Dormitory dynamic allocation and management method and system based on large model

Abstract

The invention discloses a dormitory dynamic allocation and management method and system based on a large model, and relates to the technical field of dormitory management, and the method comprises the steps: carrying out the unified registration and maintenance of dormitory building, room and bed information, and completing the management of house resource basic information; basic information, work and rest rules, social preferences and learning behaviors of students are collected, and student feature portraits are constructed through data preprocessing, data integration and feature extraction; a dormitory optimal allocation scheme is automatically generated by comprehensively applying a large language model based on a Transform architecture, a multi-modal AI analysis technology and a multi-target optimization algorithm based on house resource basic information and student feature portraits; three modes of applying for dormitory adjustment by students, batch adjustment by an administrator and automatic recommendation of dormitory adjustment are provided, and adjustment of an optimal dormitory allocation scheme is supported; dormitory management is realized by implementing dormitory daily management, maintenance service management, management data analysis and decision support. The method is used for dynamically optimizing dormitory resource allocation and improving dormitory management efficiency.

Description

Technical Field

[0001] This invention relates to the field of dormitory management technology, specifically a method and system for dynamic allocation and management of dormitories based on a large model. Background Technology

[0002] Currently, the allocation and management of university dormitories largely relies on manual operation or a static allocation model based on basic student information. It does not conduct in-depth analysis of students' living habits, social preferences, and other multi-dimensional characteristics. This makes it impossible to achieve dynamic optimization of dormitory allocation and lacks early warning of interpersonal conflicts, leading to frequent problems such as tense interpersonal relationships among students and low utilization of dormitory resources.

[0003] Although some universities and related institutions have attempted to introduce information management systems to improve the situation, these systems still have significant shortcomings: First, they lack dynamic adjustment capabilities, making it difficult to optimize allocation plans in a timely manner based on changes in student needs or accommodation feedback; second, their conflict identification and intervention mechanisms are weak, making it impossible to effectively predict and intervene in potential accommodation conflicts in a timely manner; third, their data utilization efficiency is low, failing to fully explore the value of student characteristic data to support decision-making; and fourth, their privacy protection mechanisms are inadequate, posing a risk of leakage of students' personal information.

[0004] In addition, existing information systems still have shortcomings in core aspects such as data processing efficiency, privacy and security protection, and real-time response speed. They also generally lack intelligent analysis capabilities driven by large models and edge computing technology support, making it difficult to meet the personalized and intelligent development needs of modern universities for dormitory management. Summary of the Invention

[0005] This invention addresses the technical problems of traditional dormitory allocation methods, such as low intelligence, lack of dynamic optimization capabilities, and difficulty in predicting interpersonal conflicts, by providing a method and system for dynamic dormitory allocation and management based on a large model.

[0006] Firstly, the present invention provides a method for dynamic allocation and management of dormitories based on a large model, and the technical solution adopted to solve the above-mentioned technical problems is as follows:

[0007] A method for dynamic dormitory allocation and management based on a large model includes the following steps:

[0008] S1. To uniformly register and maintain information on dormitory buildings, rooms and beds, and to complete the basic information management of housing resources;

[0009] S2. Collect students' basic information, daily routines, social preferences, and learning behaviors. Through data preprocessing, data integration, and feature extraction, construct student profiles.

[0010] S3. Based on basic housing information and student profiles, the system comprehensively utilizes a large language model based on the Transformer architecture, multimodal AI analysis technology, and multi-objective optimization algorithms to automatically generate the optimal dormitory allocation plan.

[0011] S4 offers three options: student application for dormitory relocation, batch adjustment by administrators, and automatic dormitory relocation recommendation. It also supports adjustments to the optimal dormitory allocation plan.

[0012] S5. Implement dormitory management by enforcing daily dormitory management, maintenance service management, and management data analysis and decision support.

[0013] Optionally, step S2 specifically includes:

[0014] S2.1 Collect basic student information through the academic affairs system, collect data on student daily routines through the campus card system, collect data on student social preferences through the social interaction platform, and collect data on student learning behavior through the teaching platform;

[0015] S2.2 Preprocess the collected multi-dimensional data, including: removing duplicate data, filling missing values, identifying and handling outliers, and unifying data format and units;

[0016] S2.3. Using the student's ID number or student ID as the unique identifier, the preprocessed multi-dimensional data is linked and integrated to form a unified student data set, ensuring that the basic information, daily routines, social preferences, and learning behavior data of the same student correspond one-to-one.

