Campus mental health risk assessment platform and method based on artificial intelligence

Abstract

This invention discloses an artificial intelligence-based campus mental health risk assessment platform and method, belonging to the field of mental health information technology. It includes a mental health data preprocessing module, a mental health data analysis module, a mental health data transmission module, and a risk assessment execution location allocation module. This invention obtains effective sub-mental health data by screening campus mental health data, allocates each effective sub-mental health data to an appropriate processing location, and evaluates the operational status of available cloud processors. It then allocates each effective sub-mental health data in the automated processing location to an appropriate available cloud processor, achieving accurate and effective acquisition and processing of high-quality data, thus solving the problem of inaccurate and ineffective acquisition and processing of mental health data in existing technologies.

Description

Technical Field

[0001] The present invention relates to the field of mental health information technology, and in particular to a campus mental health risk assessment platform and method based on artificial intelligence. Background Art

[0002] With the rapid development of society, mental health issues have gradually become an important factor affecting the student population. Especially on campus, students face challenges such as academic pressure, emotional distress, social problems, etc., all of which may lead to mental health problems. Therefore, accurate and effective mental health risk assessment and early warning have become key needs in school management and student mental health services.

[0003] The existing campus mental health risk assessment obtains the data to be assessed, classifies the data according to preset emotion types, determines the health status probability corresponding to each classified data, and generates the mental health status assessment results of the subject to be assessed based on the health status probability.

[0004] For example, the invention patent with publication number: CN114469092A announces a method, bracelet and system for identifying psychological stress, including: obtaining skin sweat samples through wearable collection equipment; performing cortisol concentration analysis on skin sweat samples to determine sweat cortisol concentration; obtaining user individual baseline values ​​based on user data for multiple time periods throughout the day; performing concentration change calculations based on sweat cortisol concentration and user individual baseline values ​​to obtain concentration change rates; and using the concentration change rates to match in a stress level classification list to obtain user stress identification results.

[0005] For example, the patent application with publication number: CN118609807A discloses a physiological and psychological health screening method and device based on vibration imaging and rPPG, including: acquiring a facial image of the test object for a preset time; capturing and analyzing the facial image based on a preset rPPG algorithm to obtain corresponding physiological health data; capturing and analyzing the facial image based on a preset vibration imaging algorithm to obtain corresponding psychological health data; generating a comprehensive test report based on the physiological health data and the psychological health data.

[0006] However, in the process of implementing the technical solution of the invention in the embodiments of the present application, the present application found that the above technology has at least the following technical problems:

[0007] In the existing technology, when conducting mental health risk assessment, the data is usually simply classified and processed directly. In actual situations, the data obtained on campus has limitations such as large volume and obvious differentiation. Direct classification processing may not be able to match the quality of the psychological data itself with the quality of data processing requirements. This not only causes unreasonable use of data processing resources, but also makes it impossible to process psychological data efficiently and accurately. There is a problem of difficulty in accurately and effectively obtaining and processing psychological data. Summary of the invention

[0008] The embodiments of the present application solve the problem in the prior art that it is difficult to accurately and effectively obtain and process psychological data by providing an artificial intelligence-based campus mental health risk assessment platform and method, thereby achieving accurate and effective acquisition and processing of high-quality data.

[0009] The embodiment of the present application provides a campus mental health risk assessment platform and method based on artificial intelligence, including: a mental health data preprocessing module, which is used to obtain the campus mental health data transmitted by the data source for preprocessing, and screen to obtain each valid sub-mental health data; a mental health data analysis module, which is used to analyze and obtain the risk assessment execution processing method of each valid sub-mental health data based on each valid sub-mental health data; a mental health data transmission module, which is used to execute the risk assessment processing method based on each valid sub-mental health data, and transmit each valid sub-mental health data to the corresponding risk assessment execution location; a risk assessment execution location allocation module, which is used to allocate an automated target processing location for mental health risk assessment processing when the risk assessment execution location of the valid sub-mental health data is an automated processing location.

[0010] Furthermore, the specific process of obtaining the campus mental health data transmitted by the data source for preprocessing is as follows: the campus mental health data transmitted by the data source includes each psychological interaction video transmitted by the data source and the audio corresponding to each psychological interaction video; extracting the basic feature data of each psychological interaction video, the basic feature data of each psychological interaction video includes: video resolution, video frame rate and video compression rate; extracting the audio feature data corresponding to each psychological interaction video, the audio feature data includes: audio sampling rate, audio signal-to-noise ratio and audio bit rate; based on the basic feature data of each psychological interaction video and the audio feature data corresponding to each psychological interaction video, analyzing and obtaining the quality feature index of each psychological interaction video, the quality feature index is used to measure the quality of each psychological interaction video.

[0011] Furthermore, the quality characteristic index of each psychological interaction video is specifically obtained by:

[0012] ;

[0013] In the formula, Expressed as Quality characteristic index of psychological interaction videos, , is the total number of psychological interaction videos, Expressed as video resolution, Expressed as the video resolution threshold, Expressed as video frame rate, Expressed as the video frame rate threshold, Expressed as the video compression rate, Expressed as the video compression rate threshold, Expressed as the audio sampling rate, Expressed as the audio sampling rate threshold, Expressed as the audio signal-to-noise ratio, Expressed as the audio signal-to-noise ratio threshold, Expressed as audio bit rate, Expressed as the audio bitrate threshold, Expressed as the video resolution weight ratio, Expressed as the video frame rate weight ratio, Expressed as the weight ratio of video compression rate, Expressed as the audio sampling rate weight ratio, Expressed as the audio signal-to-noise ratio weight ratio, Expressed as the audio bit rate weight ratio, Expressed as The weight ratio of the basic feature data of the psychological interaction video, Expressed as The weight ratio of audio feature data corresponding to the psychological interaction video, is a natural constant.

