Intelligent matching method and system for teachers and students
Through knowledge graph technology, the intelligent matching system improves the matching efficiency of teachers and students, solves the problems of boring information display and low matching efficiency in the traditional dual-select system, and achieves a more accurate and efficient matching process.
Patent Information
- Application Number
- CN202411010632.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-26
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2044-07-26
AI Technical Summary
The traditional teacher-student dual selection system assists dual selection through data list form, resulting in boring information display and easy to miss important information, reducing the matching efficiency between teachers and students.
Using an intelligent matching method based on knowledge graph, the administrator imports the target user information, determines the matching requester, obtains the matcher recommendation list, and presents the matcher matcher based on the matching preferences, and intelligently sends the matching request information to establish a matching relationship.
It improves the matching efficiency of teachers and students, ensures that the matching process is more in line with user preferences, reduces the risk of information missed, and improves the accuracy and efficiency of matching.
Smart Images

Figure CN118939878A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a teacher-student intelligent matching method and system. Background Art
[0002] The importance of tutors to students' entire learning life is very large. A suitable tutor can enable students to learn a lot of professional knowledge. Therefore, fully understanding the tutor's information when choosing a tutor will help students develop better in their learning life. Most common teacher-student double-selection systems use data lists to assist double selection. Although this double-selection form greatly improves the intuitiveness of data viewing, boring data often easily makes people tired, and there is a high probability of missing important information. Because human memory is biased towards association, the traditional data display form shows limitations.
[0003] The invention patent with application number: CN202310604119.9 discloses an Internet education and teaching interactive management system based on data processing. The teacher terminal is used for teacher users to submit personal information for registration and login, and send the registration information to the server for storage; the student terminal is used for student users to submit personal information for registration and login, and send the registration information to the server for storage, and the server obtains the browsing information of the student terminal; the server analyzes the registration information of the teacher terminal to obtain the energy efficiency value of the online teacher, and the server analyzes the browsing information of the student terminal to obtain the recommended value of the online teacher, and weights the energy efficiency value of the online teacher and the recommended value of the online teacher. The above invention helps the student terminal to match the required online teacher more accurately in the server, saves the time of the student terminal in searching for teachers online, and enables the student terminal and the teacher terminal to find the most suitable match object more quickly.
[0004] However, the above-mentioned prior art assists double selection in the form of a data list, which can easily cause fatigue and miss important information, thereby reducing the matching efficiency between teachers and students.
[0005] In view of this, there is an urgent need for intelligent matching methods and systems between teachers and students to at least solve the above-mentioned deficiencies. Summary of the invention
[0006] One of the purposes of the present invention is to provide a method and system for intelligent matching of teachers and students, and to determine the matching requester who logs in to the matching page according to the target user information imported by the administrator. Based on the knowledge graph to view information, according to the matching preference of the matching requester, a recommended list of matches is obtained, and the more in line with the matching preference, the more priority the corresponding to-be-matched person is presented, and the intended matcher selected by the matching requester after viewing the recommended list of matches is obtained, and matching request information is sent to the intended matcher, and when the intended matcher accepts the matching request, a corresponding matching relationship is established, thereby improving the matching efficiency of teachers and students.
[0007] The teacher-student intelligent matching method provided by the embodiment of the present invention includes:
[0008] Step 1: Import target user information;
[0009] Step 2: Determine the matching requester based on the target user information;
[0010] Step 3: View information based on the knowledge graph and obtain a list of recommended matches according to the matching preferences of the matching requester;
[0011] Step 4: After the matching requester views the matching recommendation list, the matching requester's intended matching persons are obtained;
[0012] Step 5: Intelligently send a matching request message to the intended matcher. When the intended matcher accepts the matching request, a matching relationship between the intended matcher and the matching requester is established to complete the intelligent matching.
[0013] Preferably, step 1: importing target user information includes:
[0014] Get the teacher and student information of the preset import template entered by the administrator;
[0015] When the template is imported successfully, the import of the target user information is completed.
[0016] Preferably, step 2: determining a matching requester according to target user information includes:
[0017] According to the target user information, verify the user to be verified who logs in to the matching page, and obtain the matching requester who has passed the verification.