[0017] S2.4. Extract features from the integrated data, including: extracting static features from basic information, extracting time series features from daily routines, extracting network relationship features from social preferences, and extracting efficiency and effectiveness features from learning behavior. Construct student feature profiles based on the feature extraction results.

[0018] Optionally, step S3 specifically includes:

[0019] S3.1. Perform structured preprocessing on student feature profiles, input the preprocessed student feature profiles into a large language model based on the Transformer architecture, combine multimodal AI analysis technology to model and predict the matching degree of students' living habits, learning synergy and potential interpersonal conflict risk, and output quantitative matching degree, synergy score and conflict risk value.

[0020] S3.2, Set dormitory allocation constraints; at the same time, with the three core optimization objectives of maximizing the weighted score of matching living habits, maximizing the weighted score of learning collaboration, and minimizing the weighted value of conflict risk, the quantitative indicators output in step S3.1 are integrated into a multi-objective optimization function.

[0021] S3.3. Using a multi-objective optimization algorithm, based on the dormitory allocation constraints and multi-objective optimization function in step S3.2, the student group is grouped and iteratively calculated. Multiple candidate grouping schemes are generated through algorithm optimization, and the number of students in each group matches the number of beds in a single dormitory.

[0022] S3.4. Based on the basic information of the housing resources, the student grouping scheme generated in step S3.3 is matched with the dormitory resources, and the beds are allocated within the dormitory according to the students' personal preferences and complementary characteristics, and finally the optimal dormitory allocation scheme is generated.

[0023] Optionally, perform step S5, which includes routine dormitory management such as regular inspections and visitor registration;

[0024] Regularly conduct hygiene and disciplinary inspections of dormitories, record the inspection results, and report them to students and management.

[0025] Implement visitor registration management, receive visitor appointment requests, verify visitor identities, record the time, reason, and corresponding dormitory information of visitors entering and leaving the dormitory, and form a complete visitor record file;

[0026] Dormitory maintenance service management includes: receiving repair requests submitted by students through online channels, with the request content including a description of the repair problem, the repair location and contact information; and tracking the repair progress in real time, synchronizing the status of repair personnel dispatch, repair in progress and repair completion to students.

[0027] Optionally, step S5, dormitory management data analysis and decision support, includes: intelligently retrieving and statistically analyzing data according to four dimensions: occupancy status, disciplinary inspection, hygiene inspection, and leave status, and generating statistical reports and analysis results.

[0028] Secondly, the present invention provides a dormitory dynamic allocation and management system based on a large model, and the technical solution adopted to solve the above-mentioned technical problems is as follows:

[0029] A dormitory dynamic allocation and management system based on a large model, comprising:

[0030] The housing information management module is used to uniformly register and maintain information on dormitory buildings, rooms, and beds, and to complete the basic information management of housing resources;

[0031] The data acquisition and processing module is used to collect students' basic information, daily routines, social preferences, and learning behaviors. Through data preprocessing, data integration, and feature extraction, it constructs student profiles.

[0032] The AI ​​analysis module is used to automatically generate the optimal dormitory allocation plan based on basic housing information and student feature profiles, and by comprehensively using a large language model based on the Transformer architecture, multimodal AI analysis technology and multi-objective optimization algorithms.

[0033] The scheme adjustment module provides three methods for students to apply for dormitory transfer, administrators to make batch adjustments, and automatic dormitory transfer recommendations, and supports adjustments to the optimal dormitory allocation scheme.

[0034] The dormitory management module provides an interactive interface and supports three main functions: daily dormitory management, maintenance service management, and management data analysis and decision support.

[0035] Optionally, the data acquisition and processing modules involved specifically include:

[0036] The data collection unit is used to collect students' basic information through the academic affairs system, data related to students' daily routines through the campus card system, data related to students' social preferences through the social interaction platform, and data related to students' learning behavior through the teaching platform.