[0014] Furthermore, the specific process of screening to obtain each valid sub-psychological health data is called: obtaining a preset quality characteristic index threshold; if the quality characteristic index of a psychological interaction video is less than or equal to the quality characteristic index threshold, the psychological interaction video is defined as unqualified; if the quality characteristic index of a psychological interaction video is greater than the quality characteristic index threshold, the psychological interaction video is defined as qualified, the psychological interaction video is recorded as a valid psychological interaction video, and the valid psychological interaction video and the audio corresponding to the valid psychological interaction video are jointly defined as valid sub-psychological health data, thereby obtaining each valid sub-psychological health data statistically.

[0015] Furthermore, the specific process of the risk assessment execution processing method based on each valid sub-psychological health data and analyzing each valid sub-psychological health data is as follows: each valid sub-psychological health data includes each valid psychological interaction video and the audio corresponding to each valid psychological interaction video; based on each valid psychological interaction video, the video duration, video byte number, video frame rate and video bit rate are extracted and processed to obtain the video processing complexity index of each valid psychological interaction video; based on the audio corresponding to each valid psychological interaction video, the number of channels, audio byte number, audio sampling rate and audio dynamic range are extracted and processed to obtain the audio processing complexity index corresponding to each valid psychological interaction video; based on the quality characteristic index of each psychological interaction video, the quality characteristic index of each valid psychological interaction video is extracted, and the video processing complexity index of each valid psychological interaction video, the audio processing complexity index corresponding to each valid psychological interaction video and the quality characteristic index of each valid psychological interaction video are comprehensively processed to obtain the complexity evaluation index of each valid sub-psychological health data, and the complexity evaluation index of each valid sub-psychological health data is used to measure the processing difficulty of each valid sub-psychological interaction video; based on the complexity evaluation index of each valid sub-psychological health data, the risk assessment execution processing method of each valid sub-psychological health data is processed.

[0016] Furthermore, the specific process of obtaining the risk assessment execution processing method of each valid sub-mental health data is: based on the complexity assessment index of each valid sub-mental health data, and obtaining a preset complexity assessment index threshold, the complexity assessment index of each valid sub-mental health data is compared with the complexity assessment index threshold. After comparison, if the complexity assessment index of a certain valid sub-mental health data is less than or equal to the complexity assessment index threshold, then the risk assessment execution method of the valid sub-mental health data is defined as an automatic processing method; if the complexity assessment index of a certain valid sub-mental health data is greater than the complexity assessment index threshold, then the risk assessment execution method of the valid sub-mental health data is defined as a manual processing method.

[0017] Furthermore, the specific process of transmitting each valid sub-mental health data to the corresponding risk assessment execution location is: according to the risk assessment execution processing method of each valid sub-mental health data, if the risk assessment execution method of the valid sub-mental health data is an automatic processing method, then the valid sub-mental health data is transmitted to the automatic processing location; if the risk assessment execution method of the valid sub-mental health data is a manual processing method, then the valid sub-mental health data is transmitted to the manual processing location.

[0018] Furthermore, the specific process of assigning the automated target processing location for mental health risk assessment is as follows: statistically analyze all valid sub-mental health data whose risk assessment execution location is the automated processing location, mark them as each target automated processing psychological data, and statistically analyze the complexity assessment indicators of each target automated processing psychological data for mean processing to obtain the average data complexity of the target automated processing psychological data, and match them with the automated processing control sets corresponding to the average data complexity intervals preset in the database to obtain the automated processing control set corresponding to the interval where the average data complexity of the target automated processing psychological data is located, recorded as the designated automated processing control set; based on the available cloud processors in the automated processing location, extract the operating data of each available cloud processor, and perform comprehensive processing with the designated automated processing control set to obtain the allocation priority index of each available cloud processor, select the available cloud processor corresponding to the maximum allocation priority index, and use it as the automated target processing location for mental health risk assessment.

[0019] Furthermore, the automatic processing control set includes: control processing rate, control processing bandwidth and control processing throughput; the operating data of each available cloud processor includes: real-time processing rate, real-time processing bandwidth, real-time processing throughput, length of the task queue to be processed, real-time CPU usage and average processing response time.

[0020] An embodiment of the present application provides a campus mental health risk assessment method based on artificial intelligence, comprising the following steps: obtaining campus mental health data transmitted by a data source for preprocessing, and screening to obtain each valid sub-mental health data; based on each valid sub-mental health data, analyzing to obtain a risk assessment execution processing method for each valid sub-mental health data; based on the risk assessment execution processing method for each valid sub-mental health data, transmitting each valid sub-mental health data to a corresponding risk assessment execution location; when the risk assessment execution location of the valid sub-mental health data is an automated processing location, allocating an automated target processing location for mental health risk assessment processing.

[0021] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0022] 1. By acquiring the campus mental health data transmitted by the data source, each valid sub-mental health data is screened and different risk assessment processing methods are determined according to the complexity of each valid sub-mental health data. The effective sub-mental health data with automatic risk assessment processing method is transmitted to the appropriate place of origin for processing, reducing the limitations of large data volume and obvious differentiation, better matching the quality of the data itself with the data processing requirements, realizing efficient and accurate processing of psychological data, and effectively solving the problem of accurate and effective acquisition and processing of psychological data in the existing technology.

[0023] 2. By acquiring the campus mental health data transmitted by the data source, effective sub-mental health data are screened according to the basic feature data of each psychological interaction video and the audio feature data of the audio corresponding to each psychological interaction video, thereby obtaining various effective mental health data, and then realizing the effective screening of campus mental health data.

