Campus Security Risk Cognition Assessment Method, Device, Equipment, Storage Medium and Program Product

Abstract

The present invention discloses a method, device, equipment, storage medium and program product for campus safety risk perception assessment. The method includes: generating a sample data set based on campus image data, training a cognitive reasoning model according to the sample data set to obtain a safety analysis model, generating campus safety problem texts based on the sample data set, inputting the campus safety problem texts into the safety analysis model to obtain model answer information, constructing a cognitive assessment thinking chain model according to the environmental scenario factor information and student factor information in the model answer information, sending student research texts to student users, obtaining questionnaire answer information, and performing safety risk perception assessment on student users according to the cognitive assessment thinking chain model and the questionnaire answer information, so as to accurately assess the risk perception ability of students, provide an effective basis for campus safety behavior education, thereby providing targeted risk perception education for students and greatly reducing the risk of campus safety accidents.

Description

Technical Field

[0001] The present invention relates to the field of campus security information, and in particular to a campus safety risk cognition assessment method, device, equipment, storage medium and program product. Background Art

[0002] With the development of information networks and artificial intelligence, campus security information technology has made great progress. Current monitoring technologies mainly fall into the following three categories: (1) Monitoring of places, such as using the Internet of Things, audio and video, radar waves, millimeter waves, patrol robots and other technologies to monitor and inspect places with security risks; (2) Monitoring of people, such as using various technologies to monitor abnormal behavior and security risk behaviors of people and groups; (3) Monitoring of scenes, such as analyzing and predicting the characteristics of scenes through scene analysis, event evolution knowledge graphs, etc., to discover or predict their security risks.

[0003] These monitoring technologies primarily focus on security monitoring from the perspective of externally acquired observational information. However, they lack a deep understanding and analysis of the cognitive psychology and behavioral motivations behind the safety risk behaviors of students, who are the primary actors in campus security protection. This makes it difficult to effectively prevent various safety incidents from occurring at the source of students' conscious behavior. People's perception of objects and situations is largely influenced by their specific mental models, which are defined as cognitive structures and reasoning mechanisms residing in the brain's memory and possessing typical categorical characteristics within the population. Due to a lack of experience and relevant knowledge, students' mental models significantly inadequately understand safety risks and reason about potential safety hazards, a significant internal factor in safety incidents. Currently, it is impossible to effectively analyze and assess students' risk perceptions, resulting in an inability to provide appropriate safety behavior education and accident prevention education, and therefore an inability to effectively prevent the risk of campus safety incidents. Summary of the Invention

[0004] The main purpose of the present invention is to provide a campus safety risk awareness assessment method, device, equipment, storage medium and program product, aiming to solve the technical problem that the existing technology cannot effectively analyze and evaluate students' risk awareness status, resulting in the inability to provide students with appropriate safety behavior education and accident prevention education, and therefore cannot effectively prevent campus safety accident risks.

[0005] To achieve the above objectives, the present invention provides a campus safety risk awareness assessment method, the method comprising the following steps:

[0006] Generate a sample data set based on the collected campus image data, the sample data set including image data of multiple campus locations and a list of attributes corresponding to the image data;

[0007] Training a pre-built cognitive reasoning model based on the sample data set to obtain a security analysis model;

[0008] generating a campus safety question text based on the sample data set, and inputting the campus safety question text into the safety analysis model to obtain model answer information, wherein the model answer information includes environmental situation factor information and student factor information of the campus safety question in the campus safety question text;

[0009] Constructing a cognitive assessment thinking chain model based on the environmental situation factor information and the student factor information;

[0010] Generate a student survey text, send the student survey text to multiple student users, and obtain questionnaire answer information provided by each student user based on the student survey text;

[0011] A security risk cognition assessment is performed on each student user based on the cognitive assessment thinking chain model and the questionnaire answer information.

[0012] Optionally, the model answer information includes multiple first attribute data items of campus safety issues in the campus safety question text, and the questionnaire answer information includes multiple second attribute data items of campus safety issues in the student survey text; and performing a safety risk cognition assessment on each student user based on the cognitive assessment thinking chain model and the questionnaire answer information includes:

[0013] Performing semantic matching on each first attribute data item in the model answer information and each second attribute data item in the questionnaire answer information to obtain a semantic matching result;

[0014] Determining difference information between each second attribute data item and each first attribute data item according to the semantic matching result;

[0015] Performing a matching score on each second attribute data item according to the difference information to obtain a matching score corresponding to each second attribute data item;

[0016] A security risk cognition assessment is performed on each student user based on the matching score and the cognition assessment thinking chain model.

[0017] Optionally, performing security risk cognition assessment on each student user based on the matching score and the cognition assessment thought chain model includes:

[0018] Normalizing the matching scores corresponding to the second attribute data items to obtain normalized scores;

[0019] determining the number of clusters based on the mental model type, and generating a plurality of clusters based on the number of clusters;

[0020] clustering the normalized scores based on the clustering clusters;

[0021] Determine the proportion parameters of each mental model type of each student user based on the clustering results;

[0022] A security risk awareness assessment is performed on each student user based on the proportion parameter and the cognitive assessment thinking chain model.

[0023] Optionally, generating a sample data set based on the collected campus image data includes:

[0024] Preprocessing the campus image data to obtain a plurality of sample video clips;

[0025] Performing location analysis on each sample video clip to determine the location number and location type corresponding to each sample video clip;

[0026] Conduct safety problem analysis on each sample video clip to determine the type of safety problem, environmental situation factor information, student factor information and formation cause information corresponding to each sample video clip;

[0027] Generate an attribute list for each sample video clip based on the venue number, the venue type, the safety issue type, the environmental situation factor information, the student factor information, and the formation cause information;

[0028] Classifying the sample video clips according to the attribute list to determine abnormal samples and normal samples, and using the normal samples and the attribute lists corresponding to the normal samples to generate student survey texts;

[0029] A sample data set is generated according to the abnormal samples and the attribute list corresponding to each abnormal sample.

[0030] Optionally, generating a student survey text, sending the student survey text to multiple student users, and obtaining questionnaire answer information provided by each student user based on the student survey text may include:

[0031] Obtaining normal samples and attribute lists corresponding to the normal samples based on campus image data;

[0032] Generate a student survey text according to the normal sample and the attribute list corresponding to the normal sample;

[0033] Sending the student survey text to multiple student users, and obtaining questionnaire answer texts provided by each student user based on the student survey text;

[0034] Segmenting the questionnaire answer text to obtain multiple questionnaire answer words;

[0035] Screening the questionnaire answer words according to a preset campus safety domain vocabulary to obtain candidate words;

[0036] Performing feature quantization on the candidate word to obtain a word vector of the candidate word;

[0037] generating a questionnaire answer template according to the model answer information;

[0038] The questionnaire answer information fed back by each student user is obtained based on the questionnaire answer template and the word vector.