[0018] Preferably, step 3: viewing information based on the knowledge graph and obtaining a list of recommended matches according to the matching preferences of the matching requester includes:
[0019] View information based on the knowledge graph to determine the area of interest of the first knowledge graph;
[0020] Determine, based on the first region label of the region of interest of the first knowledge graph, an associated graph region that has not been viewed by the matching requester;
[0021] Take the first knowledge graph region of interest and the associated graph region together as the second knowledge graph region of interest, and determine the first graph subject of the second knowledge graph region of interest;
[0022] Get the system information of the matching requester;
[0023] Determine a second region label of a region of interest of the second knowledge graph corresponding to each first graph subject;
[0024] Determine the first type of coincidence according to the information type of the system information and the second area label;
[0025] If the first type overlap is greater than or equal to a preset first type overlap threshold, the second knowledge graph interest region corresponding to the first graph subject is used as the third knowledge graph interest region;
[0026] Determine the subject overlap items and subject difference items between the subjects of the first graph according to the region of interest of the third knowledge graph;
[0027] Display subject overlap items and subject difference items;
[0028] Generate a matcher recommendation list based on subject overlap items, subject difference items and system information.
[0029] Preferably, determining the associated graph area that has not been viewed by the matching requester according to the first area label of the first knowledge graph area of interest includes:
[0030] Perform label matching on the first region label and the standard region label, and if there is a match, obtain the content difference of the graph node content corresponding to the first region label of different target entities;
[0031] If the content difference is greater than or equal to the preset content difference threshold, the expression changes of the matching requester viewing the content of different graph nodes are obtained;
[0032] Determine the required graph node content based on the changing characteristics of the expression;
[0033] If the content difference is less than a preset content difference threshold, extract the expression feature of the matching requester viewing the content of the graph node;
[0034] Determine the content adjustment strategy of the graph node content based on the expression characteristics;
[0035] Determine the required graph node content based on the content adjustment strategy and graph node content;
[0036] Perform label matching on the target label and the entity label of the unviewed entity. When both label matches are matched, compare the content similarity between the required graph node content of the target label that matches the label and the label content of the entity label.
[0037] If each content similarity is greater than or equal to a preset content similarity threshold, the knowledge graph of the corresponding unviewed entity is used as the associated graph area.
[0038] Preferably, displaying the subject overlap items and subject difference items includes:
[0039] According to the subject overlap items, determine the overlapped subjects;
[0040] Determine the first graph node of the subject overlap item in the region of interest of the third knowledge graph, and use the first graph subject corresponding to the subject overlap item as the second graph subject;
[0041] Merge and display the first graph nodes of the second graph body;
[0042] Get the second graph node of the second graph body;
[0043] Display the subject difference items corresponding to the second graph node.
[0044] Preferably, a matcher recommendation list is generated based on subject overlap items, subject difference items and system information, including:
[0045] Determine the second type of overlap between the second graph subject and the matching requester according to the subject overlap item and the system information, and use the second type of overlap as the first recommendation value, and associate it with the second graph subject;
[0046] Obtaining the visual dwell time of the subject difference item, determining a second recommended value according to the visual dwell time and the total visual dwell time of the subject difference item, and associating it with the second atlas subject;
[0047] Obtaining viewing information of the second graph node, and determining a viewing feature vector according to the viewing information;
[0048] Obtaining a target viewing feature vector space; the target viewing feature vector space includes: a target viewing feature vector when an active matching party among matching persons who have a matching relationship in the past views the information to be matched by the matched party;
[0049] Matching the viewing feature vector with the target viewing feature vector in the target viewing feature vector space, determining the vector angle cosine corresponding to the vector angle with the smallest vector angle, and taking it as the third recommended value, and associating it with the second atlas body;
[0050] The first recommendation value, the second recommendation value and the third recommendation value corresponding to the second atlas body are accumulated to obtain a recommendation index, and the corresponding second atlas body is presented in a preset display list in descending order of the recommendation index.