[0037] The data preprocessing unit is used to preprocess the collected multi-dimensional data, including: removing duplicate data, filling missing values, identifying and handling outliers, and unifying data format and units;

[0038] The data integration unit is used to link and integrate preprocessed multi-dimensional data using the student's ID number or student ID number as a unique identifier, forming a unified student data set to ensure that the basic information, daily routines, social preferences, and learning behavior data of the same student correspond one-to-one.

[0039] The feature extraction unit is used to extract features from the integrated data, including: extracting static features from basic information, extracting time series features from daily routines, extracting network relationship features from social preferences, and extracting efficiency and effectiveness features from learning behaviors;

[0040] The profile building unit is used to build student feature profiles based on the feature extraction results.

[0041] Optionally, the AI ​​analysis modules involved specifically include:

[0042] The feature processing and prediction unit is used to perform structured preprocessing on student feature profiles. The preprocessed student feature profiles are then input into a large language model based on the Transformer architecture. Combined with multimodal AI analysis technology, the model is used to model and predict the matching degree of students' living habits, learning synergy, and potential interpersonal conflict risks. The output is a quantitative matching degree, synergy score, and conflict risk value.

[0043] The constraint setting unit is used to set dormitory allocation constraints;

[0044] The multi-objective function construction unit is used to integrate the quantitative indicators output by the feature processing and prediction units into a multi-objective optimization function with three core optimization objectives: maximizing the weighted score of life habit matching, maximizing the weighted score of learning collaboration, and minimizing the weighted value of conflict risk.

[0045] The grouping iterative calculation unit is used to perform grouping iterative calculations on the student group based on dormitory allocation constraints and multi-objective optimization functions using a multi-objective optimization algorithm. The algorithm optimizes and generates multiple candidate grouping schemes, and the number of students in each group matches the number of beds in a single dormitory.

[0046] The allocation scheme generation unit combines basic housing information with the student grouping scheme generated by the grouping iteration calculation unit and the dormitory resources. It then completes the bed allocation within the dormitory based on students' personal preferences and complementary characteristics, ultimately generating the optimal dormitory allocation scheme.

[0047] Optionally, the dormitory management module provides an interactive interface to support daily dormitory management functions, including regular inspections and visitor registration. Specifically, it involves regularly conducting hygiene and disciplinary inspections of the dormitory, recording the inspection results and providing feedback to students and management departments; implementing visitor registration management, receiving visitor appointment requests, verifying the identity of visitors, recording the time, reason, and corresponding dormitory information of visitors entering and leaving the dormitory, and forming a complete visitor record file.

[0048] The dormitory management module provides an interactive interface to support dormitory maintenance service management functions, including: receiving repair requests submitted by students through online channels, with the request content including a description of the repair problem, the repair location and contact information; and tracking the repair progress in real time, synchronizing the status of repair personnel dispatch, repair in progress and repair completion to students.

[0049] Optionally, the dormitory management module provides an interactive interface to support dormitory management data analysis and decision support functions, specifically including: intelligently retrieving and statistically analyzing data according to four dimensions: occupancy status, disciplinary inspection, hygiene inspection, and leave status, and generating statistical reports and analysis results.

[0050] The present invention provides a method and system for dynamic dormitory allocation and management based on a large model, which has the following advantages compared with the prior art:

[0051] 1. This invention can achieve in-depth analysis and modeling of multi-dimensional data such as students' daily routines and social preferences, thereby dynamically optimizing dormitory resource allocation, improving students' life satisfaction and dormitory relationship stability, enhancing the scientific and personalized nature of dormitory allocation, and reducing interpersonal conflicts caused by differences in living habits.

[0052] 2. This invention integrates functions such as housing management, dormitory adjustment, dormitory inspection, visitor registration, dormitory repair reporting, and multi-dimensional dormitory query, and is suitable for intelligent dormitory management needs in scenarios such as universities, boarding schools, and employee dormitories of large enterprises. Attached Figure Description

[0053] Appendix Figure 1 This is a flowchart of the method according to Embodiment 1 of the present invention;

[0054] Appendix Figure 2 This is a module connection block diagram of Embodiment 2 of the present invention. Detailed Implementation

[0055] To make the technical solution, the technical problem solved, and the technical effect of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with specific embodiments.