[0024] 3. By conducting a complexity assessment on each valid sub-mental health data, the processing method of each valid sub-mental health data is determined, and each valid sub-mental health data is transmitted to the corresponding place for risk assessment and execution, and an assessment is conducted on each available cloud processor at the location of automated processing, and an available cloud processor is allocated to each valid sub-mental health data. This not only promotes the rational use of data processing resources, but also ensures that each valid sub-mental health data can be responded to and processed quickly and effectively. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 A schematic diagram of an artificial intelligence-based campus mental health risk assessment platform provided in an embodiment of the present application;

[0026] Figure 2 A complexity evaluation index curve diagram provided for an embodiment of the present application;

[0027] Figure 3 A flow chart of the campus mental health risk assessment method based on artificial intelligence provided in an embodiment of the present application. DETAILED DESCRIPTION

[0028] The embodiments of the present application solve the problem that it is difficult to accurately and effectively obtain and process psychological data in the prior art by providing an artificial intelligence-based campus mental health risk assessment platform and method. The campus mental health data transmitted by the data source is obtained, and each valid sub-mental health data is screened and obtained. Different risk assessment processing methods are determined according to the complexity of each valid sub-mental health data. The valid sub-mental health data with an automatic risk assessment processing method is transmitted to a suitable place of origin for processing, thereby achieving accurate and effective acquisition and processing of high-quality data, and effectively solving the problem that it is difficult to accurately and effectively obtain and process psychological data in the prior art.

[0029] The technical solution in the embodiment of the present application is to solve the above-mentioned problem of difficulty in accurately and effectively obtaining and processing psychological data. The overall idea is as follows:

[0030] By acquiring campus mental health data transmitted by the data source, each valid sub-mental health data is screened and obtained, and different risk assessment and processing methods are determined according to the complexity of each valid sub-mental health data, the valid sub-mental health data with automatic risk assessment processing method is transmitted to the appropriate destination, and the appropriate available cloud processors are allocated to the valid sub-mental health data according to the situation of each available cloud processor, thereby achieving accurate and effective acquisition and processing of high-quality data, and effectively solving the problem of accurate and effective acquisition and processing of psychological data in the existing technology.

[0031] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0032] like Figure 1 As shown, it is a system diagram of an artificial intelligence-based campus mental health risk assessment platform provided in an embodiment of the present application, including: a mental health data preprocessing module, which is used to obtain the campus mental health data transmitted by the data source for preprocessing, and screen to obtain each valid sub-mental health data; a mental health data analysis module, which is used to analyze and obtain the risk assessment execution processing method of each valid sub-mental health data based on each valid sub-mental health data; a mental health data transmission module, which is used to execute the processing method based on the risk assessment of each valid sub-mental health data, and transmit each valid sub-mental health data to the corresponding risk assessment execution location; a risk assessment execution location allocation module, which is used to allocate an automated target processing location for mental health risk assessment processing when the risk assessment execution location of the valid sub-mental health data is an automated processing location.

[0033] Furthermore, the specific process of obtaining the campus mental health data transmitted by the data source for preprocessing is as follows: the campus mental health data transmitted by the data source includes each psychological interaction video transmitted by the data source and the audio corresponding to each psychological interaction video; extracting the basic feature data of each psychological interaction video, the basic feature data of each psychological interaction video includes: video resolution, video frame rate and video compression rate; extracting the audio feature data corresponding to each psychological interaction video, the audio feature data includes: audio sampling rate, audio signal-to-noise ratio and audio bit rate; based on the basic feature data of each psychological interaction video and the audio feature data corresponding to each psychological interaction video, analyzing and obtaining the quality feature index of each psychological interaction video, the quality feature index is used to measure the quality of each psychological interaction video.

[0034] In this embodiment, the video resolution is the pixel size of the video image, representing the clarity of the video image, the video frame rate is the number of frames played per second, and the video compression rate is the ratio of the size of the original video data to the size of the compressed video data. The video resolution and the video frame rate are obtained through the media information tool, and the video compression rate is obtained by checking the size of the original video file and the compressed file through the media information tool and performing a ratio process on the size of the original video data to the size of the compressed video data.

[0035] The audio sampling rate is the number of times the audio is sampled per second, the audio signal-to-noise ratio is the ratio between the useful signal and the noise, and the audio bit rate is the amount of data transmitted or stored per unit time. The audio sampling rate, audio signal-to-noise ratio and audio bit rate can be obtained through audio analysis tools.

[0036] Furthermore, the quality characteristic index of each psychological interaction video is obtained by:

[0037] ;

[0038] In the formula, Expressed as Quality characteristic index of psychological interaction videos, , is the total number of psychological interaction videos, Expressed as video resolution, Expressed as the video resolution threshold, Expressed as video frame rate, Expressed as the video frame rate threshold, Expressed as the video compression rate, Expressed as the video compression rate threshold, Expressed as the audio sampling rate, Expressed as the audio sampling rate threshold, Expressed as the audio signal-to-noise ratio, Expressed as the audio signal-to-noise ratio threshold, Expressed as audio bit rate, Expressed as the audio bitrate threshold, Expressed as the video resolution weight ratio, Expressed as the video frame rate weight ratio, Expressed as the weight ratio of video compression rate, Expressed as the audio sampling rate weight ratio, Expressed as the audio signal-to-noise ratio weight ratio, Expressed as the audio bit rate weight ratio, Expressed as The weight ratio of the basic feature data of the psychological interaction video, Expressed as The weight ratio of audio feature data corresponding to the psychological interaction video, is a natural constant.

[0039] In this embodiment, the video resolution threshold, the video frame rate threshold, the video compression rate threshold, the audio sampling rate threshold, the audio signal-to-noise ratio threshold, and the audio bit rate threshold are obtained from a database.