[0039] Optionally, after performing the security risk cognition assessment on each student user according to the cognition assessment thinking chain model and the questionnaire answer information, the method further includes:

[0040] Determine the thinking style type, cognitive ability type, and emotional intelligence type of each student user based on the proportion parameters of each mental model type of each student user;

[0041] Extracting features of the thinking style type, the cognitive ability type, and the emotional intelligence type to obtain thinking feature data, cognitive feature data, and emotional intelligence feature data;

[0042] Performing multimodal fusion on the thinking feature data, the cognitive feature data, and the emotional intelligence feature data to obtain comprehensive student status data;

[0043] Inputting the student comprehensive status data into a pre-built development prediction analysis model for prediction analysis to obtain development trajectory information of each student user, wherein the development prediction analysis model is constructed based on development sample data, which includes regression analysis data and causal association data between thinking feature data, cognitive feature data and emotional intelligence feature data;

[0044] Obtain historical behavior information and family background information of each student user;

[0045] Generate an initial profile of each student user based on the historical behavior information and the family background information;

[0046] Correcting the initial profile according to the development trajectory information to obtain a target profile for each student user;

[0047] Based on the target portrait and the security risk awareness assessment result, safety education suggestion information is generated for each student user, and the safety education suggestion information is sent to each student user.

[0048] In addition, to achieve the above-mentioned purpose, the present invention further proposes a campus safety risk cognition assessment device, which includes:

[0049] A data processing module, configured to generate a sample data set based on the collected campus image data, wherein the sample data set includes image data of multiple campus locations and a list of attributes corresponding to the image data;

[0050] A model training module, configured to train a pre-built cognitive reasoning model based on the sample data set to obtain a security analysis model;

[0051] a model analysis module, configured to generate a campus safety question text based on the sample data set, and input the campus safety question text into the safety analysis model to obtain model answer information, wherein the model answer information includes environmental situation factor information and student factor information of the campus safety question in the campus safety question text;

[0052] A thinking model building module, used to build a cognitive assessment thinking chain model based on the environmental situation factor information and the student factor information;

[0053] A student survey module is used to generate a student survey text, send the student survey text to multiple student users, and obtain questionnaire answer information provided by each student user based on the student survey text;

[0054] The risk cognition assessment module is used to perform a security risk cognition assessment on each student user based on the cognition assessment thinking chain model and the questionnaire answer information.

[0055] In addition, to achieve the above-mentioned purpose, the present application also proposes a campus safety risk awareness assessment device, which includes: a memory, a processor, and a computer program stored on the memory and runnable on the processor, and the computer program is configured to implement the steps of the campus safety risk awareness assessment method as described above.

[0056] In addition, to achieve the above-mentioned purpose, the present application also proposes a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the campus safety risk awareness assessment method as described above are implemented.

[0057] In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, it implements the steps of the campus safety risk awareness assessment method as described above.

[0058] The present invention generates a sample data set based on collected campus image data, the sample data set including image data of multiple campus locations and a list of attributes corresponding to the image data. A pre-built cognitive reasoning model is trained based on the sample data set to obtain a safety analysis model. A campus safety question text is generated based on the sample data set, and the campus safety question text is input into the safety analysis model to obtain model answer information. The model answer information includes environmental situational factor information and student factor information of the campus safety question in the campus safety question text. A cognitive assessment thinking chain model is constructed based on the environmental situational factor information and the student factor information, a student survey text is generated, and the student survey text is sent to multiple student users. Questionnaire answer information provided by each student user based on the student survey text is obtained, and safety risk cognition assessment is performed on each student user based on the cognitive assessment thinking chain model and the questionnaire answer information. The present invention constructs a safety analysis model and evaluates the student user's survey results based on the output of the safety analysis model, thereby accurately assessing students' risk cognition ability, providing an effective basis for campus safety behavior education, and providing targeted risk cognition education for students, significantly reducing the risk of campus safety accidents. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] The accompanying drawings are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application, and together with the description, serve to explain the principles of the present application.

[0060] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0061] Figure 1 Schematic diagram of the structure of a campus safety risk awareness assessment device in a hardware operating environment according to an embodiment of the present invention;

[0062] Figure 2 This is a flow chart of the first embodiment of the campus safety risk awareness assessment method of the present invention;

[0063] Figure 3 This is a flow chart of a second embodiment of the campus safety risk awareness assessment method of the present invention;

[0064] Figure 4 This is a structural block diagram of the first embodiment of the campus safety risk awareness assessment device of the present invention.

[0065] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0066] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0067] Reference Figure 1 , Figure 1 This is a schematic diagram of the structure of a campus safety risk awareness assessment device in the hardware operating environment involved in an embodiment of the present invention.

[0068] like Figure 1 As shown, the campus safety risk awareness assessment device may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. The communication bus 1002 is used to enable communication between these components. The user interface 1003 may include a display and an input unit, such as a keyboard. Optionally, the user interface 1003 may also include a standard wired interface or a wireless interface. The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a wireless fidelity (WI-FI) interface). The memory 1005 may be a high-speed random access memory (RAM) or a stable non-volatile memory (NVM), such as a disk drive. The memory 1005 may also be a storage device independent of the processor 1001.

[0069] Those skilled in the art will understand that Figure 1 The structure shown in does not constitute a limitation on the campus safety risk awareness assessment device, and may include more or fewer components than shown in the figure, or a combination of certain components, or a different arrangement of components.

[0070] like Figure 1 As shown, the memory 1005 as a computer-readable storage medium may include an operating system, a network communication module, a user interface module, and a campus safety risk awareness assessment program.

[0071] exist Figure 1In the campus safety risk cognition assessment device shown, the network interface 1004 is mainly used for data communication with the network server; the user interface 1003 is mainly used for data interaction with the user; the processor 1001 and the memory 1005 in the campus safety risk cognition assessment device of the present invention can be set in the campus safety risk cognition assessment device, and the campus safety risk cognition assessment device calls the campus safety risk cognition assessment program stored in the memory 1005 through the processor 1001, and executes the campus safety risk cognition assessment method provided by the embodiment of the present invention.

[0072] The embodiment of the present invention provides a campus safety risk cognition assessment method, referring to Figure 2 , Figure 2 This is a flow chart of the first embodiment of the campus safety risk awareness assessment method of the present invention.

[0073] In this embodiment, the campus safety risk awareness assessment method includes the following steps:

[0074] Step S10: Generate a sample data set based on the collected campus image data.

[0075] It should be noted that the execution entity of this embodiment can be a computing service device with data processing, network communication, and program execution capabilities, such as a tablet computer, personal computer, mobile phone, etc., or a terminal electronic device capable of performing the above functions. The following uses a campus safety risk awareness assessment device (assessment device) as an example to illustrate this embodiment and the following embodiments.