[0051] The teacher-student intelligent matching system provided by the embodiment of the present invention includes:
[0052] Target user information import subsystem, used to import target user information;
[0053] A matching requester determination subsystem is used to determine a matching requester based on target user information;
[0054] The matcher recommendation list acquisition subsystem is used to view information based on the knowledge graph and obtain the matcher recommendation list according to the matching preference of the matching requester;
[0055] The potential match acquisition subsystem is used to acquire the potential match of the matching requester after the matching requester has viewed the matching recommendation list;
[0056] The intelligent matching subsystem is used to intelligently send matching request information to the intended matcher. When the intended matcher accepts the matching request, a matching relationship between the intended matcher and the matching requester is established to complete the intelligent matching.
[0057] Preferably, the target user information import subsystem performs the following operations:
[0058] Get the teacher and student information of the preset import template entered by the administrator;
[0059] When the template is imported successfully, the import of the target user information is completed.
[0060] Preferably, the matching requester determines that the subsystem performs the following operations:
[0061] According to the target user information, verify the user to be verified who logs in to the matching page, and obtain the matching requester who has passed the verification.
[0062] Preferably, the matcher recommendation list acquisition subsystem performs the following operations:
[0063] View information based on the knowledge graph to determine the area of interest of the first knowledge graph;
[0064] Determine, based on the first region label of the region of interest of the first knowledge graph, an associated graph region that has not been viewed by the matching requester;
[0065] Take the first knowledge graph region of interest and the associated graph region together as the second knowledge graph region of interest, and determine the first graph subject of the second knowledge graph region of interest;
[0066] Get the system information of the matching requester;
[0067] Determine a second region label of a region of interest of the second knowledge graph corresponding to each first graph subject;
[0068] Determine the first type of coincidence according to the information type of the system information and the second area label;
[0069] If the first type overlap is greater than or equal to a preset first type overlap threshold, the second knowledge graph interest region corresponding to the first graph subject is used as the third knowledge graph interest region;
[0070] Determine the subject overlap items and subject difference items between the subjects of the first graph according to the region of interest of the third knowledge graph;
[0071] Display subject overlap items and subject difference items;
[0072] Generate a matcher recommendation list based on subject overlap items, subject difference items and system information.
[0073] The beneficial effects of the present invention are:
[0074] The present invention determines the matching requester who logs in to the matching page according to the target user information imported by the administrator. Based on the knowledge graph, the information is viewed and the matching requester's matching preference is used to obtain a matching recommendation list. The more the matching preference is met, the more priority the corresponding matching person will be presented. The intended matching person selected by the matching requester after viewing the matching recommendation list is obtained, and the matching request information is sent to the intended matching person. When the intended matching person accepts the matching request, a corresponding matching relationship is established, thereby improving the matching efficiency of teachers and students.
[0075] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures specifically pointed out in the present application documents.
[0076] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS
[0077] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:
[0078] Figure 1 Schematic diagram of a teacher-student intelligent matching method according to an embodiment of the present invention;
[0079] Figure 2 Schematic diagram of the teacher-student intelligent matching system in an embodiment of the present invention. DETAILED DESCRIPTION
[0080] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.
[0081] The embodiment of the present invention provides a method for intelligent matching of teachers and students. Figure 1 As shown, including:
[0082] Step 1: Import target user information; the target user information is the information of the person who needs to be matched, such as the research direction and experimental project of the graduate student's tutor, the undergraduate major and honors of the graduate student, etc. The target user information is imported by the administrator;
[0083] Step 2: Determine the matching requester based on the target user information; the matching requester is the person who issues the matching request, such as a student who has logged into the tutor-graduate double selection system and entered information into the system;
[0084] Step 3: Based on the knowledge graph viewing information, according to the matching preference of the matching requester, obtain a list of recommended matches; wherein the knowledge graph viewing information is: the viewing process information of the knowledge graph of the person to be matched by the matching requester, such as: the number of clicks of a certain knowledge graph node by the matching requester; the knowledge graph is: a structured display of the target person's information; the matching preference is: the bias factor of the matching requester when selecting the matching object; the list of recommended matches is: a list of matching objects sorted in descending order according to the degree of compliance with the matching preference;
[0085] Step 4: After the matching requester views the matching person recommendation list, the matching requester's intended matching person is obtained; wherein the intended matching person is: the person selected by the matching requester in the matching person recommendation list for matching;
[0086] Step 5: Intelligently send the matching request information to the intended matcher. When the intended matcher accepts the matching request, a matching relationship between the intended matcher and the matching requester is established to complete the intelligent matching. The matching request information includes: the information of the person who initiated the matching request and the matching requirements.
