CN112185493A - Personality preference diagnosis device and project recommendation system based on same - Google Patents
Personality preference diagnosis device and project recommendation system based on same Download PDFInfo
- Publication number
- CN112185493A CN112185493A CN202010873535.5A CN202010873535A CN112185493A CN 112185493 A CN112185493 A CN 112185493A CN 202010873535 A CN202010873535 A CN 202010873535A CN 112185493 A CN112185493 A CN 112185493A
- Authority
- CN
- China
- Prior art keywords
- classifier
- type
- features
- training
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H10/00—ICT specially adapted for the handling or processing of patient-related medical or healthcare data
- G16H10/60—ICT specially adapted for the handling or processing of patient-related medical or healthcare data for patient-specific data, e.g. for electronic patient records
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H20/00—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance
- G16H20/70—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to mental therapies, e.g. psychological therapy or autogenous training
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H70/00—ICT specially adapted for the handling or processing of medical references
- G16H70/60—ICT specially adapted for the handling or processing of medical references relating to pathologies
Abstract
The application discloses personality preference diagnostic device and project recommendation system based on the same includes: an acquisition module configured to: acquiring physiological signals of a person to be diagnosed and basic information of the person to be diagnosed; a feature extraction module configured to: processing the physiological signal to obtain a first class of characteristics; processing the electroencephalogram signals in the physiological signals to obtain a second type of characteristics; processing the basic information of the personnel to be diagnosed to obtain a third type of characteristics; a feature fusion module configured to: fusing the first type of features, the second type of features and the third type of features to obtain fused features; a personality preference diagnostic module configured to: and inputting the fusion features into a pre-trained classifier, and outputting a personality preference diagnosis result.
Description
Technical Field
The application relates to the technical field of artificial intelligence mode recognition, in particular to a personality preference diagnosis device and a project recommendation system based on the same.
Background
The statements in this section merely provide background information related to the present disclosure and may not constitute prior art.
Personality preference (personality), also known as personality, refers to the internal tendencies and psychological characteristics of an individual in its behavior in social adaptation to others, events, oneself, and the like. The expression is the integration of abilities, temperament, characters, needs, motivations, interests, idealities, value and physique, and is self with dynamic consistency and continuity, and is a unique psychosomatic tissue formed by individuals in the socialization process. Accurate personality trait analysis has important instructive value in real life, such as making a personalized thought and behavior modification scheme for prisoners, establishing an ideal college student employment guidance selection strategy according to the character characteristics, making clear and targeted psychological dispersion routes for different professional groups, and clinically guiding the revision of an auxiliary psychological rehabilitation scheme for acute and severe patients or chronic patients. Personality assessment is the quantitative analysis of psychotropism and behavioral tendencies that play a stabilizing regulatory role in a person's behavior by specific methods to predict and guide the person's future behavior.
The american psychologist Woodworth in 1918 compiled the world's first self-aged personality scale, which became the beginning of emotional psychological studies by the self-aged scale method. The Meiers-Bridgs type index (MBTI) is compiled on the basis of 8 personality types divided by Rongge of Swiss psychologists, and has four different psychological characteristic dimensions, as shown in figure 1, the personality quantization indexes of a testee fall on a certain point of a scale, and the close end points indicate the personality preference of the testee in which aspect. Statement information gathering is done in questionnaires, with subjects responding to questions such as: 1. is you good at and doing a particular job? 2. Can you schedule things to do? 3. Can you act after planning? And selecting an evaluation score between 0 and 5 according to the sensitivity intensity, and performing score accumulation processing on different dimensions to judge the personality preference by a quantification means. By adopting the scheme to carry out psychological measurement on the same target individual, the conclusion of the questionnaire has good stability and consistency, the personality traits of the measured individual can be accurately reflected, and the questionnaire has good performance on quantization indexes such as stability coefficients, equivalent coefficients and intrinsic consistency coefficients. Meanwhile, the psychological diagnosis conclusion obtained by measurement is proved to be higher in goodness of fit with the actual state of an individual to be examined in the long-term use process, and the method is good in accuracy and usefulness and has ideal effectiveness. At present, the MBTI means is widely adopted in the field of psychological research at home and abroad and is cited by a plurality of documents, so that the MBTI means becomes the most popular personality assessment tool at international and becomes one of the gold standards for personality preference analysis.
On the other hand, MBTI is difficult to get rid of inherent defects of psychological personality diagnosis in a self-statement mode, such as fuzzy comprehension of questions in questionnaire answering, subjective camouflage of testers due to various reasons, incapability of adapting to economic and social development due to lack of a reasonable updating mechanism for a long time in standard design, lack of sufficient theory and large-scale data support for various revisions and the like, and all the defects can cause the reduction of reliability and effectiveness. The experimental method is another important way in the field of psychological research, and the emotion of a testee is induced by audio-visual stimulation, the physiological behavior characteristics of the testee are synchronously obtained, the internal information is mined, and the emotional state or the psychological characteristics of the testee are analyzed and evaluated from different angles. The objectivity of the method is remarkably improved compared with a self-aging mode, but the method has the defects that effective data and reasonable labels are lacked, physiological signals are weak and are easily interfered by noise, the optimal process of an emotion exciting strategy and a feature extraction algorithm is complex, and the bottleneck of performance improvement is difficult to overcome due to the lack of an ideal multi-source information fusion strategy, so that novel psychological research means combining the self-aging test mode gradually gets extensive attention of the academic world. Supervised learning is the mainstream machine learning method, and samples with known certain characteristics or certain characteristics are used as a training set, training data is analyzed to establish a mathematical model (logistic regression, linear regression and the like), and a specific inference function is generated to predict unknown samples.
Disclosure of Invention
In order to solve the defects of the prior art, the application provides a personality preference diagnosis device and a project recommendation method and system based on the personality preference diagnosis device;
in a first aspect, the present application provides a personality preference diagnostic apparatus;
a personality preference diagnostic apparatus comprising:
an acquisition module configured to: acquiring physiological signals of a person to be diagnosed and basic information of the person to be diagnosed;
a feature extraction module configured to: processing the physiological signal to obtain a first class of characteristics; processing the electroencephalogram signals in the physiological signals to obtain a second type of characteristics; processing the basic information of the personnel to be diagnosed to obtain a third type of characteristics;
a feature fusion module configured to: fusing the first type of features, the second type of features and the third type of features to obtain fused features;
a personality preference diagnostic module configured to: and inputting the fusion features into a pre-trained classifier, and outputting a personality preference diagnosis result.