[0056] Example 1:

[0057] Combined with appendix Figure 1 This embodiment proposes a method for dynamic dormitory allocation and management based on a large model, which includes the following steps:

[0058] S1. To uniformly register and maintain information on dormitory buildings, rooms, and beds, and to complete the basic information management of housing resources.

[0059] S2. Collect students' basic information, daily routines, social preferences, and learning behaviors. Through data preprocessing, data integration, and feature extraction, construct student profiles. This process specifically includes:

[0060] S2.1 Collect basic student information through the academic affairs system, collect data on student daily routines through the campus card system, collect data on student social preferences through the social interaction platform, and collect data on student learning behavior through the teaching platform;

[0061] S2.2 Preprocess the collected multi-dimensional data, including: removing duplicate data, filling missing values, identifying and handling outliers, and unifying data format and units;

[0062] S2.3. Using the student's ID number or student ID as the unique identifier, the preprocessed multi-dimensional data is linked and integrated to form a unified student data set, ensuring that the basic information, daily routines, social preferences, and learning behavior data of the same student correspond one-to-one.

[0063] S2.4. Extract features from the integrated data, including: extracting static features from basic information, extracting time series features from daily routines, extracting network relationship features from social preferences, and extracting efficiency and effectiveness features from learning behavior. Construct student feature profiles based on the feature extraction results.

[0064] S3. Based on basic housing information and student profiles, and by comprehensively utilizing a large language model based on the Transformer architecture, multimodal AI analysis technology, and multi-objective optimization algorithms, the optimal dormitory allocation plan is automatically generated. This process specifically includes:

[0065] S3.1. Perform structured preprocessing on student feature profiles, input the preprocessed student feature profiles into a large language model based on the Transformer architecture, combine multimodal AI analysis technology to model and predict the matching degree of students' living habits, learning synergy and potential interpersonal conflict risk, and output quantitative matching degree, synergy score and conflict risk value.

[0066] S3.2, Set dormitory allocation constraints; at the same time, with the three core optimization objectives of maximizing the weighted score of matching living habits, maximizing the weighted score of learning collaboration, and minimizing the weighted value of conflict risk, the quantitative indicators output in step S3.1 are integrated into a multi-objective optimization function.

[0067] S3.3. Using a multi-objective optimization algorithm, based on the dormitory allocation constraints and multi-objective optimization function in step S3.2, the student group is grouped and iteratively calculated. Multiple candidate grouping schemes are generated through algorithm optimization, and the number of students in each group matches the number of beds in a single dormitory.

[0068] S3.4. Based on the basic information of the housing resources, the student grouping scheme generated in step S3.3 is matched with the dormitory resources, and the beds are allocated within the dormitory according to the students' personal preferences and complementary characteristics, and finally the optimal dormitory allocation scheme is generated.

[0069] S4 offers three options: student application for dormitory relocation, batch adjustment by administrators, and automatic dormitory relocation recommendation. It also supports adjustments to the optimal dormitory allocation plan.

[0070] Specifically, students can apply for a dormitory change after moving into their dormitories. They can specify the reasons (such as "conflict with roommates' schedules", "too far from the teaching building", "special health needs") and upload relevant supporting documents (such as hospital diagnosis certificates and screenshots of class schedules). After the administrator approves the application, the system will automatically match available beds that meet the student's needs and provide a list of options. Once the student confirms the change, the dormitory change will be completed, ensuring that individual needs are flexibly met.

[0071] Batch Adjustment by Administrators: In scenarios such as new student enrollment, class adjustments at the beginning of the semester, and dormitory renovations, administrators can adjust dormitories in batches according to their permissions, enabling centralized student management. Administrators can import student lists and filter target dormitory areas through the backend (e.g., "adjust students from 3 classes of a certain major to dormitory building No. 2"). The system automatically checks bed availability, avoids conflicts, and then performs batch dormitory adjustments, simultaneously generating adjustment notices and pushing them to relevant students, thus improving management efficiency.