[0087] The working principle and beneficial effects of the above technical solution are:
[0088] The present invention determines the matching requester who logs in to the matching page according to the target user information imported by the administrator. Based on the knowledge graph, the information is viewed and the matching requester's matching preference is used to obtain a matching recommendation list. The more the matching preference is met, the more priority the corresponding matching person will be presented. The intended matching person selected by the matching requester after viewing the matching recommendation list is obtained, and the matching request information is sent to the intended matching person. When the intended matching person accepts the matching request, a corresponding matching relationship is established, thereby improving the matching efficiency of teachers and students.
[0089] In one embodiment, step 1: importing target user information includes:
[0090] Get the teacher and student information of the preset import template entered by the administrator;
[0091] When the template is imported successfully, the import of the target user information is completed.
[0092] The working principle and beneficial effects of the above technical solution are:
[0093] Importing is when the administrator imports the basic information of tutors and students in batches through Excel after the system is built. Exporting is when the administrator or responsible professor exports the information of certain students or tutors in batches in Excel format. This embodiment realizes the conversion between English fields in the table and Chinese in Excel headers by establishing a field table, thereby improving the speed of information importing.
[0094] In one embodiment, step 2: determining a matching requester according to target user information includes:
[0095] According to the target user information, verify the user to be verified who logs in to the matching page, and obtain the matching requester who has passed the verification.
[0096] The working principle and beneficial effects of the above technical solution are:
[0097] After the user logs in, when jumping to web pages and obtaining information, he can only enter pages that match his own permissions or use functions that match his permissions, and some important data of the user will be stored in the session of the server. Cookies only store the session ID. When the system needs some important data of the user, it directly calls the value stored in the session through the session ID, avoiding the injection of data into the outside world, thereby improving the security of the system.
[0098] In one embodiment, step 3: viewing information based on the knowledge graph and obtaining a list of recommended matches according to the matching preferences of the matching requester includes:
[0099] Determine the first knowledge graph interest area according to the knowledge graph viewing information; wherein the first knowledge graph interest area is: the knowledge graph of the person that the matching requester wants to match and views;
[0100] Determine the associated graph area that has not been viewed by the matching requester according to the first area label of the first knowledge graph area of interest; wherein, when determining the associated graph area that has not been viewed by the matching requester according to the first area label of the first knowledge graph area of interest, when determining the associated graph area according to the first area label, determine the expression information of the matching requester when viewing the current graph node content of the first area label, determine the graph node content required by the matching requester according to the expression information, and determine the knowledge graph interest area of the unviewed entity whose graph node content corresponding to the unviewed entity meets the required graph node content as the associated graph area;
[0101] The first knowledge graph region of interest and the associated graph region are taken together as the second knowledge graph region of interest, and the first graph subject of the second knowledge graph region of interest is determined; wherein the first graph subject is: the construction entity of the knowledge graph corresponding to the second knowledge graph region of interest, such as: tutors A and B;
[0102] Obtaining system information of the matching requester; wherein the system information is: target user information corresponding to the matching requester;
[0103] Determine a second region label of the second knowledge graph interest region corresponding to each first graph subject; wherein the second region label is: the knowledge graph node type of the second knowledge graph interest region corresponding to each first graph subject;
[0104] According to the information type of the system information and the second area label, the first type coincidence degree is determined; wherein the first type coincidence degree is: the suitability of the system information of the information type and the label content of the second area label corresponding to the information type, for example: the information type is: student major, and the corresponding second area label is: research direction; the student major is: communication engineering, and the research direction is: long-distance wireless communication, and the first type coincidence degree is: 95%;
[0105] If the first type overlap is greater than or equal to a preset first type overlap threshold, the second knowledge graph interest region corresponding to the first graph body is used as the third knowledge graph interest region; wherein the preset first type overlap threshold is manually preset;
[0106] According to the third knowledge graph interest region, determine the subject overlap items and subject difference items between the first graph subjects; wherein the subject overlap items are: the parts of the knowledge graph nodes corresponding to the third knowledge graph interest region with the same node content; the subject difference items are: the parts of the knowledge graph nodes corresponding to the third knowledge graph interest region with different node content;
[0107] Display subject overlap items and subject difference items;
[0108] Generate a matcher recommendation list based on subject overlap items, subject difference items and system information.