[0076] It should be noted that the sample dataset includes image data of multiple campus locations and a list of attributes corresponding to the image data. The attribute list can be R(s, p, d), where s is the location number, p is the type of security issue, and d is the location description.

[0077] It should be noted that the aforementioned campus image data may be video footage of one or more campus scenes captured by an image sensor. The campus image data may include normal image data and abnormal image data. Normal image data may be video footage that does not contain campus safety issues, while abnormal image data may be video footage that contains campus safety issues.

[0078] In some embodiments, the assessment device can collect 180 seconds of continuous video footage from each location on campus as a sample dataset. For safety issue footage, the footage is captured starting 30 seconds before the issue occurs. The safety issue footage in the sample dataset serves as training data for the multimodal large model, while the normal footage serves as data for assessing students' mental models. The sample dataset should include all safety issue and normal footage from the location.

[0079] Furthermore, in order to improve data quality and thus improve model performance, the above step S10 may include:

[0080] Step S101: pre-processing the campus image data to obtain a plurality of sample video clips;

[0081] Step S102: Performing location analysis on each sample video clip to determine the location number and location type corresponding to each sample video clip;

[0082] Step S103: Analyze the security issues of each sample video clip to determine the security issue type, environmental situation factor information, student factor information, and formation cause information corresponding to each sample video clip;

[0083] Step S104: generating an attribute list of each sample video clip based on the venue number, the venue type, the safety issue type, the environmental situation factor information, the student factor information, and the formation cause information;

[0084] Step S105: classifying the sample video clips according to the attribute list to determine abnormal samples and normal samples, and the normal samples and the attribute lists corresponding to the normal samples are used to generate student survey texts;

[0085] Step S106: Generate a sample data set according to the abnormal samples and the attribute list corresponding to each abnormal sample.

[0086] It should be noted that the venue number can be an identification number corresponding to the campus scene in the sample video clip; the venue type can be a scene type corresponding to the campus scene in the sample video clip (for example, the scene type can be a corridor, stairs, sports field, etc.); the safety problem type can be the problem type of the safety problem occurring in the sample video clip (for example, the safety problem type can be a student falling and getting injured, a fight between students, etc.); the environmental situational factor information can be the environmental factors and scene factors of the above-mentioned safety problem types occurring in the campus scene (for example, the scene environmental factors that cause students to fall can be flooded roads, snowy roads, high crowd density and / or insufficient lighting, etc.); the student factor information can be the factors of the students themselves who cause the above-mentioned safety problem types in the campus scene (for example, walking too fast, crowding, lack of patience, poor eyesight and / or unfamiliarity with the corridor, etc.); the cause information can be the essential cause of the above-mentioned safety problem (for example, lack of relevant safety regulations, weak safety awareness of students, insufficient behavioral norms education, lack of predictive ability of students, etc.).

[0087] For example, the types of safety problems p(k) that may occur in the place numbered 000024 are: "falling and getting injured", "fighting", "broken handrails", "fire danger", etc.; among them, the environmental and situational factors e(l) of "falling and getting injured" are: "stagnant water or snow", "high crowd density", "insufficient lighting", etc.; and under the influence of the environmental and situational factors of "high crowd density", the student factors t that lead to "falling and getting injured" are: "congestion and pushing", "lack of patience", etc.; when "crowd density is high", the reasons r(n) for "falling and getting injured" caused by "congestion and pushing" are: lack of relevant safety regulations, weak safety awareness of students, insufficient education on behavioral norms, and lack of predictive ability of students.

[0088] It should be noted that the above-mentioned attribute list can be a list of the security problem type, environmental situation factor information, student factor information, and formation cause information corresponding to each sample video clip. In some embodiments, the evaluation device can organize the security analysis model and the student user's answer information into a list form, such as a list T[s, d, p(k), e(l), t(m), r(n)]. After training the above video clip samples using the GPT-4 multimodal large model, the list style compiled based on its answers to the above three questions is shown in Table 1 below. Table 1 is a list style table of T[s, d, p(k), e(l), t(m), r(n)]:

[0089] Table 1. Sample table of T[s, d, p(k), e(l), t(m), r(n)]

[0090]

[0091] In some embodiments, preprocessing may include performing image grayscale conversion, image scaling, image denoising, image enhancement, image cropping, edge detection, binarization, image filling, color space conversion, data enhancement, etc. on campus image data.

[0092] Step S20: training the pre-built cognitive reasoning model according to the sample data set to obtain a security analysis model.

[0093] In some embodiments, the evaluation device collects 180 seconds of continuous video footage R (s, p, d) of each location from campus video materials as a sample data set; the above sample data set should include all safety issue video clips and normal video clips that have occurred in the location, and the safety issue video clips are captured starting from 30 seconds before the problem occurs; s is the location number, p is the type of safety issue, d is the location description, and p is 0 in normal video clips; the safety issue video clips in the above sample data set are used as multimodal large model prompt training data, and the normal video clips are used as student mental model evaluation data.

[0094] In some embodiments, the evaluation device uses a multimodal large model to perform prompt training on video clips with safety issues in the R (s, p, d) training data according to the following prompt template, taking the venue as a unit.

[0095] The model's training instructions include: identifying safety issues on campus and analyzing their influencing factors and causes;

[0096] The training prompts of the model include: location number s and description d introduction, type of safety problem p, environmental and situational factors e, student's own factors t, and cause of formation r.

[0097] The input data of the model include: video clips with security issues in R (s, p, d);

[0098] The output format of the model includes: type of safety problem p, environmental and situational factors e, student's own factors t and formation reasons r.

[0099] Step S30: Generate a campus safety question text based on the sample data set, and input the campus safety question text into the security analysis model to obtain model answer information.

[0100] It should be noted that the model answer information includes the environmental situation factor information and student factor information of the campus safety issue in the campus safety issue text.

[0101] In some embodiments, the campus safety question text includes, but is not limited to, the following questions:

[0102] Q1: "What other safety issues may exist in this place? Please classify the safety issues and the causes of each type of problem";

[0103] Q2: "What environmental and situational factors are related to the formation of safety problems in this place? Please list the environmental and situational factors related to each type of problem";

[0104] Q3: "What student factors are related to the formation of safety problems in this place? Please give the student factors related to each type of problem?"

[0105] Step S40: Constructing a cognitive assessment thinking chain model based on the environmental situation factor information and the student factor information.

[0106] In some embodiments, the evaluation device organizes the types of safety issues that may be caused in each place, environmental and situational factors, student factors, and causes of safety issues based on the model answer information output by the safety analysis model, and obtains a list T[s, d, p(k), e(l), t(m), r(n)]; wherein s is the place number, d is the description of the place, p(k), e(l), t(m), r(n) are the types of safety issues that may exist in the place, environmental and situational factors, student factors, and causes of safety issues, respectively, and k, l, m, n are their serial numbers; according to the thinking logic of sdp(k)-e(l)-t(m)-r(n), that is, discovering possible safety issues from campus places and their descriptions, thinking about the environmental and situational factors, student factors, and the reasons for the joint action of the two types of factors that occur in the above safety issues, constructing a cognitive structure and reasoning mechanism thinking chain model, and evaluating students' safety risk cognitive mental model.