[0040] In this embodiment, , and are the influence weight ratios of the preset video resolution, video frame rate and video compression rate obtained from the preset database, respectively, which are used to represent the numerical values ​​of the influence degree of the video resolution, video frame rate and video compression rate on the basic feature data of each psychological interaction video. When used, the influence weight ratios corresponding to the video resolution, video frame rate and video compression rate can be directly obtained from the preset database, and the corresponding relationship can be a preset mapping relationship. For example, the video resolution, video frame rate and video compression rate form a mapping set with the weights corresponding to the video resolution, video frame rate and video compression rate in the preset database. After the real-time video resolution, video frame rate and video compression rate are input into the mapping set, the corresponding weight ratios of the video resolution, video frame rate and video compression rate can be obtained. In this example , and The value range is , , and The sum of is 1. , and The influence weight ratios of the preset audio sampling rate, audio signal-to-noise ratio and audio bit rate are respectively obtained from the preset database, and are used to represent the numerical values ​​of the influence degree of the audio sampling rate, audio signal-to-noise ratio and audio bit rate on the audio feature data. When used, the influence weight ratios corresponding to the audio sampling rate, audio signal-to-noise ratio and audio bit rate can be directly obtained from the preset database. The corresponding relationship can be a preset mapping relationship. For example, the audio sampling rate, audio signal-to-noise ratio and audio bit rate form a mapping set with the weights corresponding to the audio sampling rate, audio signal-to-noise ratio and audio bit rate in the preset database. After the real-time audio sampling rate, audio signal-to-noise ratio and audio bit rate are input into the mapping set, the corresponding weight ratios of the audio sampling rate, audio signal-to-noise ratio and audio bit rate can be obtained. In this example , and The value range is , , and The sum of is 1. and The influence weight ratios of the preset basic feature data and audio feature data of each psychological interaction video obtained from the preset database are respectively used to represent the numerical value of the influence degree of the basic feature data and audio feature data of each psychological interaction video on the quality feature index of each psychological interaction video. When used, the influence weight ratios corresponding to the basic feature data and audio feature data of each psychological interaction video can be directly obtained from the preset database. The corresponding relationship can be a preset mapping relationship. For example, the basic feature data and audio feature data of each psychological interaction video and the weights corresponding to the basic feature data and audio feature data of each psychological interaction video in the preset database form a mapping set. After the real-time basic feature data and audio feature data of each psychological interaction video are input into the mapping set, the corresponding weight ratios of the basic feature data and audio feature data of each psychological interaction video can be obtained. In this example and The value range is , and The sum of is 1.

[0041] In this embodiment, the higher the video resolution, the higher the video frame rate must be to ensure the smoothness of video playback, and the compression rate must be lowered to ensure video quality. The higher the audio sampling rate, the more data sampled per second, the higher the audio bit rate, the higher the audio sampling accuracy, the more audio details can be captured, and the higher the audio signal-to-noise ratio. The above formula uses exponential operation for nonlinear transformation, smoothes the influence of outliers, and makes the obtained quality feature index more reliable and effective.

[0042] Furthermore, the specific process of screening out each valid sub-psychological health data is called: obtaining a preset quality characteristic index threshold; if the quality characteristic index of a psychological interaction video is less than or equal to the quality characteristic index threshold, the psychological interaction video is defined as unqualified; if the quality characteristic index of a psychological interaction video is greater than the quality characteristic index threshold, the psychological interaction video is defined as qualified, the psychological interaction video is recorded as a valid psychological interaction video, and the valid psychological interaction video and the audio corresponding to the valid psychological interaction video are jointly defined as valid sub-psychological health data, thereby obtaining each valid sub-psychological health data statistically.

[0043] In this embodiment, the preset quality characteristic index threshold is obtained. , if the quality feature index of a psychological interaction video is less than or equal to the quality feature index threshold, that is When the quality characteristic index of a psychological interaction video is greater than the quality characteristic index threshold, the quality of the psychological interaction video is unqualified, that is, the campus psychological health data is unqualified, and the campus psychological health data is screened out. When the quality of the psychological interaction video is qualified, the campus mental health data is qualified, and the campus mental health data is used as valid sub-mental health data.

[0044] Furthermore, based on each valid sub-mental health data, the specific process of analyzing and obtaining the risk assessment execution processing method of each valid sub-mental health data is as follows: each valid sub-mental health data includes each valid psychological interaction video and the audio corresponding to each valid psychological interaction video; based on each valid psychological interaction video, the video duration, video byte number, video frame rate and video bit rate are extracted therefrom and processed to obtain the video processing complexity index of each valid psychological interaction video; based on the audio corresponding to each valid psychological interaction video, the number of channels, audio byte number, audio sampling rate and audio dynamic range are extracted therefrom and processed to obtain the audio processing complexity index corresponding to each valid psychological interaction video; based on the quality characteristic index of each psychological interaction video, the quality characteristic index of each valid psychological interaction video is extracted therefrom, and the video processing complexity index of each valid psychological interaction video, the audio processing complexity index corresponding to each valid psychological interaction video and the quality characteristic index of each valid psychological interaction video are comprehensively processed to obtain the complexity evaluation index of each valid sub-mental health data, and the complexity evaluation index of each valid sub-mental health data is used to measure the processing difficulty of each valid sub-psychological interaction video; based on the complexity evaluation index of each valid sub-mental health data, the risk assessment execution processing method of each valid sub-mental health data is processed.

[0045] In this embodiment, the specific process of obtaining the video processing complexity index of each effective psychological interaction video is as follows:

[0046] ;

[0047] In the formula, Expressed as The video processing complexity index of effective psychological interaction videos, , is the total number of effective psychological interaction videos, Represents the video duration. Represented as the video length threshold, Expressed as the number of video bytes, Expressed as the video byte count threshold, Expressed as video frame rate, Expressed as the video frame rate threshold, Expressed as the video bit rate, Expressed as the video bit rate threshold, Expressed as the weight ratio of video duration, Expressed as the weight ratio of video bytes, Expressed as the video frame rate weight ratio, It is expressed as a video bit rate weight ratio. The video duration threshold, video byte number threshold, video frame rate threshold and video bit rate threshold are obtained from the database.

[0048] Video duration refers to the length of time the video is played, video bytes refer to the size of the video file, video frame rate refers to the number of frames played per second, and video bit rate refers to the amount of data transmitted per second. Video duration, video bytes, video frame rate, and video bit rate can all be obtained through video processing software.