[0109] The working principle and beneficial effects of the above technical solution are:
[0110] The present invention obtains the viewing process information of the knowledge graph of the person to be matched by the matching requester, and determines the first knowledge graph interest area corresponding to the knowledge graph of the viewing entity that the matching requester has viewed according to the viewing process information. Determine the first area label of the first knowledge graph interest area, and filter out the associated graph area. The first knowledge graph interest area and the associated graph area are used together as the second knowledge graph interest area. According to the system information of the matching requester and the second area label of each first graph subject, determine the first type of overlap, the first type of overlap represents the suitability of the system information of the information type and the label content of the second area label corresponding to the information type, if appropriate, filter the third knowledge graph interest area corresponding to the first graph subject. Determine the subject overlap items and subject difference items between the first graph subjects corresponding to the third knowledge graph interest area. Highlight the subject difference items, and generate a matcher recommendation list based on the subject overlap items, subject difference items and system information, thereby improving the suitability of the generation of the matcher recommendation list.
[0111] In one embodiment, determining, according to the first region label of the first knowledge graph region of interest, the associated graph region that has not been viewed by the matching requester includes:
[0112] The first region label is matched with the standard region label. If a match exists, the content difference of the graph node content corresponding to the first region label of different target entities is obtained; the standard region label is: the overlapping types of graph node types of the knowledge graphs of the two entities most recently viewed by the matching requester;
[0113] If the content difference is greater than or equal to a preset content difference threshold, the expression change of the matching requester when viewing the content of different graph nodes is obtained; wherein the preset content difference threshold is preset manually, such as: 0.5; the expression change is: the expression change information when viewing the graph node content of the first area label of different target entities, for example: the expression change when viewing the research direction of tutor A "wireless communication" to viewing the research direction of tutor B "optical fiber communication" is: "frown" to "brows relaxed";
[0114] According to the change characteristics of the expression change, the required graph node content is determined; wherein the change characteristics are: from which orientation the expression changes to which orientation the expression changes, for example: frowning to frowning means changing from an expression with an orientation of "rejection" to an expression with an orientation of "acceptance", and the corresponding required graph node content is: "optical fiber communication";
[0115] If the content difference is less than a preset content difference threshold, extract the facial expression feature of the matching requester viewing the content of the graph node; wherein the facial expression feature is: a characteristic representation of the facial expression information of the matching requester, such as "frown", "pout", etc.;
[0116] Determine the content adjustment strategy of the graph node content according to the facial expression features; wherein, when determining the content adjustment strategy of the graph node content according to the facial expression features, when the facial expression features indicate "rejection", the content adjustment strategy is: adjust in the direction where the graph node content is different, for example: when the graph node content is "wireless communication" and the facial expression features indicate "rejection", the required graph node content is "wired communication"; when the facial expression features indicate "acceptance", the content adjustment strategy is: adjust in the direction where the graph node content is the same, for example: when the graph node content is "wireless communication" and the facial expression features indicate "acceptance", the required graph node content is "wireless communication";
[0117] Determine the required graph node content based on the content adjustment strategy and graph node content;
[0118] Perform label matching on the target label and the entity label of the unviewed entity. When both label matches are matched, compare the content similarity between the required graph node content of the target label that matches the label and the label content of the entity label.
[0119] If each content similarity is greater than or equal to a preset content similarity threshold, the knowledge graph of the corresponding unviewed entity is used as the associated graph area. The preset content similarity threshold is manually preset, for example, 0.9.
[0120] The working principle and beneficial effects of the above technical solution are:
[0121] When the present invention determines the associated graph area that the matching requester has not viewed, the standard area labels of the overlapping types of graph node types of the knowledge graphs of the two entities that the matching requester has recently viewed are obtained, and the graph node content types that the matching requester focuses on in the first area label are screened out. The content differences of the graph node contents of different target entities corresponding to the standard area labels are obtained. When the content differences are large, the expression changes of the matching requester can provide the content orientation information of the required graph node content, and the required graph node content can be directly determined based on the expression change characteristics. When the content differences are small, the expression features are extracted to first determine the content orientation. According to the different results of the content orientation determination, a content adjustment strategy is formulated based on the current graph node content, and the required graph node content after the content adjustment strategy is applied is obtained, and then the subsequent screening of the associated graph area is performed, which improves the screening efficiency of the associated graph area and is more reasonable.