[0107] Step S50: Generate a student survey text, send the student survey text to multiple student users, and obtain questionnaire answer information provided by each student user based on the student survey text.

[0108] In some embodiments, the evaluation device may generate a student survey text based on the normal image data in the sample data set, and send the student survey text to multiple student users to obtain questionnaire answer information provided by each student user based on the student survey text.

[0109] In some embodiments, the evaluation device generates a student survey text based on the normal image data in the sample data set, combines the normal image data and the student survey text to perform a visual display of the student users, and simultaneously evaluates more than 30 students; first, the venue number s and venue description d are displayed, and then each student is asked to fill out a questionnaire on the types of safety issues that may exist in the venue p(k), the environmental and situational factors e(l) of each type of safety issue, the student's own factors t(m), and the causes of the safety issues r(n), to obtain questionnaire survey data, process the questionnaire survey data, and obtain questionnaire answer information.

[0110] Furthermore, in order to accurately obtain the questionnaire answer information fed back by the student user, the above step S50 may include:

[0111] Step S501: obtaining normal samples and attribute lists corresponding to the normal samples based on campus image data;

[0112] Step S502: generating a student survey text according to the normal sample and the attribute list corresponding to the normal sample;

[0113] Step S503: sending the student survey text to multiple student users, and obtaining questionnaire answer texts provided by each student user based on the student survey text;

[0114] Step S504: segmenting the questionnaire answer text to obtain multiple questionnaire answer words;

[0115] Step S505: Screening the questionnaire answer words according to a preset campus safety domain vocabulary to obtain candidate words;

[0116] Step S506: quantizing the features of the candidate word to obtain a word vector of the candidate word;

[0117] Step S507: generating a questionnaire answer template based on the model answer information;

[0118] Step S508: Obtaining questionnaire answer information fed back by each student user based on the questionnaire answer template and the word vector.

[0119] It should be noted that the above-mentioned normal samples can be image data in the sample data set that do not have any safety issues.

[0120] In some embodiments, the evaluation device may use methods including but not limited to manual methods, semantic network technology, and Transformer model training technology to perform semantic matching scoring on the questionnaire data filled out by each student and the list T[s, d, p(k), e(l), t(m), r(n)] obtained by sorting out the answers of the large model, including p(k) types of safety issues, e(l) environmental and situational factors, t(m) student's own factors, and r(n) causes of safety issues.

[0121] In some embodiments, the evaluation device may pre-build a vocabulary library in the field of campus safety, which is used to filter out valid words in the questionnaire answers to eliminate invalid words, traverse the word segmentation results, filter out words existing in the vocabulary library as candidate words, and use a pre-trained word vector model (such as Word2Vec, GloVe or BERT) to convert the candidate words into word vectors.

[0122] Step S60: Conducting a security risk cognition assessment on each student user based on the cognition assessment thinking chain model and the questionnaire answer information.

[0123] In a specific implementation, the evaluation device can compare the student user's questionnaire answer information with the model answer information of the security analysis model, and conduct a security risk cognition assessment on each student user based on the difference information of the two answers and combined with the cognitive evaluation thinking chain model.

[0124] This embodiment generates a sample data set based on collected campus image data. The sample data set includes image data of multiple campus locations and a list of attributes corresponding to the image data. A pre-built cognitive reasoning model is trained based on the sample data set to obtain a safety analysis model. A campus safety question text is generated based on the sample data set, and the campus safety question text is input into the safety analysis model to obtain model answer information. The model answer information includes environmental situational factor information and student factor information of the campus safety question in the campus safety question text. A cognitive assessment thinking chain model is constructed based on the environmental situational factor information and the student factor information. A student survey text is generated, and the student survey text is sent to multiple student users. Questionnaire answer information provided by each student user based on the student survey text is obtained. A safety risk cognition assessment is performed on each student user based on the cognitive assessment thinking chain model and the questionnaire answer information. This embodiment constructs a safety analysis model and evaluates the student user's survey results based on the output of the safety analysis model, thereby accurately assessing students' risk cognition ability, providing an effective basis for campus safety behavior education, and providing targeted risk cognition education for students, significantly reducing the risk of campus safety accidents.

[0125] refer to Figure 3 , Figure 3 This is a flow chart of the second embodiment of the campus safety risk awareness assessment method of the present invention.

[0126] Based on the above first embodiment, in this embodiment, step S60 further includes:

[0127] Step S601: semantically match each first attribute data item in the model answer information with each second attribute data item in the questionnaire answer information to obtain a semantic matching result.

[0128] It should be noted that the model answer information includes multiple first attribute data items of campus safety issues in the campus safety issue text, and the questionnaire answer information includes multiple second attribute data items of campus safety issues in the student survey text.

[0129] In some embodiments, the model answer information includes a first answer list, which includes multiple first attribute data items. For example, the first answer list is T1[s, d, p(k), e(l), t(m), r(n)], where s is the venue number, d is the venue description, p(k) is the type of safety issue, e(l) is the environmental and situational factors, t(m) is the student's own factors, and r(n) is the data item of the cause of the safety issue.

[0130] In some embodiments, the questionnaire answer information includes a second answer list, which includes multiple second attribute data items. For example, the second answer list is T2[s, d, p(k), e(l), t(m), r(n)], where s is the venue number, d is the venue description, p(k) is the type of safety issue, e(l) is the environmental and situational factors, t(m) is the student's own factors, and r(n) is the data item of the cause of the safety issue.

[0131] In some embodiments, the evaluation device can construct a student's cognitive structure and reasoning mechanism thinking chain model according to the thinking logic of sdp(k)-e(l)-t(m)-r(n); for example, the types of safety problems p(k) that may occur in the place numbered 000024 are: "falling and getting injured", "fighting", "broken handrails", "fire danger", etc.; among them, the environmental and situational factors e(l) of "falling and getting injured" are: "accumulated water or snow", "high crowd density", "insufficient lighting", etc.; and under the influence of the environmental and situational factors of "high crowd density", the student's own factors t that lead to "falling and getting injured" are: "congestion and pushing", "lack of patience", etc.; when the "crowd density is high", the reasons r(n) for "falling and getting injured" caused by "congestion and pushing" are: lack of relevant safety regulations, weak students' safety awareness, insufficient behavioral norms education, and students' lack of predictive ability.

[0132] Step S602: Determine the difference information between each second attribute data item and each first attribute data item according to the semantic matching result.