[0049] In this embodiment, , , and are respectively the influence weight ratios of preset video duration, video byte count, video frame rate and video bit rate obtained from the preset database, which are used to represent the numerical values ​​of the influence degree of video duration, video byte count, video frame rate and video bit rate on the video processing complexity index. When used, the influence weight ratios corresponding to the video duration, video byte count, video frame rate and video bit rate can be directly obtained from the preset database. The corresponding relationship can be a preset mapping relationship. For example, the video duration, video byte count, video frame rate and video bit rate form a mapping set with the weights corresponding to the video duration, video byte count, video frame rate and video bit rate in the preset database. After the real-time video duration, video byte count, video frame rate and video bit rate are input into the mapping set, the corresponding weight ratios of video duration, video byte count, video frame rate and video bit rate can be obtained. In this example , , and The value range is , , , and The sum of is 1.

[0050] In this embodiment, there is a certain correlation between the video duration, the number of video bytes, the video frame rate and the video bit rate. The longer the video duration, the larger the video file, the larger the number of video bytes, the higher the video bit rate, and the amount of data transmitted or stored per second increases. The larger the number of video bytes and the higher the video bit rate, the higher the required video frame rate. The larger the video frame rate and the video bit rate, the more data each frame contains and the more frames need to be processed, which increases the complexity of the video processing. The above formula uses logarithmic operations for nonlinear transformation processing, which makes the video processing complexity index more explanatory.

[0051] The specific process of obtaining the audio processing complexity index of each effective psychological interaction video is as follows:

[0052] ;

[0053] In the formula, Expressed as The audio processing complexity index of the audio corresponding to the effective psychological interaction video, Expressed as the number of channels, Expressed as the channel number threshold, Expressed as the number of audio bytes, Expressed as audio byte count threshold, Expressed as the audio sampling rate, Expressed as the audio sampling rate threshold, Expressed as audio dynamic range, Expressed as the audio dynamic range threshold, Expressed as the weight ratio of the number of channels, Expressed as the weight ratio of the number of audio bytes, Expressed as the audio sampling rate weight ratio, Expressed as the audio dynamic range weight ratio, are natural constants, and the channel number threshold, audio byte number threshold, audio sampling rate threshold, and audio dynamic range threshold are obtained from the database.

[0054] The number of channels refers to the number of independent sound channels in the audio signal, the number of audio bytes refers to the size of the audio file, the audio sampling rate refers to the number of times the audio is sampled per second, and the audio dynamic range refers to the level difference between the maximum undistorted sound and the minimum audible sound in the audio signal. The number of channels is obtained through audio editing software, the number of audio bytes can be obtained by viewing the audio file properties, the audio sampling rate is obtained through audio analysis tools, and the audio dynamic range is obtained using audio editing software.

[0055] In this embodiment, there is a certain correlation between the number of channels, the number of audio bytes, the audio sampling rate and the audio dynamic range. As the number of channels increases, the number of audio bytes will also increase accordingly. The higher the audio sampling rate, the more bytes generated by each channel will also increase, that is, the number of audio bytes will increase. The higher the number of audio bytes, the larger the corresponding audio dynamic range. The larger the number of channels, the number of audio bytes, the audio sampling rate and the audio dynamic range, the greater the difficulty of processing audio data, that is, the higher the audio processing complexity. The above formula uses square root operation for nonlinear transformation to make the audio processing complexity index more reliable.

[0056] In this embodiment, , , and They are respectively the influence weight ratios of the preset number of channels, number of audio bytes, audio sampling rate and audio dynamic range obtained from the preset database, and are used to represent the numerical values ​​of the influence degree of the number of channels, number of audio bytes, audio sampling rate and audio dynamic range on the audio processing complexity index. When used, the influence weight ratios corresponding to the number of channels, number of audio bytes, audio sampling rate and audio dynamic range can be directly obtained from the preset database, and the corresponding relationship can be a preset mapping relationship. For example, the number of channels, number of audio bytes, audio sampling rate and audio dynamic range form a mapping set with the weights corresponding to the number of channels, number of audio bytes, audio sampling rate and audio dynamic range in the preset database. After the real-time number of channels, number of audio bytes, audio sampling rate and audio dynamic range are input into the mapping set, the corresponding weight ratios of the number of channels, number of audio bytes, audio sampling rate and audio dynamic range can be obtained. In this example , , and The value range is , , , and The sum of is 1.

[0057] The specific method for obtaining the complexity evaluation index of each effective psychological interaction video is as follows:

[0058] ;

[0059] In the formula, Expressed as Complexity evaluation index for effective psychological interaction videos, Expressed as The video processing complexity index of effective psychological interaction videos, Expressed as The audio processing complexity index of the corresponding audio of the effective psychological interaction video, Expressed as The quality characteristic index of effective psychological interaction videos, It is expressed as the weight ratio of the video processing complexity index in the complexity evaluation index. It is expressed as the weight ratio of the audio processing complexity index in the complexity evaluation index. No. The weight ratio of the quality feature index of an effective psychological interaction video.

[0060] In this embodiment, the video complexity index and the audio complexity index are combined to obtain the complexity evaluation index of each effective sub-psychological health data, and the complexity evaluation index measures the processing difficulty of each effective sub-psychological interaction video.

[0061] In this embodiment, , and are the influence weight ratios of the preset video resolution, video frame rate and video compression rate obtained from the preset database, respectively, which are used to represent the numerical values ​​of the influence degree of the video processing complexity index, audio processing complexity index and quality characteristic index on the complexity evaluation index of each effective psychological interaction video. When used, the influence weight ratios corresponding to the video processing complexity index, audio processing complexity index and quality characteristic index can be directly obtained from the preset database. The corresponding relationship can be a preset mapping relationship. For example, the video processing complexity index, audio processing complexity index and quality characteristic index form a mapping set with the weights corresponding to the video processing complexity index, audio processing complexity index and quality characteristic index in the preset database. After the real-time video processing complexity index, audio processing complexity index and quality characteristic index are input into the mapping set, the corresponding weight ratios of the video processing complexity index, audio processing complexity index and quality characteristic index can be obtained. In this example , and The value range is , , and The sum of is 1.