In a second aspect, the present application provides an item recommendation system;
an item recommendation system comprising:
an acquisition module configured to: acquiring physiological signals of a person to be diagnosed and basic information of the person to be diagnosed;
a feature extraction module configured to: processing the physiological signal to obtain a first class of characteristics; processing the electroencephalogram signals in the physiological signals to obtain a second type of characteristics; processing the basic information of the personnel to be diagnosed to obtain a third type of characteristics;
a feature fusion module configured to: fusing the first type of features, the second type of features and the third type of features to obtain fused features;
a personality preference diagnostic module configured to: inputting the fusion characteristics into a classifier trained in advance, and outputting a personality preference diagnosis result;
an item recommendation module configured to: and performing distance calculation on the personality preference diagnosis result and the fusion characteristic of the same personality preference diagnosis result in the same population of the personnel to be diagnosed, screening the item of the personnel corresponding to the minimum distance, and recommending the item as the optimal item to the personnel to be diagnosed.
In a third aspect, the present application further provides an electronic device, including: one or more processors, one or more memories, and one or more computer programs; wherein a processor is connected to the memory, the one or more computer programs are stored in the memory, and when the electronic device runs, the processor executes the one or more computer programs stored in the memory, so as to make the electronic device execute the functions of the modules according to the first aspect or the second aspect.
In a fourth aspect, the present application further provides a computer-readable storage medium for storing computer instructions, which when executed by a processor, perform the functions of the modules of the first or second aspect.
Compared with the prior art, the beneficial effects of this application are:
(1) personality tendency analysis is an important subject in the field of psychological research, and has wide application prospects in the aspect of ideological and behavioral rehabilitation and correction guided by psychological traits, such as important values in the aspects of modification of prisoners, clinical rehabilitation psychological assistance, university student employment guidance, professional crowd psychological health care and the like. The traditional personality tendency analysis still takes a subjective question-answering mode as a main mode, quantitative analysis is carried out on different personality dimensions according to a scoring mode of an international universal classical questionnaire, and the MBTI standard is a widely accepted measure with good reliability and effectiveness and is also adopted in practice in a large quantity. However, the psychological research strategy of the self-statement mode has strong subjectivity and interference of a plurality of factors in the operation process, and the performance is reduced to a certain extent. The method combines a psychological research mode of an experimental method, uses emotion excitation as a means, uses multi-source information as data support, uses a classic self-aging table conclusion as a labeling mode, establishes a data set with physiological and psychological correlation, adopts a proper machine learning algorithm to train a more objective and accurate personality trait intelligent evaluation system, effectively integrates two different emotion analysis means of experiment and self-aging, and realizes advantage complementation on the basis of overcoming the defect of a single mode.
(2) And a feature fusion strategy is adopted to associate, correlate and synthesize the data and the information acquired from a plurality of information sources so as to obtain more accurate personality feature estimation. By standardizing and processing the uniform dimension, the problem of overlarge weight of individual characteristics caused by inconsistent units is avoided. Based on multiple sources and isomerism of physiological information, brain electrical information, personal identity information (age, sex, culture hierarchy and the like)The method is characterized in that the characteristics of standard deviation, difference, sample entropy, wavelet transformation and the like are extracted in a targeted manner, and then the characteristics are comprehensively combed and integrated to establish a fusion sample characteristic space { gamma1，γ2，γ3In which is γ1Being a physiological signal feature subspace, gamma2Is a subspace of features of the electroencephalogram signal, gamma3A background information feature subspace. The method combines the spatial redundancy or the temporal redundancy or the complementation of multi-source information according to a certain criterion to support the identification and classification work aiming at the character types and obtain the consistency explanation and description of the tested object.
(3) Only if the distribution rule of the observed data is accurately grasped, the matched pattern recognition algorithm can be selected, and the ideal classification performance is obtained. With regard to the complexity of multi-modal fusion type feature set data and the unknown distribution rule thereof, on the basis of a traditional linear support vector machine classifier (SVM), as shown in FIG. 6B, the application introduces the transformation Z phi (x) to map a nonlinear feature space to a higher-dimensional linear space, learns a classification model from training data in a new feature space by using the linear SVM, and establishes a novel classifier which can support the completion of the recognition of the nonlinear distribution data. Through a performance optimization mechanism, linear or nonlinear classifiers are selected according to data distribution characteristic adaptability and combined into four unrelated secondary classifiers, diagnosis of personality preference of four different dimensions is completed, and the limitation that a traditional SVM is only suitable for completing a linear secondary classification task is effectively overcome.
(4) The character type matching according to the judgment conclusion is carried out by the Hash table rule, and the time efficiency of searching is effectively improved.
(5) And intelligently recommending psychological diagnosis derivative products. According to the method, a similarity recommendation algorithm is adopted, as shown in fig. 7, according to the personality intelligent analysis conclusion of a subject, the closest point on a two-dimensional space coordinate is calculated through quantification modes such as Euclidean distance and Machattan distance by combining social attributes (prisoners, college students, patients, staff and the like), and psychological diagnosis derivative products such as a personalized psychological behavior correction scheme or an occupation planning scheme and the like which are suitable for the needs of the subject are recommended. If the college graduates are intelligently analyzed through personality preference, employment recommendations matched with the characters of the college graduates can be obtained, and employment work is scientifically guided; if prisoners are intelligently analyzed by personality preference, more optimized thought and behavior modification strategies adaptive to the characters of the prisoners can be obtained, and scientific references are provided for the development of modification work of police officers; if the chronic disease patient is intelligently analyzed through personality preference, a set of mental health nursing scheme suitable for the psychological traits of the chronic disease patient can be formed, a doctor is guided to further take scientific mental health nursing auxiliary supporting measures while carrying out pathological treatment, and the rehabilitation effect is improved.