[0133] It should be noted that the aforementioned difference information can be obtained by the evaluation device by semantically matching each first attribute data item in the first answer list in the model answer information with each second attribute data item in the second answer list in the questionnaire answer information, and obtaining difference information between each second attribute data item and each first attribute data item based on the semantic differences between the data items. For example, semantic matching can be performed between the types of security questions answered by the model and the types of security questions answered by the student user to obtain difference information between the student user's answers and the model's answers.

[0134] Step S603: performing matching scoring on each second attribute data item according to the difference information to obtain a matching score corresponding to each second attribute data item.

[0135] In some embodiments, the evaluation device may use methods including but not limited to manual methods, semantic network technology, and Transformer model training technology to perform semantic matching scoring on the questionnaire data filled out by each student and the list T[s, d, p(k), e(l), t(m), r(n)] obtained by sorting out the answers of the large model, including p(k) types of safety issues, e(l) environmental and situational factors, t(m) student's own factors, and r(n) causes of safety issues.

[0136] For example, there are K safety issues in p(k). If all of them are mentioned in the student questionnaire data, they will be counted as K points, and 1 point will be deducted for each missing one. If a new issue is found that is mentioned in the questionnaire data but not in p(k), its rationality will be judged manually. If it is reasonable, 1 point will be added and added to p(k). The other e(l), t(m), and r(n) data items are scored in this way, and a score list of four data items P(i), E(i), T(i), and R(i) is obtained, where i is the serial number of the student who filled out the questionnaire. The score ratio of P(i), E(i), T(i), and R(i) data is calculated according to the full score after adding reasonable answers, and normalized to convert it into a numerical value between 0 and 1. Based on the statistical analysis of the above numerical values, including but not limited to mean and variance statistical indicators, an overall evaluation of the students' cognitive mental model of safety issues is conducted.

[0137] For example, there are 12 security issues in p(k) in the above list T. If all of them are mentioned in the student questionnaire data, they will be scored as 12 points, and 1 point will be deducted for each missing one. If a new issue is found in the questionnaire data but not in p(k), its rationality will be judged manually. If it is reasonable, 1 point will be added and added to p(k). According to the same calculation method, the scoring list of the four data items P(i), E(i), T(i), and R(i) is obtained as shown in Table 2 below, where i is the serial number of the student who completed the questionnaire:

[0138] Table 2, Data item scoring list

[0139]

[0140] Step S604: Perform a security risk cognition assessment on each student user based on the matching score and the cognition assessment thinking chain model.

[0141] In some embodiments, the evaluation device may use the K-means method to cluster the normalized P(i), E(i), T(i), and R(i) data, conduct a safety risk awareness assessment on each student user based on the clustering results and the cognitive assessment thinking chain model, determine the student user's mental type, and then generate a reasonable education plan for the student user's corresponding mental type, and provide safety education and training to the student user based on the education plan.

[0142] The score ratio of P(i), E(i), T(i), and R(i) data is calculated based on the full score after adding reasonable answers, and normalized to convert them into values ​​between 0 and 1. For example, after normalization of the data in Table 1 above, the data pattern obtained is shown in Table 3 below. Table 3 is the data table after normalization:

[0143] Table 3, normalized data table

[0144]

[0145] Based on the statistical analysis of the above numerical values, an overall assessment of students' mental models of safety issue cognition was conducted. For example, the normalized data means of P(i), E(i), and T(i) of the five student users in the table above were 0.429, 0.305, 0.431, and 0.327, respectively. This indicates that the overall level of these students' mental models of safety issue cognition is relatively low, and they find it difficult to recognize most safety risks, especially the environmental and situational factors, as well as the possible causes of safety issues caused by the combination of various factors. In addition, the standard deviation of the above data can also be calculated to analyze the degree of dispersion of students' mental models.

[0146] Furthermore, in order to improve the accuracy of cognitive assessment, the above step S604 may include:

[0147] Step S6041: normalizing the matching scores corresponding to the second attribute data items to obtain normalized scores;

[0148] Step S6042: determining the number of clusters based on the mental model type, and generating multiple clusters based on the number of clusters;

[0149] Step S6043: clustering the normalized scores based on the cluster clusters;

[0150] Step S6044: Determine the proportion parameters of each mental model type of each student user based on the clustering results;

[0151] Step S6045: Conduct a security risk awareness assessment on each student user based on the proportion parameter and the cognitive assessment thinking chain model.

[0152] In some embodiments, the evaluation device may use the K-means method to cluster the normalized P(i), E(i), T(i), and R(i) data. Based on experience, the number of clusters is set to 4, and the center values ​​PC, EC, TC, and RC of the four clusters and the number of samples in each cluster are obtained respectively. The number of samples is then converted into a percentage C% of the total number. The four clusters represent four typical mental model types of students. For example, the calculation results for a group of students are shown in Table 4 below, which is an analysis data table of each type of mental model:

[0153] Table 4, analysis data of mental model

[0154]

[0155] The students' safety risk awareness is evaluated according to the cluster center value and sample percentage of each type of mental model. For example, based on the calculation results in Table 1 above, the following evaluation is given:

[0156] Type 1 mental model: accounting for 43.9%, with insufficient knowledge of the types of possible safety issues (PC=0.312), and serious lack of knowledge of the environmental and situational factors that cause safety issues (EC=0.209), but with a certain degree of knowledge of their own factors (TC=0.395), and serious lack of cognitive reasoning about the causes of safety issues (RC=0.281); to a certain extent, this reflects the characteristics of their thinking mode of insufficient divergent thinking and very incomplete reasoning, and may also be due to the fact that they are freshmen who are unfamiliar with the campus environment.

[0157] Type 2 mental model: accounting for 28.1%, with a serious lack of awareness of the types of possible security issues (PC=0.273), a certain level of awareness of the environmental and situational factors that cause security issues (EC=0.354), a relatively high level of awareness of their own factors (TC=0.436), and a certain level of cognitive reasoning ability about the causes of security issues (RC=0.369); to a certain extent, this reflects their lack of divergent thinking and a focus on their own causes.

[0158] Type 3 mental model: accounting for 18.4%, has a certain degree of cognition of the types of possible security issues (PC=0.396), a relatively high degree of cognition of the environmental and situational factors that cause security issues (EC=0.407), a relatively weak cognition of internal factors (TC=0.311), and relatively weak cognitive reasoning ability about the causes of security issues (RC=0.304). To a certain extent, this reflects the characteristics of a thinking model with a certain degree of divergent thinking, focusing on external causes, but with weak reasoning ability.

[0159] Type 4 mental model: accounting for 9.6%, with relatively high awareness of the types of possible security issues, environmental and situational factors that cause security issues, and personal factors (PC=0.427, EC=0.482, TC=0.449), and relatively strong cognitive reasoning ability about the causes of security issues (RC=0.483); to a certain extent, this reflects the characteristics of a thinking model with relatively good divergent thinking, the ability to comprehensively consider various factors, and relatively comprehensive cognitive reasoning.