[0062] like Figure 2 As shown, it is a complexity evaluation index curve diagram provided by the embodiment of the present application, with the video complexity index as the independent variable and the complexity evaluation index as the dependent variable. Through the increase and decrease analysis of curve 1, curve 2 and curve 3, it can be known that when the video complexity index is larger, the complexity evaluation index is larger; when the video complexity index is smaller, the complexity evaluation index is smaller. The relevant data statistics of the complexity evaluation index are shown in Table 1:

[0063] Table 1 Statistics of relevant data of complexity evaluation indicators

[0064]

[0065] It can be seen from Table 1 that for curve 1, curve 2 and curve 3, when the number of user feedback errors and the video complexity index of curve 1, curve 2 and curve 3 are 1.00, when the audio complexity index of curve 1, curve 2 and curve 3 are 0.50, 1.00 and 1.50 respectively, and other parameters are consistent, the complexity evaluation indexes of curve 1, curve 2 and curve 3 are 1.43, 1.54 and 1.66 respectively. It can be seen that when the video complexity index is consistent with other parameters, the larger the audio complexity index, the larger the complexity evaluation index.

[0066] Furthermore, the specific process of obtaining the risk assessment execution processing method of each effective sub-mental health data is: based on the complexity assessment index of each effective sub-mental health data, and obtaining the preset complexity assessment index threshold, the complexity assessment index of each effective sub-mental health data is compared with the complexity assessment index threshold. After comparison, if the complexity assessment index of a certain effective sub-mental health data is less than or equal to the complexity assessment index threshold, then the risk assessment execution method of the effective sub-mental health data is defined as the automatic processing method; if the complexity assessment index of a certain effective sub-mental health data is greater than the complexity assessment index threshold, then the risk assessment execution method of the effective sub-mental health data is defined as the manual processing method.

[0067] In this embodiment, the preset complexity evaluation index threshold is obtained. , when the complexity evaluation index is less than or equal to the complexity evaluation index threshold, that is , the risk assessment execution mode of the effective sub-mental health data is defined as the automatic processing mode. When the complexity assessment index is greater than the complexity assessment index threshold, that is, , the risk assessment execution method of the effective sub-mental health data is defined as a manual processing method.

[0068] Furthermore, the specific process of transmitting each valid sub-mental health data to the corresponding risk assessment execution location is: according to the risk assessment execution processing method of each valid sub-mental health data, if the risk assessment execution method of the valid sub-mental health data is an automatic processing method, then the valid sub-mental health data is transmitted to the automatic processing location; if the risk assessment execution method of the valid sub-mental health data is a manual processing method, then the valid sub-mental health data is transmitted to the manual processing location.

[0069] In this embodiment, the automatic processing method is that the artificial intelligence platform automatically processes the effective sub-mental health data, and automatically conducts a mental health risk assessment on the effective sub-mental health data, specifically: using natural language processing tools (NLP), text block tools (TextBlob) and statistical analysis tools (SAS) to process the effective sub-mental health data; when the risk assessment execution method of a certain effective sub-mental health data is the automatic processing method, the effective sub-mental health data is transmitted to the automatic processing location to wait for the allocation of a processor; when the risk assessment execution method of a certain effective sub-mental health data is the manual processing method, the effective sub-mental health data is transmitted to the manual processing location for manual processing; the effective sub-mental health data is sent to the relevant person in charge and the relevant person in charge is reminded to directly conduct a mental health risk assessment, or the relevant person in charge is reminded to further process the effective sub-mental health data, and then the effective sub-mental health data is transmitted to the automatic processing location.

[0070] Furthermore, the specific process of assigning the automated target processing location for mental health risk assessment is as follows: statistically analyze all valid sub-mental health data whose risk assessment execution location is the automated processing location, mark them as each target automated processing psychological data, and statistically analyze the complexity assessment indicators of each target automated processing psychological data for mean processing to obtain the average data complexity of the target automated processing psychological data, match them with the automated processing control sets corresponding to the average data complexity intervals preset in the database, and obtain the automated processing control set corresponding to the interval where the average data complexity of the target automated processing psychological data is located, recorded as the designated automated processing control set; based on the available cloud processors in the automated processing location, extract the operating data of each available cloud processor, and perform comprehensive processing with the designated automated processing control set to obtain the allocation priority index of each available cloud processor, select the available cloud processor corresponding to the maximum allocation priority index, and use it as the automated target processing location for mental health risk assessment.

[0071] In this embodiment, the method for obtaining the allocation priority index of each available cloud processor is:

[0072] ;

[0073] In the formula, Expressed as The allocation priority index of available cloud processors, n , is the total number of available cloud processors, Expressed as The real-time processing rate of available cloud processors, Expressed as the control treatment rate, Expressed as real-time processing bandwidth of available cloud processors, is represented as the control processing bandwidth, Expressed as real-time processing throughput of available cloud processors, Expressed as control processing throughput, Expressed as The length of the queue of pending tasks for available cloud processors, It is expressed as the length threshold of the queue of pending tasks. Expressed as Real-time CPU usage of available cloud processors, Expressed as real-time CPU usage threshold, Expressed as The average processing response time of available cloud processors, Expressed as the average processing response time threshold, Expressed as the real-time processing rate weight ratio, Expressed as the weight ratio of real-time processing bandwidth, Expressed as the weight ratio of real-time processing throughput, It is expressed as the weight ratio of the length of the queue of pending tasks. Expressed as the real-time CPU usage weight ratio, It is expressed as a weighted ratio of an average processing response time, and the control processing rate, control processing bandwidth, control processing throughput, pending task queue length threshold, real-time CPU usage threshold, and average processing response time threshold are obtained from a database.