[0072] Automatic dormitory relocation recommendation: When new students enroll, it provides administrators with an initial allocation reference, reducing manual workload; when there are vacant beds in the dormitory during the semester, it proactively pushes recommended solutions to students with potential dormitory relocation needs (such as when a student is detected to have submitted feedback on "dormitory distance issues" multiple times), guiding them to initiate an application, and achieving a dynamic balance between resource utilization and demand satisfaction.

[0073] S5. Implement dormitory management by enforcing daily dormitory management, maintenance service management, and management data analysis and decision support.

[0074] Dormitory routine management includes regular inspections and visitor registration;

[0075] Regularly conduct hygiene and disciplinary inspections of dormitories, record the inspection results, and report them to students and management.

[0076] Implement visitor registration management, receive visitor appointment requests, verify visitor identities, record the time, reason, and corresponding dormitory information of visitors entering and leaving the dormitory, and form a complete visitor record file.

[0077] Dormitory maintenance service management includes: receiving repair requests submitted by students online, with requests including a description of the repair problem, repair location, and contact information; tracking repair progress in real time, synchronizing the status of repair personnel dispatch, repair in progress, and repair completion to students; and receiving student feedback on the maintenance service after repairs are completed, with evaluations covering both repair quality and efficiency, forming a closed-loop management system to ensure the normal use of dormitory facilities.

[0078] Dormitory management data analysis and decision support includes: intelligently retrieving and statistically analyzing data according to four dimensions: occupancy status (e.g., occupancy rate of each building / room, number of empty beds), disciplinary inspection (e.g., number of disciplinary violations in each dormitory, distribution of violation types), hygiene inspection (e.g., hygiene score of each dormitory, number of times unqualified), and leave status (e.g., student leave duration, leave frequency and related data with the dormitory). This generates statistical reports and analysis results, presenting key dormitory management indicators and providing data support for managers to identify management problems and formulate targeted measures (e.g., strengthening hygiene supervision of a certain building, optimizing the allocation of empty bed resources).

[0079] Example 2:

[0080] Combined with appendix Figure 2 This embodiment proposes a dormitory dynamic allocation and management system based on a large model, which includes:

[0081] The housing information management module is used to uniformly register and maintain information on dormitory buildings, rooms, and beds, and to complete the basic information management of housing resources;

[0082] The data acquisition and processing module is used to collect students' basic information, daily routines, social preferences, and learning behaviors. Through data preprocessing, data integration, and feature extraction, it constructs student profiles.

[0083] The AI ​​analysis module is used to automatically generate the optimal dormitory allocation plan based on basic housing information and student feature profiles, and by comprehensively using a large language model based on the Transformer architecture, multimodal AI analysis technology and multi-objective optimization algorithms.

[0084] The scheme adjustment module provides three methods for students to apply for dormitory transfer, administrators to make batch adjustments, and automatic dormitory transfer recommendations, and supports adjustments to the optimal dormitory allocation scheme.

[0085] The dormitory management module provides an interactive interface and supports three main functions: daily dormitory management, maintenance service management, and management data analysis and decision support.

[0086] In this embodiment, the data acquisition and processing module specifically includes:

[0087] The data collection unit is used to collect students' basic information through the academic affairs system, data related to students' daily routines through the campus card system, data related to students' social preferences through the social interaction platform, and data related to students' learning behavior through the teaching platform.

[0088] The data preprocessing unit is used to preprocess the collected multi-dimensional data, including: removing duplicate data, filling missing values, identifying and handling outliers, and unifying data format and units;

[0089] The data integration unit is used to link and integrate preprocessed multi-dimensional data using the student's ID number or student ID number as a unique identifier, forming a unified student data set to ensure that the basic information, daily routines, social preferences, and learning behavior data of the same student correspond one-to-one.

[0090] The feature extraction unit is used to extract features from the integrated data, including: extracting static features from basic information, extracting time series features from daily routines, extracting network relationship features from social preferences, and extracting efficiency and effectiveness features from learning behaviors;

[0091] The profile building unit is used to build student feature profiles based on the feature extraction results.

[0092] In this embodiment, the AI ​​analysis module specifically includes:

[0093] The feature processing and prediction unit is used to perform structured preprocessing on student feature profiles. The preprocessed student feature profiles are then input into a large language model based on the Transformer architecture. Combined with multimodal AI analysis technology, the model is used to model and predict the matching degree of students' living habits, learning synergy, and potential interpersonal conflict risks. The output is a quantitative matching degree, synergy score, and conflict risk value.

[0094] The constraint setting unit is used to set dormitory allocation constraints;

[0095] The multi-objective function construction unit is used to integrate the quantitative indicators output by the feature processing and prediction units into a multi-objective optimization function with three core optimization objectives: maximizing the weighted score of life habit matching, maximizing the weighted score of learning collaboration, and minimizing the weighted value of conflict risk.

[0096] The grouping iterative calculation unit is used to perform grouping iterative calculations on the student group based on dormitory allocation constraints and multi-objective optimization functions using a multi-objective optimization algorithm. The algorithm optimizes and generates multiple candidate grouping schemes, and the number of students in each group matches the number of beds in a single dormitory.

[0097] The allocation scheme generation unit combines basic housing information with the student grouping scheme generated by the grouping iteration calculation unit and the dormitory resources. It then completes the bed allocation within the dormitory based on students' personal preferences and complementary characteristics, ultimately generating the optimal dormitory allocation scheme.

[0098] In this embodiment, the dormitory management module provides an interactive interface to support the implementation of daily dormitory management functions, including regular inspections and visitor registration. Specifically, the module conducts regular hygiene and disciplinary inspections of the dormitory, records the inspection results and provides feedback to students and management departments. It also implements visitor registration management, receives visitor appointment requests, verifies the identity of visitors, records the time, reason and corresponding dormitory information of visitors entering and leaving the dormitory, and forms a complete visitor record file.

[0099] The dormitory management module provides an interactive interface to support dormitory maintenance service management functions. Specifically, it includes: receiving repair requests submitted by students online, with requests containing a description of the repair problem, repair location, and contact information; tracking repair progress in real time, synchronizing the status of repair personnel dispatch, repair in progress, and repair completion to students. After repairs are completed, students can provide feedback on the maintenance service through the interactive interface, including evaluations of repair quality and efficiency, forming a closed-loop management system for maintenance services to ensure the normal use of dormitory facilities.

[0100] The dormitory management module provides an interactive interface that supports data analysis and decision support functions for dormitory management. Specifically, it includes: intelligently retrieving and statistically analyzing data according to four dimensions: occupancy status (such as occupancy rate of each building / room, number of empty beds), disciplinary inspection (such as the number of disciplinary violations in each dormitory, distribution of violation types), hygiene inspection (such as hygiene score of each dormitory, number of times unqualified), and leave status (such as student leave duration, leave frequency and dormitory-related data). This generates statistical reports and analysis results, presenting key dormitory management indicators and providing data support for managers to identify management problems and formulate targeted measures (such as strengthening hygiene supervision of a certain building and optimizing the allocation of empty bed resources).

[0101] In summary, the dormitory dynamic allocation and management method and system based on a large model of this invention collects multi-dimensional data such as students' daily routines, social preferences, and learning habits. Through data preprocessing, data integration, and feature extraction, student feature profiles are constructed. Based on basic housing information and student feature profiles, the system comprehensively utilizes a large language model based on the Transformer architecture, multimodal AI analysis technology, and multi-objective optimization algorithms to automatically generate optimal dormitory allocation schemes, improving resource allocation efficiency and student satisfaction. The system also integrates core management functions such as housing management, dormitory adjustment, dormitory inspection, visitor registration, dormitory repair reporting, and multi-dimensional dormitory queries, thereby improving logistics management efficiency.

[0102] The above specific examples illustrate the principles and implementation methods of the present invention in detail. These embodiments are merely for the purpose of helping to understand the core technical content of the present invention. Based on the above specific embodiments of the present invention, any improvements and modifications made to the present invention by those skilled in the art without departing from the principles of the present invention should fall within the patent protection scope of the present invention.

Claims

1. A method for dynamic dormitory allocation and management based on a large model, characterized in that, Includes the following steps: S1. To uniformly register and maintain information on dormitory buildings, rooms and beds, and to complete the basic information management of housing resources; S2. Collect students' basic information, daily routines, social preferences, and learning behaviors. Through data preprocessing, data integration, and feature extraction, construct student profiles. S3. Based on basic housing information and student profiles, the system comprehensively utilizes a large language model based on the Transformer architecture, multimodal AI analysis technology, and multi-objective optimization algorithms to automatically generate the optimal dormitory allocation plan. S4 offers three options: student application for dormitory relocation, batch adjustment by administrators, and automatic dormitory relocation recommendation. It also supports adjustments to the optimal dormitory allocation plan. S5. Implement dormitory management by enforcing daily dormitory management, maintenance service management, and management data analysis and decision support.

2. The method for dynamic dormitory allocation and management based on a large model according to claim 1, characterized in that, Step S2 specifically includes: S2.1 Collect basic student information through the academic affairs system, collect data on student daily routines through the campus card system, collect data on student social preferences through the social interaction platform, and collect data on student learning behavior through the teaching platform; S2.2 Preprocess the collected multi-dimensional data, including: removing duplicate data, filling missing values, identifying and handling outliers, and unifying data format and units; S2.

3. Using the student's ID number or student ID as the unique identifier, the preprocessed multi-dimensional data is linked and integrated to form a unified student data set, ensuring that the basic information, daily routines, social preferences, and learning behavior data of the same student correspond one-to-one. S2.

4. Extract features from the integrated data, including: extracting static features from basic information, extracting time series features from daily routines, extracting network relationship features from social preferences, and extracting efficiency and effectiveness features from learning behavior. Construct student feature profiles based on the feature extraction results.

3. The method for dynamic dormitory allocation and management based on a large model according to claim 1, characterized in that, Step S3 specifically includes: S3.

1. Perform structured preprocessing on student feature profiles, input the preprocessed student feature profiles into a large language model based on the Transformer architecture, combine multimodal AI analysis technology to model and predict the matching degree of students' living habits, learning synergy and potential interpersonal conflict risk, and output quantitative matching degree, synergy score and conflict risk value. S3.2, Set dormitory allocation constraints; at the same time, with the three core optimization objectives of maximizing the weighted score of matching living habits, maximizing the weighted score of learning collaboration, and minimizing the weighted value of conflict risk, the quantitative indicators output in step S3.1 are integrated into a multi-objective optimization function. S3.

3. Using a multi-objective optimization algorithm, based on the dormitory allocation constraints and multi-objective optimization function in step S3.2, the student group is grouped and iteratively calculated. Multiple candidate grouping schemes are generated through algorithm optimization, and the number of students in each group matches the number of beds in a single dormitory. S3.

4. Based on the basic information of the housing resources, the student grouping scheme generated in step S3.3 is matched with the dormitory resources, and the beds are allocated within the dormitory according to the students' personal preferences and complementary characteristics, and finally the optimal dormitory allocation scheme is generated.

4. The method for dynamic dormitory allocation and management based on a large model according to claim 1, characterized in that, Step S5 involves routine dormitory management, including regular inspections and visitor registration. Regularly conduct hygiene and disciplinary inspections of dormitories, record the inspection results, and report them to students and management. Implement visitor registration management, receive visitor appointment requests, verify visitor identities, record the time, reason, and corresponding dormitory information of visitors entering and leaving the dormitory, and form a complete visitor record file; Dormitory maintenance service management includes: receiving repair requests submitted by students through online channels, with the request content including a description of the repair problem, the repair location, and contact information; The system tracks the repair progress in real time, synchronizing the status of repair personnel dispatch, repair in progress, and repair completion to students.

5. The method for dynamic dormitory allocation and management based on a large model according to claim 4, characterized in that, Step S5, dormitory management data analysis and decision support, includes: intelligently retrieving and statistically analyzing data according to four dimensions: occupancy status, disciplinary inspection, hygiene inspection, and leave status, and generating statistical reports and analysis results.

6. A dormitory dynamic allocation and management system based on a large model, characterized in that, It includes: The housing information management module is used to uniformly register and maintain information on dormitory buildings, rooms, and beds, and to complete the basic information management of housing resources; The data acquisition and processing module is used to collect students' basic information, daily routines, social preferences, and learning behaviors. Through data preprocessing, data integration, and feature extraction, it constructs student profiles. The AI ​​analysis module is used to automatically generate the optimal dormitory allocation plan based on basic housing information and student feature profiles, and by comprehensively using a large language model based on the Transformer architecture, multimodal AI analysis technology and multi-objective optimization algorithms. The scheme adjustment module provides three methods for students to apply for dormitory transfer, administrators to make batch adjustments, and automatic dormitory transfer recommendations, and supports adjustments to the optimal dormitory allocation scheme. The dormitory management module provides an interactive interface and supports three main functions: daily dormitory management, maintenance service management, and management data analysis and decision support.

7. The dormitory dynamic allocation and management system based on a large model according to claim 6, characterized in that, The data acquisition and processing module specifically includes: The data collection unit is used to collect students' basic information through the academic affairs system, data related to students' daily routines through the campus card system, data related to students' social preferences through the social interaction platform, and data related to students' learning behavior through the teaching platform. The data preprocessing unit is used to preprocess the collected multi-dimensional data, including: removing duplicate data, filling missing values, identifying and handling outliers, and unifying data format and units; The data integration unit is used to link and integrate preprocessed multi-dimensional data using the student's ID number or student ID number as a unique identifier, forming a unified student data set to ensure that the basic information, daily routines, social preferences, and learning behavior data of the same student correspond one-to-one. The feature extraction unit is used to extract features from the integrated data, including: extracting static features from basic information, extracting time series features from daily routines, extracting network relationship features from social preferences, and extracting efficiency and effectiveness features from learning behaviors; The profile building unit is used to build student feature profiles based on the feature extraction results.

8. The dormitory dynamic allocation and management system based on a large model according to claim 6, characterized in that, The AI ​​analysis module specifically includes: The feature processing and prediction unit is used to perform structured preprocessing on student feature profiles. The preprocessed student feature profiles are then input into a large language model based on the Transformer architecture. Combined with multimodal AI analysis technology, the model is used to model and predict the matching degree of students' living habits, learning synergy, and potential interpersonal conflict risks. The output is a quantitative matching degree, synergy score, and conflict risk value. The constraint setting unit is used to set dormitory allocation constraints; The multi-objective function construction unit is used to integrate the quantitative indicators output by the feature processing and prediction units into a multi-objective optimization function with three core optimization objectives: maximizing the weighted score of life habit matching, maximizing the weighted score of learning collaboration, and minimizing the weighted value of conflict risk. The grouping iterative calculation unit is used to perform grouping iterative calculations on the student group based on dormitory allocation constraints and multi-objective optimization functions using a multi-objective optimization algorithm. The algorithm optimizes and generates multiple candidate grouping schemes, and the number of students in each group matches the number of beds in a single dormitory. The allocation scheme generation unit combines basic housing information with the student grouping scheme generated by the grouping iteration calculation unit and the dormitory resources. It then completes the bed allocation within the dormitory based on students' personal preferences and complementary characteristics, ultimately generating the optimal dormitory allocation scheme.

9. A dormitory dynamic allocation and management system based on a large model according to claim 6, characterized in that, The dormitory management module provides an interactive interface to support daily dormitory management functions, including regular inspections and visitor registration. Specifically, it involves regularly conducting hygiene and disciplinary inspections of the dormitory, recording the inspection results and providing feedback to students and management departments; implementing visitor registration management, receiving visitor appointment requests, verifying the identity of visitors, recording the time, reason, and corresponding dormitory information of visitors entering and leaving the dormitory, and forming a complete visitor record file. The dormitory management module provides an interactive interface to support dormitory maintenance service management functions, including: receiving repair requests submitted by students through online channels, with the request content including a description of the repair problem, the repair location and contact information; and tracking the repair progress in real time, synchronizing the status of repair personnel dispatch, repair in progress and repair completion to students.

10. A dormitory dynamic allocation and management system based on a large model according to claim 9, characterized in that, The dormitory management module provides an interactive interface that supports dormitory management data analysis and decision support functions. Specifically, it includes intelligently retrieving and statistically analyzing data according to four dimensions: occupancy status, disciplinary inspection, hygiene inspection, and leave status, and generating statistical reports and analysis results.

Images