(6) And (4) a quality control system for clear quantification. The performance evaluation of the psychological diagnosis conclusion is mainly based on the subjective feeling of the testee, although some diagnosis modes have various credibility and effectiveness evaluation indexes, the unified standard still lacks, and the objective accuracy of the evaluation can not obtain the consistency between the doctors and patients (testees). As shown in fig. 8, in the present application, a cross-validation strategy based on a data set is adopted in a reliability assessment level, and a confusion matrix strategy aiming at an actual application scenario is adopted in a validity assessment level, so that the conclusion accuracy of intelligent personality preference analysis is quantized, and the system obtains better trust.
(7) The method is based on the basic theory of neuroreflexion, a characteristic space of a sample is formed by collecting various human physiological signals under a situation inducing environment and combining comprehensive information such as the sex, age, academic calendar and the like of a collected person after digitization, meanwhile, the MBTI standard is utilized to carry out multi-dimensional personality trait labeling on the same population, a complete supervised learning data set is established, self statement and an experiment are effectively fused by two different psychological research means through a proper machine learning algorithm, and a higher-level personality preference intelligent evaluation system is established.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this application, illustrate embodiments of the application and, together with the description, serve to explain the application and are not intended to limit the application.
FIG. 1 is a block diagram of a "Myers-Bridgs type index" (MBTI) framework according to an embodiment of the present application;
FIG. 2 is a flowchart of a personality analysis dataset construction based on multi-modal information according to a first embodiment of the present application;
FIG. 3 is a logical framework diagram of a personality analysis dataset based on multimodal information according to an embodiment of the present application;
FIG. 4 is a diagram of a feature fusion structure according to a first embodiment of the present application;
FIG. 5 is a framework diagram of an application structure of a personality preference intelligent diagnosis system according to a first embodiment of the present application;
FIG. 6 is a flowchart of a personality preference intelligent diagnosis system decision making process according to a first embodiment of the present application;
FIG. 7 is a diagram of intelligent recommendation of psychological diagnosis derivative products according to a first embodiment of the present application;
FIG. 8 is a schematic diagram of a training set and a test set according to a first embodiment of the present application;
fig. 9 is a schematic diagram of system validity reliability quantification assessment in the first embodiment of the present application.
Detailed Description
It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the disclosure. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments according to the present application. As used herein, the singular forms "a", "an", and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise, and it should be understood that the terms "comprises" and "comprising", and any variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
In the embodiment of the present application, "and/or" is only one kind of association relation describing an association object, and means that three kinds of relations may exist. For example, a and/or B, may represent: a exists alone, A and B exist simultaneously, and B exists alone. In addition, in the description of the present application, "a plurality" means two or more than two.
In addition, in order to facilitate clear description of technical solutions of the embodiments of the present application, in the embodiments of the present application, terms such as "first" and "second" are used to distinguish the same items or similar items having substantially the same functions and actions. Those skilled in the art will appreciate that the words "first", "second", etc. do not necessarily define a quantity or order of execution and that the words "first", "second", etc. do not necessarily differ.
The embodiments and features of the embodiments in the present application may be combined with each other without conflict.
Example one
The present embodiment provides a personality preference diagnosis apparatus;
a personality preference diagnostic apparatus comprising:
an acquisition module configured to: acquiring physiological signals of a person to be diagnosed and basic information of the person to be diagnosed;
a feature extraction module configured to: processing the physiological signal to obtain a first class of characteristics; processing the electroencephalogram signals in the physiological signals to obtain a second type of characteristics; processing the basic information of the personnel to be diagnosed to obtain a third type of characteristics;
a feature fusion module configured to: fusing the first type of features, the second type of features and the third type of features to obtain fused features;
a personality preference diagnostic module configured to: and inputting the fusion features into a pre-trained classifier, and outputting a personality preference diagnosis result.
As one or more embodiments, the acquiring a physiological signal of a person to be diagnosed includes: electrocardiosignals, respiratory data, facial blood oxygen content, skin electrical signals, blood oxygen saturation and electroencephalogram signals.
As one or more embodiments, the basic information of the person to be diagnosed includes: age, gender and cultural degree.
As one or more embodiments, the processing the physiological signal to obtain the first class of features includes:
extracting high-frequency power, standard deviation, first derivative and/or difference characteristics from the electrocardiosignal;
extracting multi-scale entropy and/or power spectral density characteristics of the skin electric signal;
extracting high-frequency power, low-frequency power, standard deviation, mean value, first derivative, difference, slope and/or power ratio from the respiratory data;
extracting standard deviation, mean value, multi-scale entropy and/or power spectral density of the blood oxygen saturation;
for facial blood oxygen content, extracting high frequency power, low frequency power, mean, first derivative, multi-scale entropy, power spectral density, slope and/or power ratio features.
As one or more embodiments, the electroencephalogram signal in the physiological signal is processed to obtain a second type of feature; the method comprises the following steps:
approximate entropy, sample entropy, power spectrum estimation and/or wavelet transformation characteristics are extracted from the electroencephalogram signals.
As one or more embodiments, the basic information of the person to be diagnosed is processed to obtain a third type of feature; the method comprises the following steps:
and acquiring the characteristics of age, gender and cultural degree of the person to be diagnosed.
As one or more embodiments, after the obtaining module, before the feature extracting module, the method further includes: a standardization processing module; the normalization processing module configured to: and carrying out unified dimension processing on the acquired data.
As one or more embodiments, the feature fusion module is configured to: and performing series fusion on the first class of features, the second class of features and the third class of features to obtain fused features.
As one or more embodiments, the fusion features are input into a pre-trained classifier, and a personality preference diagnosis result is output; the method comprises the following specific steps:
and respectively inputting the fusion features into four pre-trained two classifiers, outputting a classification result by each two classifier, and taking all classification results as final classification results.