[0074] In this embodiment, , , , , and The influence weight ratios of the preset real-time processing rate, real-time processing bandwidth, real-time processing throughput, length of the pending task queue, real-time CPU usage and average processing response time are obtained from the preset database respectively, and are used to indicate the numerical values ​​of the influence of the real-time processing rate, real-time processing bandwidth, real-time processing throughput, length of the pending task queue, real-time CPU usage and average processing response time on the allocation priority index of each available cloud processor. When used, the influence weight ratios corresponding to the real-time processing rate, real-time processing bandwidth, real-time processing throughput, length of the pending task queue, real-time CPU usage and average processing response time can be directly obtained from the preset database, and the corresponding relationship can be a preset mapping relationship For example, the real-time processing rate, real-time processing bandwidth, real-time processing throughput, length of the pending task queue, real-time CPU usage and average processing response time form a mapping set with the weights corresponding to the real-time processing rate, real-time processing bandwidth, real-time processing throughput, length of the pending task queue, real-time CPU usage and average processing response time in the preset database. After the real-time processing rate, real-time processing bandwidth, real-time processing throughput, length of the pending task queue, real-time CPU usage and average processing response time are input into the mapping set, the corresponding real-time processing rate, real-time processing bandwidth, real-time processing throughput, length of the pending task queue, real-time CPU usage and average processing response time weight ratio will be obtained. In this example, , , , , and The value range is , , , , , and The sum of is 1.

[0075] In this embodiment, there is a correlation between the real-time processing rate, the real-time processing bandwidth, the real-time processing throughput, the length of the pending task queue, the real-time CPU usage and the average processing response time. When the real-time processing bandwidth increases, the real-time processing rate of the processor will increase, thereby increasing the real-time processing bandwidth. When the length of the pending task queue increases, the CPU needs to process more tasks, the CPU usage of the location will increase, and the response time of the location will also increase accordingly. When the control processing rate becomes lower, the processor cannot process the arrived tasks in time, and the length of the pending task queue will also increase.

[0076] Furthermore, the automatic processing control set includes: control processing rate, control processing bandwidth and control processing throughput; the operating data of each available cloud processor includes: real-time processing rate, real-time processing bandwidth, real-time processing throughput, length of the pending task queue, real-time CPU usage and average processing response time.

[0077] In this embodiment, the control processing rate is the amount of data that the processor can process per unit time, the control processing bandwidth is the amount of data that the processor can transmit per unit time, and the control processing throughput is the amount of work completed by the processor per unit time. The control processing rate and the control processing throughput are obtained through a performance monitoring tool, and the control processing bandwidth is obtained through a network traffic analysis tool.

[0078] The length of the pending task queue is the number of tasks waiting to be processed in the system, the local CPU usage is the ratio of the CPU time occupied by a specific process or service to the total CPU time, and the local response time is the time required from sending a request to receiving a response. The length of the pending task queue and the local CPU usage are obtained through the operating system monitoring tool, and the local response time can be obtained through the application performance monitoring tool.

[0079] like Figure 3 As shown, it is a flow chart of the campus mental health risk assessment method based on artificial intelligence provided in the embodiment of the present application. The campus mental health risk assessment method based on artificial intelligence provided in the embodiment of the present application includes the following steps: obtaining the campus mental health data transmitted by the data source for preprocessing, and screening to obtain each valid sub-mental health data; based on each valid sub-mental health data, analyzing to obtain the risk assessment execution processing method of each valid sub-mental health data; based on the risk assessment execution processing method of each valid sub-mental health data, transmitting each valid sub-mental health data to the corresponding risk assessment execution location; when the risk assessment execution location of the valid sub-mental health data is an automated processing location, allocating the automated target processing location for mental health risk assessment processing.

[0080] In summary, the embodiment of the present application obtains campus mental health data transmitted by a data source, screens out each valid sub-mental health data, determines different risk assessment processing methods according to the complexity of each valid sub-mental health data, and transmits the valid sub-mental health data with an automatic risk assessment processing method to a suitable place of origin for processing, thereby achieving accurate and effective acquisition and processing of high-quality data, and effectively solving the problem of difficulty in accurately and effectively acquiring and processing psychological data in the prior art.

