain the set of learning experience report semantic dynamic aggregation weight factors; The third-level sub-unit for set semantic aggregation processing is used to use the set of semantic dynamic aggregation weight factors of the learning experience report as weighted weights, calculate the weighted sum of the first set of semantic encoding feature vectors of the learning experience report to obtain the first set semantic aggregation representation vector of the learning experience report.

8. The information platform management system according to claim 7, characterized in that: The dynamic aggregation energy factor calculation three-level subunit is used for: Multiplying the static energy factor of the learning experience report semantic encoding feature vector and the static energy factor of the learning experience report semantic clustering initial center vector by the first weighting parameter to obtain the first learning experience report semantic dynamic aggregation energy factor; Multiplying the square of the spatial span between the learning experience report semantic encoding feature vector and the learning experience report semantic clustering initial center vector by a second weighting parameter to obtain a second learning experience report semantic dynamic aggregation energy factor; The first learning experience report semantic dynamic aggregation energy factor is divided by the second learning experience report semantic dynamic aggregation energy factor to obtain the learning experience report semantic dynamic aggregation energy factor.

9. The information platform management system according to claim 8, characterized in that: The emotion confrontation semantic representation generating subunit is used to: calculate the position difference between the first set semantic aggregation representation vector of the learning information report and the second set semantic aggregation representation vector of the learning information report to obtain the learning experience report emotion confrontation semantic representation vector as the learning experience report emotion confrontation semantic representation; The sentiment analysis result generating subunit is used to: input the learning experience report sentiment adversarial semantic representation vector into a classifier-based sentiment identifier to obtain the sentiment analysis result.

10. An information platform management method, characterized in that: include: Extracting a collection of learning experience reports from the information storage unit; Extracting a text portion from each learning experience report in the set of learning experience reports to obtain a set of learning experience report text descriptions; Randomly dividing the set of learning experience report text descriptions into two sets to obtain a first set of learning experience report text descriptions and a second set of learning experience report text descriptions; Performing semantic understanding on the first set of the learning experience report text descriptions and the second set of the learning experience report text descriptions to obtain a first set of learning experience report semantic coding feature vectors and a second set of learning experience report semantic coding feature vectors; Performing feature distribution field semantic dynamic aggregation on the first set of semantically encoded feature vectors of the learning experience report and the second set of semantically encoded feature vectors of the learning experience report to obtain a first set of semantically aggregated representation vectors of the learning experience report and a second set of semantically aggregated representation vectors of the learning experience report; Inputting the first set of semantic aggregation representation vectors of the learning information report and the second set of semantic aggregation representation vectors of the learning information report into an emotional theme adversarial network to obtain emotional adversarial semantic representations of learning experience reports; The sentiment analysis result is generated based on the emotion adversarial semantic representation of the learning experience report.

Images