Further, the personality preference diagnosis module comprises a pre-training sub-module of the classifier, and the pre-training sub-module of the classifier comprises:
an acquisition unit configured to: acquiring physiological signals and basic information of each tester with known Meiers Briggs Type Index (MBTI);
a feature extraction unit configured to: processing physiological signals of each tester with known Meiers Briggs Type Index (MBTI) to acquire a first type of characteristics; processing the electroencephalogram signal of each tester with known Meiers Briggs Type Index (MBTI) to acquire a second type of characteristic; processing the basic information of each tester with known Meiers Briggs Type Index (MBTI) to acquire a third type of characteristics;
a feature fusion unit configured to: fusing the first type of features, the second type of features and the third type of features to obtain fused features;
a data set building unit configured to: obtaining a data set after the fusion characteristics of the testers correspond to known Meiers Briggs Type Indexes (MBTI) of the testers one by one, and dividing the data set into a training set and a test set according to a proportion;
a first training unit configured to: training a linear support vector machine by using the data with the out-of-band and in-tilt dimension labels in the training set to obtain a first classifier, and training a nonlinear support vector machine by using the data with the out-of-band and in-tilt dimension labels in the training set to obtain a second classifier; respectively inputting data with the label of the camber and inclination dimensions into a first classifier and a second classifier by using a test set, comparing the classification precision of the first classifier and the second classifier, and taking the classifier with high classification precision as a second classifier for judging the camber and inclination dimensions;
a second training unit configured to: training the linear support vector machine by using the data with the sensory intuitive dimension labels in the training set to obtain a third classifier, and training the nonlinear support vector machine by using the data with the sensory intuitive dimension labels in the training set to obtain a fourth classifier; respectively inputting data of the test set with the sensory intuitive dimension labels into a third classifier and a fourth classifier, comparing the classification precision of the third classifier and the fourth classifier, and taking the classifier with high classification precision as a sensory intuitive dimension judgment second classifier;
a third training unit configured to: training the linear support vector machine by using data with thinking emotion dimension labels in the training set to obtain a fifth classifier, and training the nonlinear support vector machine by using data with thinking emotion dimension labels in the training set to obtain a sixth classifier; respectively inputting data with thought emotion dimension labels in the test set into a fifth classifier and a sixth classifier, comparing the classification precision of the fifth classifier and the sixth classifier, and taking the classifier with high classification precision as a thought emotion dimension judgment second classifier;
a fourth training unit configured to: training the linear support vector machine by using data with a perceptual dimension label judged in the training set to obtain a seventh classifier, and training the nonlinear support vector machine by using data with a perceptual dimension label judged in the training set to obtain an eighth classifier; and respectively inputting the data of the perceptual dimension judgment labels into a seventh classifier and an eighth classifier by using the test set, comparing the classification precision of the seventh classifier and the eighth classifier, and taking the classifier with high classification precision as a perceptual dimension judgment two classifier.
Music-induced emotion is a recognized fact in academia, research on physiological mechanisms of music emotion has achieved considerable results, and research on behavioral tendency and autonomic neurophysiological response of music emotion has become a hot point of attention in the current psychological field. The music background (Western classical music, Chinese traditional string music and the like) is used as a basic mode of mood excitation, a calm and gentle atmosphere is created, and the interference of external environmental factors can be reduced to the maximum extent. Approximately 300 volunteers were recruited to complete the dataset construction work.
As shown in fig. 2, the sample space includes a feature subspace and a psychological labeling subspace, the feature sources include multi-source physiological information in an emotion excited state, such as electrocardiogram, respiratory rate, facial blood oxygen content, electroencephalogram, and the like, and digitally processed background information of the subject (the academic calendar hierarchy is respectively expressed as integer types of 1,2, 3, and the like according to primary school, junior middle school, senior middle school, university, and the like, the male is expressed as 1, and the female is expressed as 2), the size of the subject is kept uniform in all aspects of gender, age, cultural degree, and the like, and diversity of data is achieved.
As shown in FIG. 3, the structure of the labeling space adopts a multi-labeling mode, and the labeling is finished by performing quantitative processing on four different dimensions of camber-camber (E-I), feeling-intuition (S-N), thinking-emotion (T-F) and judgment-perception (J-P) according to self-aging information of a subject and an MBTI personality type index algorithm rule by taking an MBTI template as a standard.
Indexes such as electrocardio (-1.5mA-1.5mA), skin current (0 mus-25 mus), respiration (-50% -50%), blood oxygen (0 mus A-1.2 mus), facial blood oxygen content (0 mus A-0.15 mus A), electroencephalogram and the like in physiological data and background information indexes after digitization are different in unit and value range, inconsistency of sample characteristic dimensions can cause certain characteristics to form a leading effect, and normalization processing is carried out based on the requirement of information fusion:
wherein xmaxIs the maximum value of the same type of data, xminAll data are mapped between 0 and 1 for the minimum value of the same type of data, so that the influence of abnormal values is avoided, and the integral deflection of the data is effectively improved.
As shown in FIG. 4, a feature subset 1 is established based on a plurality of features with high frequency and clear identification, such as standard deviation, difference, power spectral density and the like, used by human physiological signals in the recent emotion calculation research field.
By researching the characteristics of the electroencephalogram signals of the human body in different physiological states and different brain function states, the psychological activity characteristics of various personality individuals can be effectively explored. Firstly, the electroencephalogram signal has inherent weakness, has the characteristics of nonlinearity, non-stationarity and randomness, and meanwhile, the background noise of the acquired electroencephalogram signal is complex, and the electroencephalogram signal has non-physiological artifacts such as power frequency interference, contact noise of an electrode and the skin, common-mode signal interference between the electrode and the ground and the like, and also has physiological noise such as eye movement artifact, muscle movement artifact, electrocardio-artifact and the like, so that the analysis of the electroencephalogram signal is seriously influenced. In order to effectively reduce noise interference, as shown in fig. 4, the present application adopts an Independent Component Analysis (ICA) strategy to estimate, from observed signals, mutually statistically independent original electroencephalogram signals mixed by unknown factors. Further, the method combines a plurality of schemes adopted for electroencephalogram time-frequency domain analysis at home and abroad in recent years, extracts approximate entropy, sample entropy, power spectrum estimation, wavelet transformation and the like, and establishes the feature subset 2. The specific implementation process is expressed as follows:
A. approximate entropy
The approximate entropy is a nonlinear kinetic parameter for quantifying the regularity and unpredictability of time series fluctuation, reflects the possibility of new information in the time series, and is expressed by an algorithm as follows:
1. the electroencephalogram signals are sampled at equal time intervals to obtain an N-dimensional time sequence, u (1), u (2).