[0081] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0082] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0083] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0084] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0085] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0086] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. The campus mental health risk assessment platform based on artificial intelligence is characterized by: include: The mental health data preprocessing module is used to obtain the campus mental health data transmitted by the data source for preprocessing and screen out each valid sub-mental health data; A mental health data analysis module is used to analyze each valid sub-mental health data and obtain a risk assessment execution processing method for each valid sub-mental health data; A mental health data transmission module, used for executing a processing method based on the risk assessment of each valid sub-mental health data, and transmitting each valid sub-mental health data to a corresponding risk assessment execution location; A risk assessment execution location allocation module, used for allocating an automated target processing location for mental health risk assessment processing when the risk assessment execution location of valid child mental health data is an automated processing location; The specific process of the risk assessment execution processing method based on each valid sub-mental health data and analyzing each valid sub-mental health data is as follows: Each valid sub-mental health data includes each valid psychological interaction video and the audio corresponding to each valid psychological interaction video; Based on each effective psychological interaction video, the video duration, video byte count, video frame rate and video bit rate are extracted and processed to obtain the video processing complexity index of each effective psychological interaction video; Based on the audio corresponding to each valid psychological interaction video, the number of channels, the number of audio bytes, the audio sampling rate and the audio dynamic range are extracted and processed to obtain the audio processing complexity index corresponding to each valid psychological interaction video; Based on the quality characteristic index of each psychological interaction video, the quality characteristic index of each effective psychological interaction video is extracted, and the video processing complexity index of each effective psychological interaction video, the audio processing complexity index corresponding to each effective psychological interaction video and the quality characteristic index of each effective psychological interaction video are comprehensively processed to obtain the complexity evaluation index of each effective sub-mental health data, and the complexity evaluation index is used to measure the processing difficulty of each effective sub-mental health data; Based on the complexity evaluation index of each valid sub-mental health data, a risk evaluation execution processing method of each valid sub-mental health data is obtained; The specific process of obtaining the risk assessment execution processing method of each valid sub-mental health data is: based on the complexity assessment index of each valid sub-mental health data, and obtaining a preset complexity assessment index threshold, the complexity assessment index of each valid sub-mental health data is compared with the complexity assessment index threshold. After comparison, if the complexity assessment index of a certain valid sub-mental health data is less than or equal to the complexity assessment index threshold, then the risk assessment execution method of the valid sub-mental health data is defined as an automatic processing method; if the complexity assessment index of a certain valid sub-mental health data is greater than the complexity assessment index threshold, then the risk assessment execution method of the valid sub-mental health data is defined as a manual processing method; The specific process of transmitting each valid sub-mental health data to the corresponding risk assessment execution location is: according to the risk assessment execution processing method of each valid sub-mental health data, if the risk assessment execution method of the valid sub-mental health data is an automatic processing method, then the valid sub-mental health data is transmitted to the automatic processing location; if the risk assessment execution method of the valid sub-mental health data is a manual processing method, then the valid sub-mental health data is transmitted to the manual processing location; The specific process of assigning the automated target processing location to carry out mental health risk assessment processing is as follows: counting all valid sub-mental health data whose risk assessment execution location is the automated processing location, marking them as each target automated processing psychological data, counting the complexity assessment indicators of each target automated processing psychological data for mean processing to obtain the average data complexity of the target automated processing psychological data, matching them with the automated processing control sets corresponding to each average data complexity interval preset in the database, obtaining the automated processing control set corresponding to the interval where the average data complexity of the target automated processing psychological data is located, and recording them as the designated automated processing control set; Based on the available cloud processors in the automated processing location, the operating data of each available cloud processor is extracted, and the data is comprehensively processed with the specified automated processing control set to obtain the allocation priority index of each available cloud processor. The available cloud processor corresponding to the maximum allocation priority index is selected and used as the automated target processing location for mental health risk assessment.

2. The campus mental health risk assessment platform based on artificial intelligence as claimed in claim 1, characterized in that: The specific process of obtaining the campus mental health data transmitted by the data source and preprocessing it is as follows: The campus mental health data transmitted by the data source includes each mental interaction video transmitted by the data source and the audio corresponding to each mental interaction video; Extracting basic feature data of each psychological interaction video, wherein the basic feature data of each psychological interaction video includes: video resolution, video frame rate and video compression rate; Extracting audio feature data corresponding to each psychological interaction video, wherein the audio feature data includes: audio sampling rate, audio signal-to-noise ratio and audio bit rate; Based on the basic feature data of each psychological interaction video and the audio feature data corresponding to each psychological interaction video, a quality feature index of each psychological interaction video is analyzed and obtained, and the quality feature index is used to measure the quality of each psychological interaction video.

3. The campus mental health risk assessment platform based on artificial intelligence as claimed in claim 2, characterized in that: The quality characteristic index of each psychological interaction video is specifically obtained by: ； In the formula, Expressed as Quality characteristic index of psychological interaction videos, , is the total number of psychological interaction videos, Expressed as video resolution, Expressed as the video resolution threshold, Expressed as video frame rate, Expressed as the video frame rate threshold, Expressed as the video compression rate, Expressed as the video compression rate threshold, Expressed as the audio sampling rate, Expressed as the audio sampling rate threshold, Expressed as the audio signal-to-noise ratio, Expressed as the audio signal-to-noise ratio threshold, Expressed as audio bit rate, Expressed as the audio bitrate threshold, Expressed as the video resolution weight ratio, Expressed as the video frame rate weight ratio, Expressed as the weight ratio of video compression rate, Expressed as the audio sampling rate weight ratio, Expressed as the audio signal-to-noise ratio weight ratio, Expressed as the audio bit rate weight ratio, Expressed as The weight ratio of the basic feature data of the psychological interaction video, Expressed as The weight ratio of audio feature data corresponding to the psychological interaction video, is a natural constant.

4. The campus mental health risk assessment platform based on artificial intelligence as claimed in claim 3 is characterized in that: The specific process of screening and obtaining each valid sub-mental health data is called: Obtain a preset quality characteristic index threshold. If the quality characteristic index of a psychological interaction video is less than or equal to the quality characteristic index threshold, the psychological interaction video is defined as unqualified. If the quality characteristic index of a psychological interaction video is greater than the quality characteristic index threshold, the psychological interaction video is defined as qualified, and the psychological interaction video is recorded as a valid psychological interaction video. The valid psychological interaction video and the audio corresponding to the valid psychological interaction video are jointly defined as valid sub-psychological health data, thereby obtaining statistics of each valid sub-psychological health data.

5. The campus mental health risk assessment platform based on artificial intelligence as claimed in claim 1, characterized in that: The automatic processing control set includes: control processing rate, control processing bandwidth and control processing throughput; The operating data of each available cloud processor includes: real-time processing rate, real-time processing bandwidth, real-time processing throughput, length of pending task queue, real-time CPU usage and average processing response time.

6. A method for applying the artificial intelligence-based campus mental health risk assessment platform as described in any one of claims 1 to 5, characterized in that: The following steps are involved: Obtain campus mental health data transmitted by the data source for preprocessing, and screen out each valid sub-mental health data; Based on each valid sub-mental health data, analyze and obtain the risk assessment execution and processing method of each valid sub-mental health data; Based on the risk assessment execution processing method of each valid sub-mental health data, each valid sub-mental health data is transmitted to the corresponding risk assessment execution location; When the risk assessment execution location of effective sub-mental health data is the automated processing location, the automated target processing location is assigned to carry out mental health risk assessment processing.

Images