[0160] Furthermore, in order to effectively avoid campus safety accidents and improve campus safety, after the above step S60, the following steps are further included:

[0161] Step S61: determining the thinking style type, cognitive ability type, and emotional intelligence type of each student user based on the proportion parameters of each mental model type of each student user;

[0162] Step S62: extracting features of the thinking style type, the cognitive ability type, and the emotional intelligence type to obtain thinking feature data, cognitive feature data, and emotional intelligence feature data;

[0163] Step S63: performing multimodal fusion on the thinking feature data, the cognitive feature data, and the emotional intelligence feature data to obtain comprehensive student status data;

[0164] Step S64: inputting the student comprehensive status data into a pre-built development prediction analysis model for prediction analysis to obtain development trajectory information of each student user;

[0165] Step S65: Obtain historical behavior information and family background information of each student user;

[0166] Step S66: generating an initial profile of each student user based on the historical behavior information and the family background information;

[0167] Step S67: Correcting the initial portrait according to the development trajectory information to obtain a target portrait for each student user;

[0168] Step S68: Generate safety education advice information for each student user based on the target portrait and the safety risk awareness assessment result, and send the safety education advice information to each student user.

[0169] It should be noted that the development prediction analysis model is constructed based on development sample data, which includes regression analysis data and causal association data between thinking feature data, cognitive feature data, and emotional intelligence feature data. Multiple linear regression or logistic regression can be used to analyze the impact of different dimensions on learning outcomes, thereby constructing sample data.

[0170] It should be noted that the evaluation device determines the thinking style type, cognitive ability type and emotional intelligence type of each student user based on the proportion parameters of each mental model type of each student user, pre-processes the thinking style types, cognitive ability types and emotional intelligence types of multiple student users (including data cleaning, data standardization and time series alignment), performs feature extraction and multimodal fusion on the pre-processed data, obtains the student's comprehensive status data, generates an initial portrait of each student user based on the historical behavior information and family background information of each student user, constructs a cross-analysis matrix based on the comprehensive status data (the cross-analysis matrix includes interaction information of multiple dimensions (such as thinking style type, cognitive ability type and emotional intelligence type)), corrects the initial portrait based on the cross-analysis matrix, and obtains the target portrait of each student user.

[0171] This embodiment obtains a semantic matching result by semantically matching each first attribute data item in the model answer information with each second attribute data item in the questionnaire answer information, determines the difference information between each second attribute data item and each first attribute data item based on the semantic matching result, performs matching scoring on each second attribute data item based on the difference information, obtains a matching score corresponding to each second attribute data item, and performs a safety risk cognition assessment on each student user based on the matching score and the cognitive assessment thinking chain model; because this embodiment achieves an accurate assessment of the risk cognition ability of student users by semantically matching and scoring the model answer information with the survey answer information of student users, it effectively improves the accuracy of the student risk cognition ability assessment, thereby providing high-quality data for campus safety risk cognition education.

[0172] In addition, an embodiment of the present invention also proposes a computer-readable storage medium, which stores a campus safety risk awareness assessment program. When the campus safety risk awareness assessment program is executed by a processor, it implements the steps of the campus safety risk awareness assessment method described above.

[0173] The computer-readable storage medium provided herein may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems, or devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to, an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including, but not limited to, wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0174] The above-mentioned computer-readable storage medium may be included in the campus safety risk awareness assessment device; or it may exist independently without being assembled into the campus safety risk awareness assessment device.

[0175] In addition, an embodiment of the present invention also proposes a computer program product, including a campus safety risk cognition assessment program, which implements the steps of the campus safety risk cognition assessment method described above when executed by a processor.

[0176] The specific implementation of the computer program product of the present invention is basically the same as the embodiments of the above-mentioned campus safety risk awareness assessment method, and will not be repeated here.

[0177] Reference Figure 4 , Figure 4 This is a structural block diagram of the first embodiment of the campus safety risk awareness assessment device of the present invention.

[0178] like Figure 4 As shown, the campus safety risk awareness assessment device proposed in the embodiment of the present invention includes:

[0179] A data processing module 10 is configured to generate a sample data set based on the collected campus image data, wherein the sample data set includes image data of multiple campus locations and a list of attributes corresponding to the image data;

[0180] A model training module 20 is used to train a pre-built cognitive reasoning model based on the sample data set to obtain a security analysis model;

[0181] A model analysis module 30 is configured to generate a campus safety question text based on the sample data set, and input the campus safety question text into the safety analysis model to obtain model answer information, wherein the model answer information includes environmental situation factor information and student factor information of the campus safety question in the campus safety question text;

[0182] A thinking model building module 40 is used to build a cognitive assessment thinking chain model based on the environmental situation factor information and the student factor information;

[0183] The student survey module 50 is used to generate a student survey text, send the student survey text to multiple student users, and obtain questionnaire answer information provided by each student user based on the student survey text;

[0184] The risk cognition assessment module 60 is used to perform a security risk cognition assessment on each student user based on the cognition assessment thought chain model and the questionnaire answer information.

[0185] Furthermore, the model answer information includes multiple first attribute data items of campus safety issues in the campus safety issue text, and the questionnaire answer information includes multiple second attribute data items of campus safety issues in the student survey text; the risk perception assessment module 60 is also used to semantically match each first attribute data item in the model answer information with each second attribute data item in the questionnaire answer information to obtain a semantic matching result; determine the difference information between each second attribute data item and each first attribute data item based on the semantic matching result; perform a matching score on each second attribute data item based on the difference information to obtain a matching score corresponding to each second attribute data item; and perform a safety risk perception assessment on each student user based on the matching score and the cognitive assessment thinking chain model.

[0186] Furthermore, the risk perception assessment module 60 is also used to normalize the matching scores corresponding to each second attribute data item to obtain a normalized score; determine the number of clusters based on the mental model type, and generate multiple clusters based on the number of clusters; cluster each normalized score based on the cluster; determine the proportion parameters of each mental model type of each student user according to the clustering results; and perform a safety risk perception assessment on each student user based on the proportion parameters and the cognitive assessment thinking chain model.

[0187] Furthermore, the data processing module 10 is also used to pre-process the campus image data to obtain multiple sample video clips; perform a location analysis on each sample video clip to determine the location number and location type corresponding to each sample video clip; perform a safety issue analysis on each sample video clip to determine the safety issue type, environmental scenario factor information, student factor information and formation cause information corresponding to each sample video clip; generate an attribute list for each sample video clip based on the location number, the location type, the safety issue type, the environmental scenario factor information, the student factor information and the formation cause information; classify the sample video clips according to the attribute list to determine abnormal samples and normal samples, and the normal samples and the attribute list corresponding to the normal samples are used to generate student survey texts; and generate a sample data set according to the abnormal samples and the attribute list corresponding to each abnormal sample.