2. And defining algorithm related parameters m and r, wherein m is an integer and represents the length of the comparison vector, and r is a real number and represents a quantization value of the similarity.
3. Reconstruct an m-dimensional vector X (1), X (2).... X (N-m +1), where X (i) [ (i), u (i +1).. u (i + m-1) ].
4. For i is more than or equal to 1 and less than or equal to N-m + 1, counting the number of vectors meeting the following conditions:
d represents the distance of the vectors X (i) from X (j).
5. Definition of
6. Approximate entropy (ApEn) is defined as:
ApEn＝Φm(r)-Φm+1(r) (3)
B. sample entropy
The lower the value of the sample entropy, the higher the sequence self-similarity, and the larger the value of the sample entropy, the more complex the sample sequence. Sample entropy is currently used for evaluating the complexity of physiological time series (EEG, EMG, etc.) and for diagnosing pathological states.
1. The electroencephalogram signal is sampled at equal time intervals to obtain an N-dimensional time series { x (N) } ═ x (1), x (2.. times (N)).
2. Forming a set of vector sequences of dimension m, X, according to the sequence numbersm(1)、Xm(2)......Xm(N-m +1) wherein Xm(i) X (i + m-1) },1 ≦ i ≦ N-m +1, representing m consecutive x values starting from the ith point.
3. Defining:
d[Xm(i),Xm(j)]＝maxk＝0,1......m-1(|x(i+k)-x(j+k|) (4)
representing two vectors Xm(i) And Xm(j) The largest absolute value of the corresponding element difference.
4. For a given Xm(i) Statistics of Xm(i) And Xm(j) (j is more than or equal to 1 and less than or equal to N-m, j is not equal to i) and the number of the distances r is recorded as Bi：
5. Definition Bm(r) is:
6. adding the dimensionality to m +1, calculating Xm+1(i) And Xm+1(j) (j is more than or equal to 1 and less than or equal to N-m, j is not equal to i) and the number of the distances is less than or equal to r and is marked as Ai：
7. Defining:
8、Bm(r) represents the probability that two sequences match m points with a similarity tolerance r, Am(r) represents the probability that two sequences match m +1 points, with sample entropy defined as:
C. power spectrum estimation
The brain wave with amplitude changing along with time is converted into a spectrogram with brain electric power changing along with frequency, so that the distribution and the change of brain electric rhythm can be more clearly reflected to match different personality traits. The specific operation is to relate the power spectrum to the square of the amplitude-frequency characteristic, expressed as the ratio of the overall mean of the square of the amplitude-frequency characteristic to the duration, i.e. the limit value at which the duration tends to be infinite.
D. Wavelet transform
The wavelet transformation can highlight the characteristics of some aspects of the electroencephalogram signals, and the signals are gradually refined in a multi-scale mode through the telescopic translation operation, so that the high-frequency time subdivision and the low-frequency subdivision are achieved, different time-frequency signal analyses are automatically adapted, the local amplification of time and space frequencies is completed, and the details of the signals are focused. x (n) represents the discrete brain electrical signal acquired, defining the wavelet transform as:
wherein psij,kRepresenting wavelet basis functions, j representing frequency resolution, k representing amount of time shift, and performing finite layer on the signalAnd (3) decomposition:
the method utilizes wavelet coefficients to represent the energy distribution of the electroencephalogram signals in time domain and frequency domain, and carries out three-layer decomposition on the electroencephalogram signals, and the expression is as follows:
x(n)＝A3+D1+D2+D3 (12)
and extracting wavelet coefficient energy mean values corresponding to different frequency bands as characteristic quantities to reflect the characteristics of the time domain and the frequency domain of the electroencephalogram signals.
After the multi-modal feature space is constructed, according to the MBTI test conclusion of the subject, the personality type preference of four different dimensions is respectively marked as shown in figure 1, and the personality intelligent diagnosis problem is generalized into four mutually independent two-classification problems based on multi-source information.
The information of the age, the sex, the culture background and the like of the testee is digitalized to establish a characteristic subset 3.
As shown in FIG. 5, the application establishes a specific rule on the fused data set to guide the effective classification of individuals with different personality traits. A Support Vector Machine (SVM) is a supervised machine learning algorithm and can be widely used for classification or regression tasks. Expressing each sample as a point in n-dimensional space (n is the number of features), the value of each feature being a specific coordinate, the classification task is done by finding a hyperplane that can well distinguish the two classes.
As shown in fig. 5, according to the MBTI standard, in the four personality tendency dimensions of E-I (0/1) and S-N, T-F, J-P, according to the decision conclusion of two categories with supervised learning rules, a personality type combination such as ENTP (0101) can be obtained, and is used as a key value of a hash table, and a personality preference type corresponding to the combination is matched, and psychology analysis derivative products (such as university student personalized employment recommendation, a criminal person-specific thought behavior modification scheme, a clinical patient assisted psychology rehabilitation scheme, and the like) are intelligently recommended in combination with the demand background of a subject. The method adopts a cross validation strategy of a data set level to finish the reliability evaluation of the intelligent diagnosis system, and adopts a confusion matrix strategy to finish the validity evaluation of the intelligent system at an application level.
The personality linear support vector machine classification algorithm based on multi-mode information expresses:
the input is a training data set T { (x)1,y1),(x2,y2)......(xN,yN) Therein of
For n-dimensional fused feature vectors of the sample, yiE {0,1}, i 1,2.. N is a personality preference label in a single dimension. Selecting a penalty function C>0 constructs and solves the convex quadratic programming problem,
an objective function:
constraint conditions are as follows:
optimal solution:
computing
Selection of alpha*A component of
Satisfy the requirement of
Computing
As shown in fig. 6, the optimal classification plane is expressed as:
ω*·x+b*＝0 (15)
obtaining a classification decision function:
f(x)＝sign(ω*·x+b*) (16)
the personality nonlinear support vector machine classification algorithm based on multi-modal information expresses:
the input is a training data set T { (x)1,y1),(x2,y2)......(xN,yN) Therein of
For n-dimensional fused feature vectors of the sample, yiE {0,1}, i 1,2.. N is a personality preference label in a single dimension. Selecting proper kernel function K (x, z) and penalty function C>0 constructing and solving a convex quadratic programming problem.
