Intelligent school selection system based on man-machine conversation interaction data

Through human-computer dialogue interaction technology, the competitiveness values ​​of users and colleges and universities are calculated, and clear and transparent school selection suggestions are provided, which solves the problems of inaccurate recommendation results and opaque processes in the existing school selection system, and improves the school selection results Accuracy and user satisfaction.

CN119088908APending Publication Date: 2024-12-06BEIJING XUECHENG GUILAI EDUCATION TECH CO LTD
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202311448607.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2023-11-02
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

The recommendation results in the existing school selection system are inaccurate, the process is opaque, and the user passively accepts the results, resulting in the user being dissatisfied with the recommendation results.

Method used

Design an intelligent school selection system based on human-computer dialogue interaction data, including a school selection operating system, an intelligent dialogue system and a matching recommendation system. Through user's natural language recognition and analysis, users' postgraduate entrance examination competitiveness value and the competition of colleges and majors The force requirements value provides clear and transparent school selection advice.

Benefits of technology

It significantly improves the accuracy of the results of the calibration selection and the user's satisfaction with the results, ensures users' clear understanding of the process of the calibration selection, provides timely feedback and specific information, and ensures the trustworthiness of the recommended results.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119088908A_ABST
    Figure CN119088908A_ABST
Patent Text Reader

Abstract

The invention relates to an intelligent school selection system based on man-machine conversation interaction data. The system comprises a school selection operation system, an intelligent conversation system, a matching recommendation system and a data entry system. According to the intelligent school selection system based on the man-machine dialog interaction data, a man-machine dialog interaction mode is adopted, specific information can be fed back and informed to a user in time in each dialog, basic information and intention information of the user are obtained step by step, it is guaranteed that the user is clear and transparent in the whole school selection process, the accuracy of a school selection result is remarkably improved, and the user experience is improved. According to the method, the satisfaction degree of a user on result delivery is ensured, corresponding parameters are given to basic information fields of college specialty, a competitiveness requirement value of each college specialty is calculated, corresponding weight values are given to each piece of characteristic information of the user, an examination competitiveness level value of each examinee is calculated, and the examination competitiveness level value of each examinee is calculated. And performing matching degree calculation according to the examination competitiveness of the examinees and the competitive power requirement value of the college specialty, and giving specialty school selection recommendation for the students.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of educational technology, and in particular to an intelligent school selection system based on human-computer dialogue interaction data. Background Art

[0002] With the development of society and the education industry, the public's pursuit of higher education has become increasingly strong, so postgraduate entrance examination has become an option for everyone. The first thing students have to face when taking the postgraduate entrance examination is to determine the school and major they apply for. However, there are more than 800 graduate schools and more than 2,000 majors in the country. Students will face a lot of choices, which will make it difficult to choose.

[0003] Among the most similar technical solutions currently available, the main method steps are: the first step is to obtain a preliminary screening of several colleges and majors through the basic requirements of the candidates for the postgraduate entrance examination colleges and majors; the second step is to match the screened colleges with the user's personal situation to give a clear recommendation. In the first step, it is necessary to disassemble the various field information of the college and major, including the college's ranking, level, region and college characteristics, as well as to establish a mapping relationship between colleges and majors to clarify which specific majors different colleges will recruit. In the second step, it is necessary to collect the user's personal information, including university, university major, grade ranking, English level, preparation time and other information fields, match the user information and college information through an algorithm, and recommend a batch of colleges and majors in a targeted manner. After these two steps, finally recommend a number of colleges and majors to the user, and rank the recommended colleges and majors according to certain rules for the user to choose and determine.

[0004] Among the most similar technical solutions currently available, the first step can generally be achieved, which mainly involves disassembling the various label data of colleges and majors, including information fields such as college rankings, levels, regions and characteristics. After the user determines his or her personal major, college level and region, etc., a batch of college and major information can be obtained through field screening. In the second step, for the collected user information, how does the system calculate these field information into digital signals and apply them to the recommendation results? The process is not clear and transparent, and users passively accept the results. In addition, the recommended results are often inaccurate, and users are not satisfied with the recommended results. Therefore, an intelligent school selection system based on human-computer dialogue interaction data is proposed to solve the above problems. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention provides an intelligent school selection system based on human-computer dialogue interaction data, which has the advantage of accurate recommendation results and solves the problems of unclear and transparent process, users passively accepting results, and often inaccurate recommendation results and users are not satisfied with the recommendation results.

[0006] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: an intelligent school selection system based on human-computer dialogue interaction data, including a school selection operating system, an intelligent dialogue system, a matching recommendation system and a data entry system;

[0007] Among them, the school selection operating system is used to: allow users to input and modify personal information and intention information, and provide information to the matching recommendation system, and finally receive and display the school selection result report information fed back by it;

[0008] Among them, the intelligent dialogue system is used to: identify and analyze the user's natural language, and ultimately understand the user's personal information and intention information;

[0009] Among them, the matching recommendation system is used to: calculate and match the difficulty of colleges and universities based on user information and intention information, and recommend colleges and universities that meet the requirements and have preparation guidance significance;

[0010] Among them, the data entry system is used to: input school information, major information and annual recruitment data, etc.

[0011] Furthermore, the school selection operating system consists of three parts, namely, school selection function entrance, school selection information input and school selection result feedback. The school selection information input includes personal information and intention information. The personal information includes current status, university, university major, CET-4 and CET-6 scores, planned preparation time and mathematical foundation. The intention information includes intended major, intended region, school level and training method.

[0012] Furthermore, the school selection result feedback will eventually give the user a school selection result recommendation, that is, a school selection report. The report first completes the corresponding postgraduate entrance examination competitiveness ability decomposition and calculation based on the user's personal information and intention information, and obtains the final postgraduate entrance examination competitiveness level value. At the same time, an analysis of the recommendation reasons is given, and finally it is labeled as "safe, sprint, and bottom line" and recommended to the user.

[0013] Furthermore, the intelligent dialogue system includes dialogue triggering logic, dialogue judgment flow, system feedback speech, and a speech recognition module, wherein the speech recognition module is used to quickly identify user information based on the user's personal information.

[0014] Furthermore, the data entry system includes information related to institutions, majors and annual enrollment information.

[0015] Furthermore, the information related to the institution includes the name of the institution, the level of the institution, the ranking of the institution, the city where the institution is located, the type of institution and the information of its subordinate departments, etc. The information related to the major includes the subject category, first-level discipline, major name, major code and subject evaluation, etc.

[0016] Furthermore, the matching recommendation system calculates the required value of the competitiveness of each school major in the postgraduate entrance examination based on the obtained school major information. The matching recommendation system divides all schools into 6 levels, namely A1 to A6, for school hierarchy information, and then assigns corresponding weight values ​​to different levels. The matching recommendation system divides professional subject evaluation into 10 levels, namely B1 to B10, for professional subject evaluation, and then assigns corresponding weight values ​​to different levels. The matching recommendation system divides the regional cities of the schools into 3 levels, namely C1 to C3, for the regional cities, and then assigns corresponding weight values ​​to different levels. The matching recommendation system divides the training methods into 2 levels, namely D1 and D2, for professional training methods, and then assigns corresponding weight values ​​to different levels. The matching system recommends school ranking information and substitutes the specific ranking value into a linear algorithm formula to obtain the ranking value E. The matching recommendation system obtains the calculation formula for each ABCD item as follows:

[0017]

[0018] Furthermore, the linear algorithm formula of the ranking value E is:

[0019] E=5.008772-school ranking×0.008772 (when the school ranking is greater than 570, E=0). The calculation formula for the final difficulty of admission to graduate school majors is: M=A+B+C+D+E+50.

[0020] Furthermore, the matching recommendation system calculates the user's postgraduate entrance examination competitiveness value based on the acquired user personal information. The matching recommendation system divides all the user's universities into 7 levels, namely X1 to X7, and then assigns corresponding weight values ​​to different levels. The matching recommendation system divides the user's academic level into 4 levels, namely Y1 to Y4, and then assigns corresponding weight values ​​to different levels. The matching recommendation system divides the CET-4 and CET-6 scores into 3 levels, namely P1 to P3, and then assigns corresponding weight values ​​to different levels. The matching recommendation system divides the user's preparation time into 3 levels, namely Q1 to Q3, and then assigns corresponding weight values ​​to different levels. The matching recommendation system divides the user's mathematical foundation into 6 levels, namely T1 to T6, and then assigns corresponding weight values ​​to different levels.

[0021] Furthermore, the matching recommendation system obtains the calculation formula for each item of XYPQT:

[0022]

[0023] The final level of the test-taking user is: N = X + Y + P + Q + T + 50.

[0024] Then subtract the user's postgraduate entrance examination competitiveness value N from the postgraduate entrance examination competitiveness requirement value M of each college's major to obtain R. If R∈[-7, 7], the matching degree is five stars and the recommended level is safe. If R∈(7, 15], the matching degree is four stars and the recommended level is sprint. If R∈[-15, -7), the matching degree is three stars and the recommended level is minimum. If R∈(-∞, -15), the matching degree is two stars and the recommended level is not recommended. If R∈(15, +∞), the matching degree is one star and the recommended level is not recommended.

[0025] Compared with the prior art, the technical solution of this application has the following beneficial effects:

[0026] This intelligent school selection system based on human-computer dialogue interaction data breaks the black box of the school selection process, allowing users to have a clearer understanding of the school selection process and thus have greater trust in the school selection results. It adopts a human-computer dialogue interaction method, and each conversation will provide timely feedback to inform users of specific information, thereby gradually obtaining users' basic information and user intention information, thereby ensuring that users have a clear and transparent understanding of the entire school selection process, significantly improving the accuracy of school selection results, and ensuring user satisfaction with the results. The basic information fields of colleges and majors are assigned corresponding parameters, and the corresponding competitiveness requirements for each college and major are calculated. Then, corresponding weight values ​​are assigned to each characteristic information of the user, and the postgraduate entrance examination competitiveness level of each candidate is calculated. Finally, the matching degree is calculated based on the candidate's postgraduate entrance examination competitiveness and the competitiveness requirements of the college and major, and a number of college and major school selection recommendations are given to students. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is a system structure block diagram of the present invention;

[0028] Figure 2 It is a structural block diagram of the school selection system in the present invention;

[0029] Figure 3 It is a structural block diagram of the data entry system in the present invention;

[0030] Figure 4 This is a structural block diagram of the speech recognition module in the present invention. DETAILED DESCRIPTION

[0031] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0032] See also Figure 1-4 , in this embodiment, an intelligent school selection system based on human-computer dialogue interaction data includes a school selection operating system, an intelligent dialogue system, a matching recommendation system and a data entry system;

[0033] Among them, the school selection operating system is used to: allow users to input and modify personal information and intention information, and provide information to the matching recommendation system, and finally receive and display the school selection result report information fed back by it;

[0034] Among them, the intelligent dialogue system is used to: identify and analyze the user's natural language, and ultimately understand the user's personal information and intention information;

[0035] Among them, the matching recommendation system is used to: calculate and match the difficulty of colleges and universities based on user information and intention information, and recommend colleges and universities that meet the requirements and have preparation guidance significance;

[0036] Among them, the data entry system is used to: input school information, major information and annual recruitment data, etc.

[0037] The standardized fields recognized by speech include current status, university, university major, university academic ranking, CET-4 and CET-6 scores, and planned preparation time.

[0038] 1. The voice recorded by the user is "Let me think about how to say it. Hmm... I'm ready. I'm currently studying at Huazhong Normal University, majoring in Chinese. I've passed the Level 4 English test and am now preparing for the postgraduate entrance examination next year. My main problem now is that I don't know which school to choose. I want to solve my school selection problem through this platform."

[0039] 2. First, use a third-party voice system to recognize the above voice into text.

[0040] 3. Then, through the self-built school selection corpus, text sentences related to school selection are extracted from the recognized text.

[0041] a. I am currently studying at Huazhong Normal University;

[0042] b. Subject language major;

[0043] c. Have passed CET-4;

[0044] d. Start preparing now and prepare to take the postgraduate entrance examination next year.

[0045] 4. Through the third-party large language model, the keywords in the proposed text sentences are matched with the self-built school selection field library, for example:

[0046] a. Central China Normal University → Central China Normal University;

[0047] b. Subject Language and Literature major → Subject Teaching (Chinese);

[0048] c. Passed CET-4 → Passed CET-4 but not CET-6;

[0049] d. Start preparing now and prepare to take the postgraduate entrance examination next year → more than one year.

[0050] 5. Final output standardized information format

[0051] a.University: Central China Normal University

[0052] b.University major: Subject teaching (Chinese)

[0053] c. CET-4 and CET-6 scores: passed CET-4 but not CET-6

[0054] d. Planned preparation time: more than one year

[0055] The school selection operating system consists of three parts, namely the school selection function entrance, school selection information input and school selection result feedback. The school selection information input includes personal information and intention information. Personal information includes current status, university, university major, CET-4 and CET-6 scores, planned preparation time and mathematical foundation. Intention information includes intended major, intended region, school level and training method.

[0056] The school selection result feedback will eventually give users a school selection result recommendation, that is, a school selection report. The report first completes the corresponding postgraduate entrance examination competitiveness ability decomposition and calculation based on the user's personal information and intention information, and obtains the final postgraduate entrance examination competitiveness level value. At the same time, it gives an analysis of the reasons for the recommendation, and finally labels it as "safe, sprint, and guaranteed" and recommends it to the user.

[0057] The intelligent dialogue system includes user personal information, user intention information, system feedback content, voice recognition content and a voice recognition module, wherein the voice recognition system is used to quickly identify user information based on the user's personal information.

[0058] The method of using the speech recognition module includes the following steps:

[0059] S1: First, the system will give the user an expression template and ask the user to imitate it. Then the user can speak according to the template.

[0060] After receiving this sentence, the S2 system will first convert the speech into text, and then combine it with the school selection corpus to extract the discourse text related to school selection;

[0061] The S3 system then breaks down the extracted speech text into keywords, and then matches the keywords to the standardized field library for school selection, thus finally obtaining the standardized fields related to school selection in the sentence;

[0062] S4 will eventually have standardized fields about user personal information, including "current status, university, university major, university academic ranking, CET-4 and CET-6 scores, and planned preparation time."

[0063] Abstract the [speech unit] in the dialogue school selection

[0064] Category A unit: User’s personal information, including:

[0065] Current status;

[0066] colleges and universities;

[0067] college major;

[0068] University academic rankings;

[0069] CET-4 and CET-6 scores;

[0070] Plan how long to prepare for the exam;

[0071] Mathematical foundations.

[0072] Category B unit: User intention information, including:

[0073] Intended major;

[0074] Intended area;

[0075] Institutional level;

[0076] Training method.

[0077] Unit C: System feedback content, including:

[0078] System opening remarks - user's first use;

[0079] The system replies that personal information is filled in;

[0080] The system replies with the screening results - number of institutions;

[0081] The system replies [options are available and the quantity is between [6,18]];

[0082] The system reply leads to the mathematical foundation;

[0083] The system responds with the landing analysis report;

[0084] The system responds to information adjustments (including personal information and screening information); the system responds to report generation prompts based on information adjustments;

[0085] The system replies to the user to try again after a while;

[0086] The system replied that the screening result was 0;

[0087] The system replies that the screening result for a certain major is 0 (the total is not 0); the system replies that there are no matches for the existing colleges;

[0088] The system replies to the analysis page and returns to the chat area;

[0089] The bottom option first appears to give feedback.

[0090] Class D unit: speech recognition module, including;

[0091] Send a voice opening statement;

[0092] Semantic recognition feedback;

[0093] Embed [speech unit] into [dialogue process design];

[0094] The system first determines the user's current status and then gives corresponding feedback and operation guidance.

[0095] If it is determined that the user is using the app for the first time: the system gives an opening statement, asks the user about their current status, and gives corresponding options to guide the user to the next step.

[0096] Users ask questions through shortcut entries: Based on the shortcut entry clicked by the user, the system gives questions and options corresponding to the shortcut entry, thereby guiding the user to the next step.

[0097] The user continues to use the system after 12 hours: The system greets the user again and sends out the questions and options that were not completed in the last conversation to continue guiding the user to the next step.

[0098] The user returns to the chat interface from the report interface: proactively communicate with the user to check whether they are satisfied with the report content and guide the user's further operations.

[0099] If the user is using it for the first time, the system will give an opening statement, which includes the [Voice Recognition Module]. The user selects this module and enters the voice recognition process.

[0100] The speech recognition unit first sends a voice opening statement to the user, asking the user to send a voice message, with the opening statement being "You can also tell Xiaozhi about your situation in one sentence~".

[0101] The system then sends the user a reference template, and the user can send a voice message based on the content expression of the template. The reference template is "I am a junior student at Henan University, majoring in business management. I have passed the CET-4 but not the CET-6. My grades are usually above average. I am going to take the postgraduate entrance examination in 2024 and plan to give myself six months to a year to prepare for the exam."

[0102] After the user sends the voice, the system will receive the corresponding voice content and convert the voice into text through a third-party voice system.

[0103] After the system obtains the converted text, it extracts school selection-related text from the text through its self-built school selection corpus and eliminates sentences that are not related to school selection.

[0104] After the system obtains the sentence content related to school selection, it extracts keywords from the text through a third-party large language model.

[0105] The system will perform standardized matching on these keywords through its self-built school selection field library and finally obtain standardized fields.

[0106] The final standardized matching of school selection fields "current status, university, university major, university grade ranking, CET-4 and CET-6 scores, and planned preparation time" are fed back to users.

[0107] The system will automatically guide the user to the next step based on the user's current conversation operations to complete the school selection goal.

[0108] Personal information: After the user replies to a piece of his or her personal information, the system will determine which piece of personal information the user is currently replying to, and then continue to ask for the next empty Class A unit option according to the preset order.

[0109] Personal information: If there are no empty options in the user's personal information at this time, the system will automatically guide the user to start replying to the intended information.

[0110] Intention information: The intention information system also has a preset order. The system will actively ask the user for intention information. After the user's feedback, the system will screen the colleges and majors based on the user's existing intention information, and make further judgments based on the screening results.

[0111] Intent information: Based on the results of the system screening, if the number of colleges and majors screened out is still relatively large, users will continue to be guided to further provide feedback and screening actions on their intention information.

[0112] Intent information: Based on the results of the system screening, if the number of selected colleges and majors is small, or the screening items of the intention information have been used up, the user will be guided to view the final school selection report.

[0113] Modified information: When a user replies to his or her personal information in a conversation, if the system finds that the information already exists, it will be judged that this is a modified information, and the system will proactively inform the user of the modified situation and continue to guide the user to view the modified school selection report.

[0114] Modify information: When a user replies to his or her intended information in a conversation, if the system finds that the information already exists, it will be judged that this is a modification of the information, and the system will proactively inform the user of the new screening results after the modification, and continue to guide the user to view the modified school selection report.

[0115] Mathematical foundation: Regarding the user's mathematical foundation information, the system first determines whether the colleges and majors selected based on the intention information require mathematics exams. If mathematics exams are required, the system will actively inquire about the user's mathematical foundation and the user will give corresponding feedback.

[0116] The above completes the architectural design of the entire school selection intelligent dialogue system, and ultimately understands the user's personal information, intention information, and gives a corresponding school selection report.

[0117] The data entry system includes information related to institutions, majors and annual enrollment information.

[0118] Information related to institutions includes the name of the institution, the level of the institution, the ranking of the institution, the city where the institution is located, the type of institution and the information on its subordinate departments, etc. Information related to majors includes subject categories, first-level disciplines, major names, major codes and subject evaluations, etc.

[0119] The matching recommendation system calculates the required value of the competitiveness of each school major in the postgraduate entrance examination based on the obtained information of the school majors. The matching recommendation system divides all the schools into 6 levels, namely A1 to A6, for the school hierarchy information, and then assigns corresponding weight values ​​to different levels. The matching recommendation system divides the professional subject evaluation into 10 levels, namely B1 to B10, and then assigns corresponding weight values ​​to different levels. The matching recommendation system divides the regional cities of the schools into 3 levels, namely C1 to C3, and then assigns corresponding weight values ​​to different levels. The matching recommendation system divides the training methods into 2 levels, namely D1 and D2, and then assigns corresponding weight values ​​to different levels. The matching system recommends the school ranking information, and will substitute the specific ranking value into a linear algorithm formula to obtain the ranking value E.

[0120] The linear algorithm formula of the ranking value E is:

[0121] E=5.008772-school ranking×0.008772 (when the school ranking is greater than 570, E=0). The calculation formula for the final difficulty of admission to graduate school majors is: M=A+B+C+D+E+50.

[0122] The matching recommendation system calculates the user's postgraduate entrance examination competitiveness value based on the obtained user personal information. The matching recommendation system divides all the user's colleges into 7 levels, namely X1 to X7, and then assigns corresponding weight values ​​to different levels. The matching recommendation system divides the user's academic level into 4 levels, namely Y1 to Y4, and then assigns corresponding weight values ​​to different levels. The matching recommendation system divides the CET-4 and CET-6 scores into 3 levels, namely P1 to P3, and then assigns corresponding weight values ​​to different levels. The matching recommendation system divides the user's preparation time into 3 levels, namely Q1 to Q3, and then assigns corresponding weight values ​​to different levels. The matching recommendation system divides the user's mathematical foundation into 6 levels, namely T1 to T6, and then assigns corresponding weight values ​​to different levels.

[0123] The matching recommendation system obtains the calculation formula for each item of XYPQT:

[0124]

[0125] The final level of the test-taking user is: N = X + Y + P + Q + T + 50.

[0126] Then subtract the user's postgraduate entrance examination competitiveness value N from the postgraduate entrance examination competitiveness requirement value M of each college's major to obtain R. If R∈[-7, 7], the matching degree is five stars and the recommended level is safe. If R∈(7, 15], the matching degree is four stars and the recommended level is sprint. If R∈[-15, -7), the matching degree is three stars and the recommended level is minimum. If R∈(-∞, -15), the matching degree is two stars and the recommended level is not recommended. If R∈(15, +∞), the matching degree is one star and the recommended level is not recommended.

[0127] Competitiveness requirements for college majors:

[0128] A value: Ranking of graduate schools, divided into 6 levels, with weights of 2, 4, 5, 6, 8, and 10 respectively.

[0129] B value: Subject evaluation, divided into 10 levels, with weight values ​​of 0, 6, 6.5, 7, 7.5, 8, 8.5, 9, 9.5, 10.

[0130] C value: region, divided into 3 levels, with weight values ​​of 0, 7, and 10 respectively.

[0131] D value: training method, divided into 2 levels, with weight values ​​of 0 and 10 respectively.

[0132] E value: 5.008772-school ranking*0.008772 (when the school ranking is greater than 570, E=0).

[0133] Competitiveness level of the test taker:

[0134] X value: colleges and universities, divided into 7 levels, with weight values ​​of 0, 2, 4, 5, 6, 8, and 10 respectively.

[0135] Y value: performance level, divided into 4 levels, with weight values ​​of 0, 3, 7, and 10 respectively.

[0136] P value: CET-4 and CET-6 scores, divided into 3 levels, with weights of 0, 5, and 10 respectively.

[0137] Q value: preparation time, divided into 3 levels, with weights of 0, 6, and 10 respectively.

[0138] T value: mathematical foundation, divided into 6 levels, with weight values ​​0, 6, 8, 3, 7, and 10 respectively.

[0139] Allocation method 1: A mathematics test is required, and the test is "Mathematics 1 or Mathematics 2". The user does not study mathematics. In this case, XYPQT will allocate the corresponding score.

[0140] X: 50%, Y: 10%, Q: 10%, T: 20%, P: 10%, the full score is 50, X=25, Y=5, P=5, Q=10, T=5.

[0141] Allocation method 2: Mathematics is required and the test is "Mathematics 3", but the user does not study mathematics. In this case, XYPQT will allocate the corresponding score.

[0142] X: 50%, Y: 12%, Q: 10%, T: 16%, P: 12%, the full score is 50, X=25, Y=6, P=5, Q=8, T=6.

[0143] Allocation method three: Mathematics is required, the user has studied mathematics, and the user is not taking a cross-examination. In this case, XYPQT will allocate the corresponding scores.

[0144] X: 50%, Y: 15%, Q: 10%, T: 10%, P: 15%, the full score is 50, X=25, Y=7.5, P=5, Q=5, T=7.5.

[0145] Allocation method four: A mathematics exam is required, the user studies mathematics, and the user is taking a cross-exam. At this time, XYPQT will allocate the corresponding scores.

[0146] X: 50%, Y: 20%, Q: 8%, T: 8%, P: 14%, the full score is 50, X=25, Y=10, P=4, Q=4, T=7.

[0147] Allocation method five: Mathematics does not need to be taken and the user is taking a cross-examination. In this case, XYPQT will allocate the corresponding scores.

[0148] X: 50%, Y: 18%, Q: 14%, P: 18%, the full score is 50, X=25, Y=9, P=7, Q=9, T=0.

[0149] Allocation method six: Mathematics does not need to be taken, and the user is not taking a cross-examination. In this case, XYPQT will allocate the corresponding scores.

[0150] X: 50%, Y: 24%, Q: 10%, P: 16%, the full score is 50, X=25, Y=12, P=5, Q=8, T=0.

[0151] Specific implementation examples:

[0152] 1) User information input

[0153] University: xx University, University major: 040101-Education, School performance: ranked 10% to 50%, CET-4 and CET-6 scores: passed CET-4 but not CET-6, Planned preparation time: more than one year, Intended major: Subject teaching (Chinese), Intended region: Beijing, Tianjin, University level: 211 and above, Training mode: full-time.

[0154] 2) User's competitiveness level

[0155] If you do not need to take the math test and the user is not taking a cross-examination, XYPQT will assign the corresponding score:

[0156] X: 50%, Y: 24%, Q: 10%, P: 16%, the full score is 50,

[0157] X=25, Y=12, P=5, Q=8, T=0.

[0158] Then for the above test preparation user n, the values ​​of Xn, Yn, Pn, Qn, and Tn are

[0159] The weight value of Xn=(X / 10)*Xn, the weight value of Yn=(Y / 10)*Yn, the weight value of Pn=(P / 10)*Pn, the weight value of Qn=(Q / 10)*Qn, and the weight value of Tn=(T / 10)*Tn.

[0160] For this user, the weight of Xn = 6; the weight of Yn = 7; the weight of Pn = 5; the weight of Qn = 10. Substitute them into the formula to calculate

[0161] Xn=15, Yn=8.4, Pn=2.5, Qn=8, Tn=0. The final competitiveness level of the user n is: Nn=Xn+Yn+Pn+Qn+Tn+50=83.9.

[0162] 3) School selection

[0163] Based on the "intended major, intended region, school level, and training method", and combined with the annual recruitment data already existing in the database, a total of three school majors were selected, namely xxx University-School of Education-Subject Teaching (Chinese), xxxx University-Faculty of Education-Subject Teaching (Chinese) and xxxx University-Faculty of Literature-Subject Teaching (Chinese).

[0164] 4) Competitiveness requirements of college majors

[0165] A: 55.5%, B: 11.5%, C: 6%, D: 17%, the full score is 50,

[0166] A=27.75, B=5.75, C=3, D=8.5.