[0122] In one embodiment, displaying subject overlap items and subject difference items includes:
[0123] According to the subject overlap item, the overlap subject is determined; wherein the overlap subject is: the first atlas subject with the subject overlap item;
[0124] Determine the first graph node of the subject overlap item in the region of interest of the third knowledge graph, and use the first graph subject corresponding to the subject overlap item as the second graph subject;
[0125] Merge and display the first graph nodes of the second graph body;
[0126] Obtain a second graph node of the second graph subject; wherein the second graph node is: a graph node corresponding to a subject difference item of the second graph subject;
[0127] Display the subject difference items corresponding to the second graph node.
[0128] The working principle and beneficial effects of the above technical solution are:
[0129] The present invention determines the overlapping subject corresponding to the subject overlap item, determines the first graph node and the second graph subject corresponding to the subject overlap item, and merges and displays the first graph node of the second graph subject. Determines the second graph node corresponding to the subject difference item of the second graph subject, highlights the subject difference item corresponding to the second graph node, improves the rationality of graph node display, and facilitates the matching requester to view valid information.
[0130] In one embodiment, a matcher recommendation list is generated based on subject overlap items, subject difference items, and system information, including:
[0131] According to the subject overlap item and the system information, determine the second type overlap degree between the second graph subject and the matching requester, and use it as the first recommendation value, and associate it with the second graph subject; wherein the second type overlap degree is: the suitability of the system information and the corresponding subject overlap item;
[0132] Obtain the visual stay time of the subject difference item, determine the second recommended value according to the visual stay time and the total visual stay time of the subject difference item, and associate it with the second atlas subject; wherein the second recommended value is: the visual stay time divided by the total visual stay time;
[0133] Obtain viewing information of the second graph node, and determine a viewing feature vector based on the viewing information; wherein the viewing feature vector is: a description vector describing the viewing information of the second graph node composed of viewing features, and the viewing features are: viewing time of the second graph node, viewing expression, and the number of clicks on the associated graph nodes of the second graph node after viewing the second graph node;
[0134] Obtaining a target viewing feature vector space; the target viewing feature vector space includes: a target viewing feature vector when an active matching party among the matching parties that have a matching relationship in history views the information to be matched of the matched party, and the construction rule of the target viewing feature vector is the same as that of the viewing feature vector;
[0135] Matching the viewing feature vector with the target viewing feature vector in the target viewing feature vector space, determining the vector angle cosine corresponding to the vector angle with the smallest vector angle, and taking it as the third recommended value, and associating it with the second atlas body;
[0136] The first recommendation value, the second recommendation value and the third recommendation value corresponding to the second atlas body are accumulated to obtain a recommendation index, and the corresponding second atlas body is presented in a preset display list in descending order of the recommendation index.
[0137] The working principle and beneficial effects of the above technical solution are:
[0138] The present invention introduces the second type of overlap between the subject overlap item and the system information as the first recommendation value; calculates the second recommendation value according to the visual dwell time of the subject difference item and the total visual dwell time of the subject difference item; obtains the viewing information of the second graph node and constructs the viewing feature vector. The target viewing feature vector space is constructed according to the viewing information of the active matching party among the matchers who have a matching relationship in history when viewing the information to be matched of the matched party, the viewing feature vector is matched with the target viewing feature vector in the target viewing feature vector space, the vector angle cosine corresponding to the vector angle with the smallest vector angle is determined as the third recommendation value, the first recommendation value, the second recommendation value and the third recommendation value associated with the second graph subject are cumulatively calculated to obtain the recommendation index, and the corresponding second graph subject is presented in a preset display list based on the order of the recommendation index from large to small, thereby improving the rationality of the construction of the matcher recommendation list.