An objective function:
constraint conditions are as follows:
optimal solution:
selection of alpha*A component of
Satisfy the requirement of
And (3) calculating:
as shown in fig. 6, the optimal classification plane is expressed as:
based on the performance of the linear and nonlinear classifiers on different personality dimensions of the data set, a personality preference intelligent judger formed by combining four classifiers is obtained.
The hash table is built according to the detailed definition of different personality type combinations in the MBTI standard. Performing character matching according to a decision conclusion, as shown in fig. 6, four two classifiers respectively decide 0,1,0,1, use them as key, define hash function f (key), bring the key into the function, determine addresses in the description hash tables storing different character types, map to ENPT character types, and meanwhile, make sure that the specific diagnosis conclusion is: 1. fast response, clever and longer than various affairs. 2. The two sides of the problem can be disputed for fun and the like. Based on personality preference intelligent diagnosis, psychology derivative product development facing different requirements can be further developed.
As shown in fig. 8, in the present application, a cross validation method is used to obtain a system reliability quantitative evaluation index, a data set is equally divided into ten mutually independent sub-regions, one of the sub-regions is selected as a test set each time, and the other nine sub-regions are selected as a training set to obtain test accuracy, the method is iterated for ten rounds, a mean value of the accuracy of each time is calculated, and the mean value is used as an evaluation of the degree of consistency of results obtained when the same method (combined SVM) is used to repeatedly measure the same object (personality demand multi-modal data set).
As shown in fig. 9, the system validity quantitative evaluation index is obtained by using a confusion matrix method, the conditional probability that the self-old classical method and the intelligent method are consistent in different decision conclusions is calculated, and the average value is used as the evaluation of the degree of the object to be measured which can be accurately measured by using the measurement method. Due to the inherent defects of the self-old classical method in the field of psychological evaluation, the conclusion value is maintained to be more than 70%, and the effectiveness of the intelligent diagnosis system is considered to be good.
Example two
The present embodiment provides an item recommendation system;
an item recommendation system comprising:
an acquisition module configured to: acquiring physiological signals of a person to be diagnosed and basic information of the person to be diagnosed;
a feature extraction module configured to: processing the physiological signal to obtain a first class of characteristics; processing the electroencephalogram signals in the physiological signals to obtain a second type of characteristics; processing the basic information of the personnel to be diagnosed to obtain a third type of characteristics;
a feature fusion module configured to: fusing the first type of features, the second type of features and the third type of features to obtain fused features;
a personality preference diagnostic module configured to: inputting the fusion characteristics into a classifier trained in advance, and outputting a personality preference diagnosis result;
an item recommendation module configured to: and performing distance calculation on the personality preference diagnosis result and the fusion characteristic of the same personality preference diagnosis result in the same population of the personnel to be diagnosed, screening the item of the personnel corresponding to the minimum distance, and recommending the item as the optimal item to the personnel to be diagnosed.
It should be understood that the items herein are, for example: employment schemes of college students, psychological correction schemes of prisoners, diagnosis and treatment schemes of sick patients or the work of staff.
EXAMPLE III
The present embodiment also provides an electronic device, including: one or more processors, one or more memories, and one or more computer programs; wherein, the processor is connected with the memory, the one or more computer programs are stored in the memory, and when the electronic device runs, the processor executes the one or more computer programs stored in the memory, so as to make the electronic device execute the functions of the modules in the first embodiment or the second embodiment.
Example four
The present embodiment also provides a computer-readable storage medium for storing computer instructions, and the computer instructions, when executed by a processor, implement the functions of the modules in the first embodiment or the second embodiment.
The above description is only a preferred embodiment of the present application and is not intended to limit the present application, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present application shall be included in the protection scope of the present application.
Claims (10)
1. A personality preference diagnosis device, comprising:
an acquisition module configured to: acquiring physiological signals of a person to be diagnosed and basic information of the person to be diagnosed;
a feature extraction module configured to: processing the physiological signal to obtain a first class of characteristics; processing the electroencephalogram signals in the physiological signals to obtain a second type of characteristics; processing the basic information of the personnel to be diagnosed to obtain a third type of characteristics;
a feature fusion module configured to: fusing the first type of features, the second type of features and the third type of features to obtain fused features;
a personality preference diagnostic module configured to: and inputting the fusion features into a pre-trained classifier, and outputting a personality preference diagnosis result.
2. The personal preference diagnostic apparatus of claim 1 wherein,
the acquiring of the physiological signal of the person to be diagnosed comprises: electrocardiosignals, respiratory data, facial blood oxygen content, skin electrical signals, blood oxygen saturation and electroencephalogram signals;
alternatively, the first and second electrodes may be,
the basic information of the person to be diagnosed comprises: age, gender and cultural degree.
3. The personal preference diagnostic apparatus of claim 1 wherein,
the processing of the physiological signals to obtain the first type of features comprises:
extracting high-frequency power, standard deviation, first derivative and/or difference characteristics from the electrocardiosignal;
extracting multi-scale entropy and/or power spectral density characteristics of the skin electric signal;
extracting high-frequency power, low-frequency power, standard deviation, mean value, first derivative, difference, slope and/or power ratio from the respiratory data;
extracting standard deviation, mean value, multi-scale entropy and/or power spectral density of the blood oxygen saturation;
extracting high-frequency power, low-frequency power, mean value, first derivative, multi-scale entropy, power spectral density, slope and/or power ratio characteristics of the facial blood oxygen content;
alternatively, the first and second electrodes may be,
processing the electroencephalogram signals in the physiological signals to obtain a second type of characteristics; the method comprises the following steps:
extracting approximate entropy, sample entropy, power spectrum estimation and/or wavelet transformation characteristics from the electroencephalogram signals;
alternatively, the first and second electrodes may be,
processing the basic information of the personnel to be diagnosed to obtain a third type of characteristics; the method comprises the following steps:
and acquiring the characteristics of age, gender and cultural degree of the person to be diagnosed.