[0167] E value: 5.008772-school ranking*0.008772 (when the school ranking is greater than 570, E=0).

[0168] Then for a specific college major m, the values ​​of Am, Bm, Cm, Dm, and Em are:

[0169] Am=weight value of (A / 10)*Am, Bm=weight value of (B / 10)*Bm, Cm=weight value of (C / 10)*Cm, Dm=weight value of (D / 10)*Dm, Em=5.008772-school ranking*0.008772 (when the school ranking is greater than 570, E=0).

[0170] xxx University-School of Education-Subject Teaching (Chinese)

[0171] The weight value of Am = 8, the weight value of Bm = 7, the weight value of Cm = 10, the weight value of Dm = 10, Am = 22.2, Bm = 4.025, Cm = 3, Dm = 8.5, Em = 3.877184, the competitiveness requirement value of the institution's major m is: Mm = Am + Bm + Cm + Dm + Em + 50 = 91.6.

[0172] xxxx University - Faculty of Education - Subject Teaching (Chinese)

[0173] The weight value of Am = 8, the weight value of Bm = 10, the weight value of Cm = 10, the weight value of Dm = 10, Am = 22.2, Bm = 5.75, Cm = 3, Dm = 8.5, Em = 4.82104, the competitiveness requirement value of the institution's major m is: Mm = Am + Bm + Cm + Dm + Em + 50 = 94.29.

[0174] xxxx University - Faculty of Literature - Subject Teaching (Chinese)

[0175] The weight value of Am = 8, the weight value of Bm = 10, the weight value of Cm = 10, the weight value of Dm = 10, Am = 22.2, Bm = 5.75, Cm = 3, Dm = 8.5, Em = 4.82104, the competitiveness requirement value of the institution's major m is: Mm = Am + Bm + Cm + Dm + Em + 50 = 94.29.

[0176] 5) Recommended level matching

[0177] R = Mm - Nn, where Nn = 83.9.

[0178] xxx University-School of Education-Subject Teaching (Chinese)

[0179] R=Mm-Nn=91.6-83.9=7.7, R∈(7,15], then the matching degree is four stars, and the recommended level is sprint.

[0180] xxxx University - Faculty of Education - Subject Teaching (Chinese)

[0181] R=Mm-Nn=94.29-83.9=10.39, R∈(7, 15], then the matching degree is four stars, and the recommended level is sprint.

[0182] xxxx University - Faculty of Literature - Subject Teaching (Chinese)

[0183] R=Mm-Nn=94.29-83.9=10.39, R∈(7, 15], then the matching degree is four stars, and the recommended level is sprint.

[0184] The technical solution of this application has the following beneficial effects:

[0185] Users have a clearer understanding of the school selection process and thus have more trust in the school selection results. By adopting a human-computer dialogue interaction method, each dialogue will provide timely feedback and inform the user of specific information, so as to gradually obtain the user's basic information and user intention information, thereby ensuring that the user has a clear and transparent understanding of the entire school selection process, significantly improving the accuracy of the school selection results, and ensuring user satisfaction with the results. The basic information fields of colleges and majors are assigned corresponding parameters, and the corresponding competitiveness requirements of each college and major are calculated. Then, corresponding weight values ​​are assigned to each characteristic information of the user, and the postgraduate entrance examination competitiveness level of each candidate is calculated. Finally, the matching degree is calculated based on the candidate's postgraduate entrance examination competitiveness and the competitiveness requirements of the college and major, and a number of college and major school selection recommendations are given to students.

[0186] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0187] While the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that many changes, modifications, substitutions and variations can be made to the embodiments without departing from the principles and spirit of the invention.

Claims

1. An intelligent school selection system based on human-computer dialogue interaction data, characterized in that: Including school selection operating system, intelligent dialogue system, matching recommendation system and data entry system; Among them, the school selection operating system is used to: allow users to input and modify personal information and intention information, and provide information to the matching recommendation system, and finally receive and display the school selection result report information fed back by it; Among them, the intelligent dialogue system is used to: identify and analyze the user's natural language, and ultimately understand the user's personal information and intention information; Among them, the matching recommendation system is used to: calculate and match the difficulty of colleges and universities based on user information and intention information, and recommend colleges and universities that meet the requirements and have guidance for exam preparation; Among them, the data entry system is used to: input school information, major information and annual recruitment data, etc.

2. The intelligent school selection system based on human-computer dialogue interaction data according to claim 1 is characterized in that: The school selection operating system consists of three parts, namely, school selection function entrance, school selection information input and school selection result feedback. The school selection information input includes personal information and intention information. The personal information includes current status, university, university major, CET-4 and CET-6 scores, planned preparation time and mathematical foundation. The intention information includes intended major, intended region, school level and training method.