[0139] The embodiment of the present invention provides a teacher-student intelligent matching system, such as Figure 2 As shown, including:
[0140] Target user information import subsystem 1, used for importing target user information;
[0141] The matching requester determination subsystem 2 is used to determine the matching requester according to the target user information;
[0142] A matcher recommendation list acquisition subsystem 3 is used to view information based on the knowledge graph and acquire a matcher recommendation list according to the matching preference of the matching requester;
[0143] The potential match acquisition subsystem 4 is used to acquire the potential match of the matching requester after the matching requester views the matching recommendation list;
[0144] The intelligent matching subsystem 5 is used to intelligently send matching request information to the intended matching party. When the intended matching party accepts the matching request, a matching relationship between the intended matching party and the matching requesting party is established to complete the intelligent matching.
[0145] In one embodiment, the target user information import subsystem performs the following operations:
[0146] Get the teacher and student information of the preset import template entered by the administrator;
[0147] When the template is imported successfully, the import of the target user information is completed.
[0148] In one embodiment, the matching requester determination subsystem performs the following operations:
[0149] According to the target user information, verify the user to be verified who logs in to the matching page, and obtain the matching requester who has passed the verification.
[0150] In one embodiment, the matcher recommendation list acquisition subsystem performs the following operations:
[0151] View information based on the knowledge graph to determine the area of interest of the first knowledge graph;
[0152] Determine, based on the first region label of the region of interest of the first knowledge graph, an associated graph region that has not been viewed by the matching requester;
[0153] Take the first knowledge graph region of interest and the associated graph region together as the second knowledge graph region of interest, and determine the first graph subject of the second knowledge graph region of interest;
[0154] Get the system information of the matching requester;
[0155] Determine a second region label of a region of interest of the second knowledge graph corresponding to each first graph subject;
[0156] Determine the first type of coincidence according to the information type of the system information and the second area label;
[0157] If the first type overlap is greater than or equal to a preset first type overlap threshold, the second knowledge graph interest region corresponding to the first graph subject is used as the third knowledge graph interest region;
[0158] Determine the subject overlap items and subject difference items between the subjects of the first graph according to the region of interest of the third knowledge graph;
[0159] Display subject overlap items and subject difference items;
[0160] Generate a matcher recommendation list based on subject overlap items, subject difference items and system information.
[0161] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.
Claims
1. Intelligent matching method between teachers and students, It is characterized in that include: Step 1: Import target user information; Step 2: Determine the matching requester based on the target user information; Step 3: View information based on the knowledge graph and obtain a list of recommended matches according to the matching preferences of the matching requester; Step 4: After the matching requester views the matching recommendation list, the matching requester's intended matching persons are obtained; Step 5: Intelligently send a matching request message to the intended matcher. When the intended matcher accepts the matching request, a matching relationship between the intended matcher and the matching requester is established to complete the intelligent matching.
2. The teacher-student intelligent matching method according to claim 1, It is characterized in that Step 1: Import target user information, including: Get the teacher and student information of the preset import template entered by the administrator; When the template is imported successfully, the import of the target user information is completed.
3. The teacher-student intelligent matching method according to claim 1, It is characterized in that Step 2: Determine the matching requester based on the target user information, including: According to the target user information, verify the user to be verified who logs in to the matching page, and obtain the matching requester who has passed the verification.
4. The teacher-student intelligent matching method according to claim 1, It is characterized in that Step 3: View information based on the knowledge graph and obtain a list of recommended matches based on the matching preferences of the matching requester, including: View information based on the knowledge graph to determine the area of interest of the first knowledge graph; Determine, based on the first region label of the region of interest of the first knowledge graph, an associated graph region that has not been viewed by the matching requester; Take the first knowledge graph region of interest and the associated graph region together as the second knowledge graph region of interest, and determine the first graph subject of the second knowledge graph region of interest; Get the system information of the matching requester; Determine a second region label of a region of interest of the second knowledge graph corresponding to each first graph subject; Determine the first type of coincidence according to the information type of the system information and the second area label; If the first type overlap is greater than or equal to a preset first type overlap threshold, the second knowledge graph interest region corresponding to the first graph subject is used as the third knowledge graph interest region; Determine the subject overlap items and subject difference items between the subjects of the first graph according to the region of interest of the third knowledge graph; Display subject overlap items and subject difference items; Generate a matcher recommendation list based on subject overlap items, subject difference items and system information.