[0188] Furthermore, the student survey module 50 is also used to obtain normal samples and attribute lists corresponding to the normal samples based on campus image data; generate student survey texts based on the normal samples and attribute lists corresponding to the normal samples; send the student survey texts to multiple student users to obtain questionnaire answer texts fed back by each student user based on the student survey texts; perform word segmentation on the questionnaire answer texts to obtain multiple questionnaire answer words; screen the questionnaire answer words according to a preset campus safety field vocabulary to obtain candidate words; perform feature quantization on the candidate words to obtain word vectors of the candidate words; generate a questionnaire answer template based on the model answer information; and obtain questionnaire answer information fed back by each student user based on the questionnaire answer template and the word vector.

[0189] Furthermore, the campus safety risk cognition assessment device further includes:

[0190] The educational suggestion module 70 is used to determine the thinking style type, cognitive ability type and emotional intelligence type of each student user based on the proportion parameters of each mental model type of each student user; perform feature extraction on the thinking style type, cognitive ability type and emotional intelligence type to obtain thinking feature data, cognitive feature data and emotional intelligence feature data; perform multimodal fusion on the thinking feature data, cognitive feature data and emotional intelligence feature data to obtain student comprehensive status data; input the student comprehensive status data into a pre-constructed development prediction analysis model for prediction analysis to obtain development trajectory information of each student user, the development prediction analysis model is constructed based on development sample data, and the development sample data includes regression analysis data and causal association data between thinking feature data, cognitive feature data and emotional intelligence feature data; obtain historical behavior information and family background information of each student user; generate an initial portrait of each student user based on the historical behavior information and the family background information; correct the initial portrait based on the development trajectory information to obtain a target portrait of each student user; generate safety education suggestion information for each student user based on the target portrait and the safety risk cognition assessment result, and send the safety education suggestion information to each student user.

[0191] This embodiment generates a sample data set based on collected campus image data. The sample data set includes image data of multiple campus locations and a list of attributes corresponding to the image data. A pre-built cognitive reasoning model is trained based on the sample data set to obtain a safety analysis model. A campus safety question text is generated based on the sample data set, and the campus safety question text is input into the safety analysis model to obtain model answer information. The model answer information includes environmental situational factor information and student factor information of the campus safety question in the campus safety question text. A cognitive assessment thinking chain model is constructed based on the environmental situational factor information and the student factor information. A student survey text is generated, and the student survey text is sent to multiple student users. Questionnaire answer information provided by each student user based on the student survey text is obtained. A safety risk cognition assessment is performed on each student user based on the cognitive assessment thinking chain model and the questionnaire answer information. This embodiment constructs a safety analysis model and evaluates the student user's survey results based on the output of the safety analysis model, thereby accurately assessing students' risk cognition ability, providing an effective basis for campus safety behavior education, and providing targeted risk cognition education for students, significantly reducing the risk of campus safety accidents.

[0192] The campus safety risk cognition assessment device provided in this application adopts the campus safety risk cognition assessment method of the above-mentioned embodiment, which can solve the technical problems of campus safety risk cognition assessment. Compared with the existing technology, the beneficial effects of the campus safety risk cognition assessment device provided in this application are the same as the beneficial effects of the campus safety risk cognition assessment method provided in the above-mentioned embodiment, and the other technical features of the campus safety risk cognition assessment device are the same as the features disclosed in the above-mentioned embodiment method, which will not be repeated here.

[0193] It should be understood that the above is only an example and does not constitute any limitation to the technical solution of the present invention. In specific applications, those skilled in the art can make settings as needed, and the present invention does not impose any limitation on this.

[0194] It should be noted that the workflow described above is merely illustrative and does not limit the scope of protection of the present invention. In practical applications, technicians in this field can select part or all of it according to actual needs to achieve the purpose of the embodiment scheme, and no limitation is made here.

[0195] In addition, for technical details not fully described in this embodiment, please refer to the campus safety risk awareness assessment method provided in any embodiment of the present invention, and will not be repeated here.

[0196] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or system comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or system. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or system comprising the element.

[0197] The serial numbers of the above embodiments of the present invention are for description only and do not represent the advantages or disadvantages of the embodiments.

[0198] Through the above description of the embodiments, those skilled in the art will clearly understand that the above-mentioned embodiments and methods can be implemented by means of software plus the necessary general-purpose hardware platform. Of course, hardware can also be used, but in many cases the former is a more preferred embodiment. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as read-only memory / random access memory, a magnetic disk, or an optical disk) and includes a number of instructions for enabling a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0199] The above are only preferred embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention description and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A campus safety risk cognition assessment method, characterized in that: The campus safety risk perception assessment method includes: Generate a sample data set based on the collected campus image data, the sample data set including image data of multiple campus locations and a list of attributes corresponding to the image data; Training a pre-built cognitive reasoning model according to the sample data set to obtain a security analysis model; Generate a campus safety problem text based on the sample data set, and input the campus safety problem text into the safety analysis model to obtain model answer information, wherein the model answer information includes environmental situation factor information and student factor information of the campus safety problem in the campus safety problem text; Constructing a cognitive assessment thinking chain model according to the environmental situation factor information and the student factor information; Generate a student survey text, send the student survey text to multiple student users, and obtain questionnaire answer information from each student user based on the student survey text; Conducting a security risk cognition assessment on each student user according to the cognitive assessment thinking chain model and the questionnaire answer information; The generating of a sample data set based on the collected campus image data includes: Preprocessing the campus image data to obtain a plurality of sample video clips; Performing location analysis on each sample video clip to determine the location number and location type corresponding to each sample video clip; Conduct safety problem analysis on each sample video clip to determine the type of safety problem, environmental situation factor information, student factor information and formation cause information corresponding to each sample video clip; Generate an attribute list of each sample video clip based on the venue number, the venue type, the safety problem type, the environmental situation factor information, the student factor information and the formation reason information; Classifying the sample video clips according to the attribute list to determine abnormal samples and normal samples, wherein the normal samples and the attribute lists corresponding to the normal samples are used to generate a student survey text; Generate a sample data set according to the abnormal samples and the attribute list corresponding to each abnormal sample; The step of constructing a cognitive assessment thinking chain model according to the environmental situation factor information and the student factor information includes: Analyze each campus location according to the environmental situation factor information and the student factor information to obtain multiple attributes; Performing logical analysis on the multiple attributes to generate a list; Analyze the environmental and situational factors, student factors, and the reasons for the combined effects of the environmental and situational factors and student factors in the safety issues in the list, and construct a cognitive assessment thinking chain model based on the analysis results; The model answer information includes a plurality of first attribute data items of campus safety issues in the campus safety issue text, and the questionnaire answer information includes a plurality of second attribute data items of campus safety issues in the student survey text; and the safety risk cognitive assessment of each student user based on the cognitive assessment thinking chain model and the questionnaire answer information includes: A security risk cognition assessment is performed on each student user according to the cognition assessment thinking chain model, multiple first attribute data items and multiple second attribute data items.