3. The intelligent school selection system based on human-computer dialogue interaction data according to claim 2 is characterized in that: The school selection result feedback will eventually give the user a school selection result recommendation, that is, a school selection report. The report first completes the corresponding postgraduate entrance examination competitiveness ability decomposition and calculation based on the user's personal information and intention information, and obtains the final postgraduate entrance examination competitiveness level value. At the same time, it gives an analysis of the reasons for the recommendation, and finally labels it as "safe, sprint, and guaranteed" and recommends it to the user.

4. The intelligent school selection system based on human-computer dialogue interaction data according to claim 1 is characterized in that: The intelligent dialogue system includes dialogue triggering logic, dialogue judgment flow, system feedback speech, and a speech recognition module, wherein the speech recognition module is used to quickly identify user information based on the user's personal information.

5. The intelligent school selection system based on human-computer dialogue interaction data according to claim 1 is characterized in that: The data entry system includes information related to institutions, information related to majors and annual enrollment information.

6. The intelligent school selection system based on human-computer dialogue interaction data according to claim 5 is characterized in that: The relevant information about the institution includes the name of the institution, the level of the institution, the ranking of the institution, the city where the institution is located, the type of institution and the information of its subordinate departments, etc. The relevant information about the major includes the subject category, first-level discipline, major name, major code and subject evaluation, etc.

7. The intelligent school selection system based on human-computer dialogue interaction data according to claim 1 is characterized in that: The matching recommendation system calculates the required competitiveness value of each school major for postgraduate entrance examination based on the obtained school major information. The matching recommendation system divides all schools into 6 levels, namely A1~A6, for school hierarchy information, and then assigns corresponding weight values ​​to different levels. The matching recommendation system divides professional subject evaluation into 10 levels, namely B1~B10, for professional subject evaluation, and then assigns corresponding weight values ​​to different levels. The matching recommendation system divides the regional cities of schools into 3 levels, namely C1~C3, for the regional cities, and then assigns corresponding weight values ​​to different levels. The matching recommendation system divides the training methods into 2 levels, namely D1 and D2, for professional training methods, and then assigns corresponding weight values ​​to different levels. The matching system recommends school ranking information and substitutes the specific ranking value into a linear algorithm formula to obtain the ranking value E. The matching recommendation system obtains the calculation formula for each ABCD item as follows:

8. The intelligent school selection system based on human-computer dialogue interaction data according to claim 7 is characterized in that: The linear algorithm formula of the ranking value E is: E=5.008772-school ranking×0.008772 (when the school ranking is greater than 570, E=0). The calculation formula for the final difficulty of admission to graduate school majors is: M=A+B+C+D+E+50.

9. The intelligent school selection system based on human-computer dialogue interaction data according to claim 1 is characterized in that: The matching recommendation system calculates the user's postgraduate entrance examination competitiveness value based on the acquired user personal information. The matching recommendation system divides all the user's colleges into 7 levels, namely X1~X7, and then assigns corresponding weight values ​​to different levels. The matching recommendation system divides the user's academic level into 4 levels, namely Y1~Y4, and then assigns corresponding weight values ​​to different levels. The matching recommendation system divides the CET-4 and CET-6 scores into 3 levels, namely P1~P3, and then assigns corresponding weight values ​​to different levels. The matching recommendation system divides the user's preparation time into 3 levels, namely Q1~Q3, and then assigns corresponding weight values ​​to different levels. The matching recommendation system divides the user's mathematical foundation into 6 levels, namely T1~T6, and then assigns corresponding weight values ​​to different levels.

10. The intelligent school selection system based on human-computer dialogue interaction data according to claim 9 is characterized in that: The matching recommendation system obtains the calculation formula for each item of XYPQT: The final level of the test-taking user is: N = X + Y + P + Q + T + 50. Then subtract the user's postgraduate entrance examination competitiveness value N from the postgraduate entrance examination competitiveness requirement value M of each college's major to obtain R. If R∈[-7, 7], the matching degree is five stars and the recommended level is safe. If R∈(7, 15], the matching degree is four stars and the recommended level is sprint. If R∈[-15, -7), the matching degree is three stars and the recommended level is minimum. If R∈(-∞, -15), the matching degree is two stars and the recommended level is not recommended. If R∈(15, +∞), the matching degree is one star and the recommended level is not recommended.

Citation Information

Patent Citations

  • Information screening method and system of assisting to select schools

    CN106875028A

  • Method for generating application information and electronic equipment

    CN109658304A

  • Postgraduate-entrance examination assisted school selection method and system

    CN112016827A

  • School and major recommendation method and system

    WO2020001031A1