4. The personal preference diagnostic apparatus as claimed in claim 1, further comprising, after the obtaining module, before the feature extraction module: a standardization processing module; the normalization processing module configured to: and carrying out unified dimension processing on the acquired data.
5. The personality preference diagnostic apparatus of claim 1, wherein the feature fusion module is configured to: and performing series fusion on the first class of features, the second class of features and the third class of features to obtain fused features.
6. The personal preference diagnostic apparatus according to claim 1, wherein the fused features are input into a classifier trained in advance, and a personal preference diagnostic result is output; the method comprises the following specific steps:
and respectively inputting the fusion features into four pre-trained two classifiers, outputting a classification result by each two classifier, and taking all classification results as final classification results.
7. The personality preference diagnostic apparatus of claim 1, wherein the personality preference diagnostic module includes a pre-training sub-module of a classifier, the pre-training sub-module of the classifier including:
an acquisition unit configured to: acquiring physiological signals and basic information of each tester with known Meiers Briggs Type Index (MBTI);
a feature extraction unit configured to: processing physiological signals of each tester with known Meiers Briggs type index MBTI to acquire a first type of characteristics; processing the electroencephalogram signal of each tester with known Meiers Briggs type index MBTI to acquire a second type of characteristic; processing the basic information of each tester with known Meiers Briggs type index MBTI to acquire a third type of characteristics;
a feature fusion unit configured to: fusing the first type of features, the second type of features and the third type of features to obtain fused features;
a data set building unit configured to: obtaining a data set after the fusion characteristics of the testers correspond to the known Meiers Briggs type index MBTI of the testers one by one, and dividing the data set into a training set and a testing set according to a proportion;
a first training unit configured to: training a linear support vector machine by using the data with the out-of-band and in-tilt dimension labels in the training set to obtain a first classifier, and training a nonlinear support vector machine by using the data with the out-of-band and in-tilt dimension labels in the training set to obtain a second classifier; respectively inputting data with the label of the camber and inclination dimensions into a first classifier and a second classifier by using a test set, comparing the classification precision of the first classifier and the second classifier, and taking the classifier with high classification precision as a second classifier for judging the camber and inclination dimensions;
a second training unit configured to: training the linear support vector machine by using the data with the sensory intuitive dimension labels in the training set to obtain a third classifier, and training the nonlinear support vector machine by using the data with the sensory intuitive dimension labels in the training set to obtain a fourth classifier; respectively inputting data of the test set with the sensory intuitive dimension labels into a third classifier and a fourth classifier, comparing the classification precision of the third classifier and the fourth classifier, and taking the classifier with high classification precision as a sensory intuitive dimension judgment second classifier;
a third training unit configured to: training the linear support vector machine by using data with thinking emotion dimension labels in the training set to obtain a fifth classifier, and training the nonlinear support vector machine by using data with thinking emotion dimension labels in the training set to obtain a sixth classifier; respectively inputting data with thought emotion dimension labels in the test set into a fifth classifier and a sixth classifier, comparing the classification precision of the fifth classifier and the sixth classifier, and taking the classifier with high classification precision as a thought emotion dimension judgment second classifier;
a fourth training unit configured to: training the linear support vector machine by using data with a perceptual dimension label judged in the training set to obtain a seventh classifier, and training the nonlinear support vector machine by using data with a perceptual dimension label judged in the training set to obtain an eighth classifier; and respectively inputting the data of the perceptual dimension judgment labels into a seventh classifier and an eighth classifier by using the test set, comparing the classification precision of the seventh classifier and the eighth classifier, and taking the classifier with high classification precision as a perceptual dimension judgment two classifier.
8. An item recommendation system, comprising:
an acquisition module configured to: acquiring physiological signals of a person to be diagnosed and basic information of the person to be diagnosed;
a feature extraction module configured to: processing the physiological signal to obtain a first class of characteristics; processing the electroencephalogram signals in the physiological signals to obtain a second type of characteristics; processing the basic information of the personnel to be diagnosed to obtain a third type of characteristics;
a feature fusion module configured to: fusing the first type of features, the second type of features and the third type of features to obtain fused features;
a personality preference diagnostic module configured to: inputting the fusion characteristics into a classifier trained in advance, and outputting a personality preference diagnosis result;
an item recommendation module configured to: and performing distance calculation on the personality preference diagnosis result and the fusion characteristic of the same personality preference diagnosis result in the same population of the personnel to be diagnosed, screening the item of the personnel corresponding to the minimum distance, and recommending the item as the optimal item to the personnel to be diagnosed.
9. An electronic device, comprising: one or more processors, one or more memories, and one or more computer programs; wherein a processor is connected to the memory, the one or more computer programs being stored in the memory, the processor executing the one or more computer programs stored in the memory when the electronic device is running, to cause the electronic device to perform the functions of the module of any of the preceding claims 1-8.