2. The campus safety risk cognition assessment method according to claim 1, characterized in that: The multiple attributes include: venue number, venue description, type of safety issue, environmental and situational factors, student factors and causes of safety issues.

3. The campus safety risk cognition assessment method according to claim 2, characterized in that: The method of performing a security risk cognitive assessment on each student user according to the cognitive assessment thinking chain model, the plurality of first attribute data items, and the plurality of second attribute data items includes: Performing semantic matching on each first attribute data item in the model answer information and each second attribute data item in the questionnaire answer information to obtain a semantic matching result; Determine difference information between each second attribute data item and each first attribute data item according to the semantic matching result; Perform matching scoring on each second attribute data item according to the difference information to obtain a matching score corresponding to each second attribute data item; A security risk cognition assessment is performed on each student user based on the matching score and the cognitive assessment thinking chain model.

4. The campus safety risk cognition assessment method according to claim 3, characterized in that: The performing of security risk cognitive assessment on each student user based on the matching score and the cognitive assessment thinking chain model includes: Normalizing the matching scores corresponding to each second attribute data item to obtain a normalized score; Determine the number of clusters based on the mental model type, and generate multiple clusters based on the number of clusters, wherein the mental model type includes multiple mental models, each cluster corresponds to a mental model, and the mental model is used to reflect the student user's cognitive ability of the types of existing safety problems, the cognitive ability of the environmental and situational factors that cause safety problems, the cognitive ability of their own factors, and the cognitive reasoning ability of the causes of safety problems; clustering the normalized scores based on the clustering clusters; Determine the proportion parameters of each mental model type of each student user based on the clustering results; A security risk cognition assessment is performed on each student user based on the proportion parameter and the cognitive assessment thinking chain model.

5. The campus safety risk cognition assessment method according to claim 4, characterized in that: The generating of the student survey text, sending the student survey text to multiple student users, and obtaining questionnaire answer information fed back by each student user based on the student survey text includes: Acquire a normal sample and a list of attributes corresponding to the normal sample based on campus image data; Generate a student survey text according to the normal sample and the attribute list corresponding to the normal sample; Sending the student survey text to multiple student users, and obtaining questionnaire answer texts from each student user based on the student survey text; Segmenting the questionnaire answer text to obtain multiple questionnaire answer words; Screening the questionnaire answer words according to a preset campus safety domain word library to obtain candidate words; Performing feature quantization on the candidate word to obtain a word vector of the candidate word; Generate a questionnaire answer template according to the model answer information; The questionnaire answer information fed back by each student user is obtained based on the questionnaire answer template and the word vector.

6. The campus safety risk cognition assessment method according to any one of claims 1 to 5, characterized in that: After the security risk cognition assessment is performed on each student user according to the cognition assessment thinking chain model and the questionnaire answer information, the method further includes: Determine the thinking style type, cognitive ability type and emotional intelligence type of each student user according to the proportion parameters of each mental model type of each student user; Extracting features of the thinking style type, the cognitive ability type, and the emotional intelligence type to obtain thinking feature data, cognitive feature data, and emotional intelligence feature data; Performing multimodal fusion on the thinking feature data, the cognitive feature data and the emotional intelligence feature data to obtain comprehensive student status data; Inputting the student comprehensive status data into a pre-built development prediction analysis model for prediction analysis to obtain development trajectory information of each student user, wherein the development prediction analysis model is built based on development sample data, and the development sample data includes regression analysis data and causal association data between thinking feature data, cognitive feature data and emotional intelligence feature data; Obtain historical behavior information and family background information of each student user; Generate an initial portrait of each student user based on the historical behavior information and the family background information; Correcting the initial portrait according to the development trajectory information to obtain a target portrait of each student user; Based on the target portrait and the safety risk awareness assessment result, safety education suggestion information for each student user is generated, and the safety education suggestion information is sent to each student user.

7. A campus safety risk cognition assessment device, characterized in that: The campus safety risk cognition assessment device comprises: A data processing module, used to generate a sample data set based on the collected campus image data, wherein the sample data set includes image data of multiple campus locations and a property list corresponding to the image data; A model training module, used to train a pre-built cognitive reasoning model according to the sample data set to obtain a security analysis model; A model analysis module, used to generate a campus safety problem text based on the sample data set, and input the campus safety problem text into the safety analysis model to obtain model answer information, wherein the model answer information includes environmental situation factor information and student factor information of the campus safety problem in the campus safety problem text; A thinking model building module, used to build a cognitive evaluation thinking chain model according to the environmental situation factor information and the student factor information; A student survey module is used to generate a student survey text, send the student survey text to multiple student users, and obtain questionnaire answer information from each student user based on the student survey text; A risk cognition assessment module, used to conduct a safety risk cognition assessment on each student user according to the cognition assessment thinking chain model and the questionnaire answer information; The data processing module is further used to pre-process the campus image data to obtain multiple sample video clips; perform a location analysis on each sample video clip to determine the location number and location type corresponding to each sample video clip; perform a safety problem analysis on each sample video clip to determine the safety problem type, environmental scenario factor information, student factor information and formation cause information corresponding to each sample video clip; generate an attribute list for each sample video clip based on the location number, the location type, the safety problem type, the environmental scenario factor information, the student factor information and the formation cause information; classify the sample video clips according to the attribute list to determine abnormal samples and normal samples, and the normal samples and the attribute list corresponding to the normal samples are used to generate a student survey text; generate a sample data set according to the abnormal samples and the attribute list corresponding to each abnormal sample; The thinking model building module is further used to analyze various campus locations according to the environmental situational factor information and the student factor information to obtain multiple attributes; perform thinking logic analysis on the multiple attributes to generate a list; analyze the environmental and situational factors, student factors, and the reasons for the joint action of the environmental and situational factors and student factors in the list, and build a cognitive evaluation thinking chain model based on the analysis results; The model answer information includes multiple first attribute data items of campus safety issues in the campus safety issue text, and the questionnaire answer information includes multiple second attribute data items of campus safety issues in the student survey text; the risk perception assessment module is also used to perform safety risk perception assessment on each student user based on the cognitive assessment thinking chain model, multiple first attribute data items and multiple second attribute data items.

8. A campus safety risk cognition assessment device, characterized in that: The campus safety risk cognition assessment device includes: a memory, a processor, and a campus safety risk cognition assessment program stored in the memory and executable on the processor, wherein the campus safety risk cognition assessment program is configured to implement the campus safety risk cognition assessment method as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a campus safety risk cognition assessment program, and when the campus safety risk cognition assessment program is executed by the processor, the campus safety risk cognition assessment method as described in any one of claims 1 to 6 is implemented.

10. A computer program product, characterized in that The computer program product includes a campus safety risk cognition assessment program, which, when executed by a processor, implements the steps of the campus safety risk cognition assessment method as described in any one of claims 1 to 6.

Images