5. The teacher-student intelligent matching method according to claim 4, It is characterized in that Displays subject overlap items and subject difference items, including: According to the subject overlap items, determine the overlapped subjects; Determine the first graph node of the subject overlap item in the region of interest of the third knowledge graph, and use the first graph subject corresponding to the subject overlap item as the second graph subject; Merge and display the first graph nodes of the second graph body; Get the second graph node of the second graph body; Display the subject difference items corresponding to the second graph node.
6. The teacher-student intelligent matching method according to claim 4, It is characterized in that Generate a matcher recommendation list based on subject overlap items, subject difference items, and system information, including: Determine the second type of overlap between the second graph subject and the matching requester according to the subject overlap item and the system information, and use the second type of overlap as the first recommendation value, and associate it with the second graph subject; Obtaining the visual dwell time of the subject difference item, determining a second recommended value according to the visual dwell time and the total visual dwell time of the subject difference item, and associating it with the second atlas subject; Obtaining viewing information of the second graph node, and determining a viewing feature vector according to the viewing information; Obtaining a target viewing feature vector space; the target viewing feature vector space includes: a target viewing feature vector when an active matching party among matching parties that have a matching relationship in the past views the information to be matched of the matched party; Matching the viewing feature vector with the target viewing feature vector in the target viewing feature vector space, determining the vector angle cosine corresponding to the vector angle with the smallest vector angle, and taking it as the third recommended value, and associating it with the second atlas body; The first recommendation value, the second recommendation value and the third recommendation value corresponding to the second atlas body are accumulated to obtain a recommendation index, and the corresponding second atlas body is presented in a preset display list in descending order of the recommendation index.
7. Intelligent matching system for teachers and students, It is characterized in that include: Target user information import subsystem, used to import target user information; A matching requester determination subsystem is used to determine a matching requester based on target user information; The matcher recommendation list acquisition subsystem is used to view information based on the knowledge graph and obtain the matcher recommendation list according to the matching preference of the matching requester; The potential match acquisition subsystem is used to acquire the potential match of the matching requester after the matching requester has viewed the matching recommendation list; The intelligent matching subsystem is used to intelligently send matching request information to the intended matcher. When the intended matcher accepts the matching request, a matching relationship between the intended matcher and the matching requester is established to complete the intelligent matching.
8. The teacher-student intelligent matching system according to claim 7, It is characterized in that The target user information import subsystem performs the following operations: Get the teacher and student information of the preset import template entered by the administrator; When the template is imported successfully, the import of the target user information is completed.
9. The teacher-student intelligent matching system according to claim 7, It is characterized in that The matching requester determination subsystem performs the following operations: According to the target user information, verify the user to be verified who logs in to the matching page, and obtain the matching requester who has passed the verification.
10. The teacher-student intelligent matching system according to claim 7, It is characterized in that The matcher recommendation list acquisition subsystem performs the following operations: View information based on the knowledge graph to determine the area of interest of the first knowledge graph; Determine, based on the first region label of the region of interest of the first knowledge graph, an associated graph region that has not been viewed by the matching requester; Take the first knowledge graph region of interest and the associated graph region together as the second knowledge graph region of interest, and determine the first graph subject of the second knowledge graph region of interest; Get the system information of the matching requester; Determine a second region label of a region of interest of the second knowledge graph corresponding to each first graph subject; Determine the first type of coincidence according to the information type of the system information and the second area label; If the first type overlap is greater than or equal to a preset first type overlap threshold, the second knowledge graph interest region corresponding to the first graph subject is used as the third knowledge graph interest region; Determine the subject overlap items and subject difference items between the subjects of the first graph according to the region of interest of the third knowledge graph; Display subject overlap items and subject difference items; Generate a matcher recommendation list based on subject overlap items, subject difference items and system information.
Citation Information
Patent Citations
Interest collection method and device, computer equipment and storage medium
CN110298245A
Broadcasting and television technology knowledge recommendation method based on user portrait and knowledge graph
CN115168744A
Mentor and graduate double-selection and information management system
CN116739854A
Personalized learning resource recommendation method based on knowledge space representation
CN117216405A