10. A computer-readable storage medium storing computer instructions which, when executed by a processor, perform the functions of the module of any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010873535.5A CN112185493A (en) | 2020-08-26 | 2020-08-26 | Personality preference diagnosis device and project recommendation system based on same |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010873535.5A CN112185493A (en) | 2020-08-26 | 2020-08-26 | Personality preference diagnosis device and project recommendation system based on same |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112185493A true CN112185493A (en) | 2021-01-05 |
Family
ID=73925812
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010873535.5A Pending CN112185493A (en) | 2020-08-26 | 2020-08-26 | Personality preference diagnosis device and project recommendation system based on same |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112185493A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113010783A (en) * | 2021-03-17 | 2021-06-22 | 华南理工大学 | Medical recommendation method, system and medium based on multi-modal cardiovascular disease information |
| CN113345590A (en) * | 2021-06-29 | 2021-09-03 | 安徽大学 | User mental health monitoring method and system based on heterogeneous graph |
| CN115601819A (en) * | 2022-11-29 | 2023-01-13 | 四川大学华西医院（Cn） | Multimode violence tendency recognition method, device, equipment and medium |
| CN116451087A (en) * | 2022-12-20 | 2023-07-18 | 石家庄七彩联创光电科技有限公司 | Character matching method, device, terminal and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1969552A (en) * | 2004-06-17 | 2007-05-23 | 皇家飞利浦电子股份有限公司 | Personalized summaries using personality attributes |
| CN101524267A (en) * | 2008-03-04 | 2009-09-09 | 黄林 | Comprehensive evaluating system and proposal for checking personal physical and psychological health |
| CN103584872A (en) * | 2013-10-29 | 2014-02-19 | 燕山大学 | Psychological stress assessment method based on multi-physiological-parameter integration |
| US20150045688A1 (en) * | 2013-08-10 | 2015-02-12 | Dario Nardi | Nardi neurotype profiler system |
| CN109567830A (en) * | 2018-10-30 | 2019-04-05 | 清华大学 | A kind of measurement of personality method and system based on neural response |
| CN110507335A (en) * | 2019-08-23 | 2019-11-29 | 山东大学 | Inmate's psychological health states appraisal procedure and system based on multi-modal information |
-
2020
- 2020-08-26 CN CN202010873535.5A patent/CN112185493A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1969552A (en) * | 2004-06-17 | 2007-05-23 | 皇家飞利浦电子股份有限公司 | Personalized summaries using personality attributes |
| CN101524267A (en) * | 2008-03-04 | 2009-09-09 | 黄林 | Comprehensive evaluating system and proposal for checking personal physical and psychological health |
| US20150045688A1 (en) * | 2013-08-10 | 2015-02-12 | Dario Nardi | Nardi neurotype profiler system |
| CN103584872A (en) * | 2013-10-29 | 2014-02-19 | 燕山大学 | Psychological stress assessment method based on multi-physiological-parameter integration |
| CN109567830A (en) * | 2018-10-30 | 2019-04-05 | 清华大学 | A kind of measurement of personality method and system based on neural response |
| CN110507335A (en) * | 2019-08-23 | 2019-11-29 | 山东大学 | Inmate's psychological health states appraisal procedure and system based on multi-modal information |
Non-Patent Citations (5)
| Title |
|---|
| 俞国良，宋振韶著: "PYTHON金融大数据风控建模实战 基于机器学习", 西安电子科技大学出版社 * |
| 杨照芳: "心跳间期和皮肤电信号中的情感的响应模式研究", 《西南大学 博士学位论文》 * |
| 王向楠: "人格、价值观、教育与保险需求", 《经济学报》 * |
| 王小甜: "不同情绪状态下脑电信号特征的研究", 《长春理工大学 硕士学位论文》 * |
| 蔡华俭 等: "心理类型量表（MBTI）的修订初步", 《应用心理学》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113010783A (en) * | 2021-03-17 | 2021-06-22 | 华南理工大学 | Medical recommendation method, system and medium based on multi-modal cardiovascular disease information |
| CN113345590A (en) * | 2021-06-29 | 2021-09-03 | 安徽大学 | User mental health monitoring method and system based on heterogeneous graph |
| CN113345590B (en) * | 2021-06-29 | 2022-12-16 | 安徽大学 | User mental health monitoring method and system based on heterogeneous graph |
| CN115601819A (en) * | 2022-11-29 | 2023-01-13 | 四川大学华西医院（Cn） | Multimode violence tendency recognition method, device, equipment and medium |
| CN116451087A (en) * | 2022-12-20 | 2023-07-18 | 石家庄七彩联创光电科技有限公司 | Character matching method, device, terminal and storage medium |
| CN116451087B (en) * | 2022-12-20 | 2023-12-26 | 石家庄七彩联创光电科技有限公司 | Character matching method, device, terminal and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110507335B (en) | Multi-mode information based criminal psychological health state assessment method and system | |
| Ganapathy et al. | Convolutional neural network based emotion classification using electrodermal activity signals and time-frequency features | |
| Ding et al. | Measurement and identification of mental workload during simulated computer tasks with multimodal methods and machine learning | |
| CN106886792B (en) | Electroencephalogram emotion recognition method for constructing multi-classifier fusion model based on layering mechanism | |
| Subasi et al. | Classification of EMG signals using wavelet neural network | |
| CN112185493A (en) | Personality preference diagnosis device and project recommendation system based on same | |
| CN110070105B (en) | Electroencephalogram emotion recognition method and system based on meta-learning example rapid screening | |
| CN113729707A (en) | FECNN-LSTM-based emotion recognition method based on multi-mode fusion of eye movement and PPG | |
| CN112890816A (en) | Health index scoring method and device for individual user | |
| Nodera et al. | Deep learning for waveform identification of resting needle electromyography signals | |
| Long et al. | A scoping review on monitoring mental health using smart wearable devices | |
| Soni et al. | Graphical representation learning-based approach for automatic classification of electroencephalogram signals in depression | |
| Napoli et al. | Uncertainty in heart rate complexity metrics caused by R-peak perturbations | |
| Ross et al. | Unsupervised multi-modal representation learning for affective computing with multi-corpus wearable data | |
| Fan et al. | A new deep convolutional neural network incorporating attentional mechanisms for ECG emotion recognition | |
| Hossain et al. | Automatic motion artifact detection in electrodermal activity data using machine learning | |
| CN114693170A (en) | Multi-source parameter fusion-based psychological assessment method for parachuting training of pilots | |
| Gómez-García et al. | Emulating the perceptual capabilities of a human evaluator to map the GRB scale for the assessment of voice disorders | |
| Choi et al. | Physiological sensing and feature extraction for emotion recognition by exploiting acupuncture spots | |
| Slanzi et al. | Predicting Web user click intention using pupil dilation and electroencephalogram analysis | |
| Arora et al. | Unraveling depression using machine intelligence | |
| Ali et al. | Learning style classification via EEG sub-band spectral centroid frequency features | |
| Pan et al. | A universal emotion recognition method based on feature priority evaluation and classifier reinforcement | |
| Govarthan et al. | Deep Learning Framework for Categorical Emotional States Assessment Using Electrodermal Activity Signals | |
| Veeranki et al. | Non-Linear Signal Processing Methods for Automatic Emotion Recognition using Electrodermal Activity |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |