CN116312968A - Psychological consultation and healing system based on man-machine conversation and core algorithm - Google Patents
Psychological consultation and healing system based on man-machine conversation and core algorithm Download PDFInfo
- Publication number
- CN116312968A CN116312968A CN202310109277.7A CN202310109277A CN116312968A CN 116312968 A CN116312968 A CN 116312968A CN 202310109277 A CN202310109277 A CN 202310109277A CN 116312968 A CN116312968 A CN 116312968A
- Authority
- CN
- China
- Prior art keywords
- psychological
- call
- healing
- voice
- user
- 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
- 230000035876 healing Effects 0.000 title claims description 31
- 238000002560 therapeutic procedure Methods 0.000 claims abstract description 87
- 238000000554 physical therapy Methods 0.000 claims abstract description 27
- 230000003993 interaction Effects 0.000 claims abstract description 10
- 238000011084 recovery Methods 0.000 claims abstract description 9
- 239000013598 vector Substances 0.000 claims description 27
- 230000006978 adaptation Effects 0.000 claims description 21
- 238000000605 extraction Methods 0.000 claims description 20
- 206010020649 Hyperkeratosis Diseases 0.000 claims description 17
- 230000009193 crawling Effects 0.000 claims description 12
- 238000001671 psychotherapy Methods 0.000 claims description 9
- 230000035945 sensitivity Effects 0.000 claims description 8
- 238000006243 chemical reaction Methods 0.000 claims description 6
- 230000001934 delay Effects 0.000 claims description 6
- 230000000694 effects Effects 0.000 claims description 6
- 230000010365 information processing Effects 0.000 claims description 6
- 238000000034 method Methods 0.000 claims description 6
- 238000012216 screening Methods 0.000 claims description 6
- 238000012360 testing method Methods 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 6
- 238000012795 verification Methods 0.000 claims description 6
- 208000003643 Callosities Diseases 0.000 claims description 4
- 230000000747 cardiac effect Effects 0.000 claims description 3
- 238000004140 cleaning Methods 0.000 claims description 3
- 230000007774 longterm Effects 0.000 claims description 3
- 238000013507 mapping Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 238000013139 quantization Methods 0.000 claims description 3
- 238000002945 steepest descent method Methods 0.000 claims description 3
- 230000009286 beneficial effect Effects 0.000 abstract description 3
- 239000006185 dispersion Substances 0.000 abstract description 3
- 230000006870 function Effects 0.000 description 12
- 230000003068 static effect Effects 0.000 description 4
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 201000010099 disease Diseases 0.000 description 2
- 230000008451 emotion Effects 0.000 description 2
- 208000019901 Anxiety disease Diseases 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000036506 anxiety Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 208000035475 disorder Diseases 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000004630 mental health Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000414 obstructive effect Effects 0.000 description 1
- 230000009323 psychological health Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
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
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/23—Updating
- G06F16/2379—Updates performed during online database operations; commit processing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
- G06F16/284—Relational databases
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/26—Speech to text systems
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Social Psychology (AREA)
- Developmental Disabilities (AREA)
- Medical Informatics (AREA)
- Primary Health Care (AREA)
- Public Health (AREA)
- Epidemiology (AREA)
- Psychology (AREA)
- Psychiatry (AREA)
- Hospice & Palliative Care (AREA)
- General Health & Medical Sciences (AREA)
- Child & Adolescent Psychology (AREA)
- Computational Linguistics (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The psychological consultation and therapy system based on the man-machine dialogue and the core algorithm comprises a user side and a system terminal, wherein the user side comprises a permission management module, a voice acquisition module, a voice sending module and an interface interaction module, and the system terminal comprises a heart physiotherapy and call recovery collection module, a voice recognition module, a psychological therapy and call recovery matching module and a psychological therapy and call recovery updating module. The invention has the beneficial effects that the on-line and off-line man-machine conversation can be realized, the psychological state of the user can be judged according to the keywords extracted from the voice of the user, and then the psychological treatment call recovery operation with the highest correlation can be matched, so that psychological dispersion is brought to the user.
Description
Technical Field
The invention relates to the field of intelligent medical treatment, in particular to a psychological consultation and healing system based on a man-machine dialogue and a core algorithm.
Background
Aiming at psychological health problems, the related research results show that negative emotions and expression disorders basically exist in groups in society, the occurrence rate of psychological diseases is about 30%, and according to the analysis of data results, the occurrence rate of diseases has the trend of continuously rising, such as negative emotions of depression, anxiety, depression and the like, the normal learning life of people is influenced, so that the health of people is greatly discounted.
Therefore, the invention provides a psychological consultation and healing system based on man-machine conversation and a core algorithm, which hopes to give people timely and effective physical and mental health care.
Disclosure of Invention
In view of the above, the present invention aims to provide a psychological consultation and healing system based on man-machine conversations and core algorithms.
The aim of the invention is realized by the following technical scheme:
a psychological consultation and healing system based on man-machine conversation and core algorithm comprises a user side and a system terminal, wherein:
the user side comprises a right management module, a voice acquisition module, a voice sending module and an interface interaction module, wherein the right management module is used for acquiring the right of a user and comprises a right establishment unit and a right distribution unit, the right management is implemented by a system administrator, the system administrator establishes required right through the right establishment unit and comprises the steps of acquiring personal information of the user, starting the voice right and distributing the user right through the right distribution unit, after the voice right is started, the voice of the user is transmitted to the voice acquisition module, the voice acquisition module stores the voice of the user to a database, in order to ensure the privacy of the user, only the history record of the last use of the user is saved, if the user is not on line for more than 30 days, all the use records of the user are cleared, the voice of the user is sent to a system terminal through the voice sending module, and a psychological consultation and cure system is obtained through feedback of the system terminal, and the interface interaction module comprises an image display unit, a video display unit and a loudspeaker unit for replying images, videos and audios;
the system terminal comprises a heart physiotherapy call-healing collecting module, a voice recognition module, a psychological call-healing matching module and a psychological call-healing updating module, wherein the psychological call-healing collecting module comprises a crawling unit, a storage unit, a psychological call-healing feature extraction unit, a scoring unit and a screening unit, the crawling unit performs real-time crawling according to the existing psychological call-healing of a network and a database, a value MySQL database is stored by the storage unit, the light information processing of the heart physiotherapy call-healing feature extraction unit is carried out to lighten the storage pressure of the MySQL database, the extracted features need to be marked, scoring processing is carried out on each tag in the scoring unit according to the features in the tag, the psychological call-healing with lower score value is screened by the screening unit, so that the storage pressure of the MySQL database is further released, the voice recognition module is used for receiving the voice sent by the voice sending module at the user side, the voice recognition module comprises a voice-text conversion unit and a voice-text feature extraction unit, the voice-text conversion unit is used for converting the voice of the user into text, the voice-text feature extraction unit is used for extracting the feature information such as key words of the voice text of the user, and the like, the feature information is matched with the psychological therapy call with highest relativity in a MySQL database of the safe physical therapy call through a psychological therapy call matching module, the psychological therapy call matching module comprises a feature training unit, a testing unit and a verification unit, the training, the testing and the verification of the psychological therapy call are respectively completed, the psychological therapy call updating module periodically updates the MySQL database after marking the user with each service, the psychological therapy call with lower cleaning score is cleaned, and crawl the latest psychotherapy call as a replacement.
Further, the rights management module is used for acquiring the rights of the user, and comprises a rights establishment unit and a rights distribution unit, rights management is carried out by a human system administrator, the system administrator establishes required rights through the rights establishment unit, the personal information of the user comprises age, gender, occupation, marital and main appeal, the rights of the user are distributed through the rights distribution unit, the rights management is refined, the rights setting granularity is improved, the administrator key is updated regularly, and the administrator rights list is maintained regularly.
Further, the user voice collected by the voice collection module is transmitted to the voice recognition module of the system terminal through the voice transmission module by utilizing wifi, and features in the user voice are recognized.
Furthermore, the central physiotherapy and call-healing feature extraction unit of the psychological therapy and call-healing collection module is used for carrying out light information processing, so that the storage pressure of the MySQL database is reduced, and the extracted features need to be labeled, and the labels comprise whether the psychological therapy and call-healing are carried out, the adaptation degree of the psychological therapy and call-healing, the professional word adaptation degree of the psychological therapy and call-healing, and the psychological therapy and call-healing source.
Further, the feature extraction unit and scoring unit of the central physiotherapy and callout feature extraction module of the psychological therapy and callout collection module are used for extracting features of the label and scoring, if the crawling unit crawls I-line callout together, whether the I-line callout in the label is the feature value of the psychological therapy and callout is marked as alpha i Wherein I is more than or equal to 1 and less than or equal to I, I is an integer, and alpha i E {0,1}, if α i =0, then the speech is not a psychotheraphy call if α i =1, then the call is not a psycho-therapeutic call, and the fitness of the ith psycho-therapeutic call in the label is noted as β i, wherein βi ∈[0,1]When beta is i When the value of (2) is closer to 0, the adaptation degree of the ith psychology cure in the label is not obvious to the voice recovery effect of the user, when beta i When the value of (2) is closer to 1, the adaptation degree of the ith psychological therapy call-healing operation in the label is obvious in the voice reply effect of the user, and the matching degree of the ith psychological therapy call-healing operation professional word in the label is marked as gamma i, wherein γi ∈[0,1]When gamma is i When the value of (2) is closer to 0, the confidence of the matching degree of the ith psychology and guaranty operation professional word in the label is lower, and when gamma is i When the value of (2) is closer to 1, the confidence of the matching degree of the i-th psychology and callout professional word in the label is higher, and the quantized psychology and callout source in the label is marked as delta i And in order to ensure that the adaptation degree label of the psychology and the matching degree of the psychology professional word do not fall into a local solution interval, the source of the psychology after the expansion quantization converges to [0,100 ]]Within the interval, wherein delta i ∈[0,100]When delta i The closer the value of (2) is to 0, the description is in the labelThe matching degree of the i-th psychology therapy callout professional word is low in reliability, when delta i When the value of (2) is closer to 100, the matching degree reliability of the i-th psychology and callout professional word in the label is higher, and the objective function and constraint conditions can be obtained according to the matching degree reliability:
max(β),max(γ),max(δ)
such that:
assuming whether the adaptation degree is the psychological therapy call-healing degree, the matching degree of the psychological therapy call-healing professional word and the weight of the quantized psychological therapy call-healing source are respectively w 1 ，w 2 ，w 3 Has w 1 +w 2 +w 3 =1, the weighted normalized label is as follows: e (E) β ＝w 1 *β，E γ ＝w 2 *γ，E δ ＝w 3 *δ
wherein ,Eβ For weighted adaptation of psychotherapy calluses E γ For weighted psychology therapy call-healing professional word matching degree E δ For weighted quantized heart physiotherapy callus sources, an ideal solution A of an objective function is obtained by utilizing a multi-criterion decision and TOPSIS algorithm * And non-ideal solution A of objective function - The following are provided:
wherein ,
ideal solution for psychological therapy call healing degree>
Ideal solution for word matching degree of psychology therapy and callosity>
For quantifying ideal solution of heart physiotherapy and callus technique source, < ->
Non-ideal solution for psychological therapy call-healing fit->
For the non-ideal solution of the matching degree of the psychological therapy and the medical words, the ++>
To quantify the non-ideal solution of the cardiophysiotherapy callus source, the Euclidean distance of the ideal solution and the non-ideal solution is as follows:
then calculate the psychological therapy call-healing as
The optimal solution can be written as:
F＝max{F * }
such that:
w 1 ,w 2 ,w 3 ∈[0,1]
w 1 +w 2 +w 3 ＝1
further, the voice recognition module employs a hierarchical digital dynamic network to recognize voice input by the user, assuming that n is a net input at the mth layer in the hierarchical digital dynamic network m (t) is noted as:
wherein ,
is a set of layer subscripts connected directly forward to the mth layer, DL m,l Is the set of all delays on the tapped delay line between the first and the mth layers, LW m,l Is the layer weight between the l input vectors and the mth layer, IW m,l Is the input weight between the first input vector and the mth layer, I m Is the set under the input vector connected to the mth layer, DI m,l Is the set of all delays on the tapped delay line between the i-th input vector and the m-th layer, p l (t-d) is the first input vector at time t-d, b m Is the bias of the m-th layer, the output of the m-th layer is:
a m (t)＝f m (n m (t))
to train the network, the steepest descent method is used, using the sum of squares of the errors:
wherein, F is the sum of squares of the errors, e is the error of each iteration, Q is the total iteration number, τ is the target output, a is the actual output, and two components of the gradient are:
wherein ,lw1,1 Is the layer weight between 1 input vector and layer 1, iw 1,1 Is the input weight between the 1 st input vector and the 1 st layer, the derivative of the network output to the weight is:
the operation formula of the network is as follows: a (t) =lw 1,1 a(t-1)+iw 1,1 p (t), the derivative of the network output with respect to the weight can be written as:
further, the hierarchical digital dynamic network is improved by using BP back propagation rule, assuming that U is a set of all output layer numbers, U e U, set X is a set of all input layer numbers, X e X, and X and performance index F are a one-to-one mapping relationship, and the calculation gradient is as follows by the chain rule:
for each iteration, the sum term contains the term corresponding to each output layer if the performance index F (x) is not a particular output a u An explicit function of (t), the explicit derivative is 0, wherein,
the chain rule can be utilized to further popularize:
for calculating->
Assume that the net input for layer x is +.>
There is->
wherein ,/>
Inputting weight value a for the kth of the x layer u (t-d) is the output of the layer u at the number of iterations of t-d, defining the sensitivity of the layer u as:
explicit derivatives can be found:
wherein ,
the sensitivity of the kth weight of the nth layer to the ith weight of the mth layer is represented.
Further, the I-string speech operation that the psychological therapy and speech recovery matching module climbs through the heart physiotherapy and speech recovery collecting module is taken as a sample one and is marked as i= { I 1 ,i 2 ,…,i I The output layer generated by the voice recognition module outputs U user voice texts as a sample two, and the U 'is recorded as U' = { U 1 ,u 2 ,…,u U Then the overall average of sample one and sample two is noted as
The overall covariance is cov (I ', U'), then the overall pearson correlation coefficient can be written as:
defining a correlation coefficient threshold Tu, tu E [0,1 ]]If ρ I`,U` >Tu, then select the psychotherapy call for replying to the user.
Further, if max (ρ I`,U` )<Tu indicates that the timeliness of the crawled psychological therapy call-up operation can not meet the demands of users, and then the call-up operation with low long-term use rate of the MySQL database is required to be cleaned through the cardiac physiotherapy call-up operation updating module, and the crawl unit is restarted to obtain the latest psychological therapy call-up operation.
Further, the interface interaction module comprises an image display unit, a video display unit and a loudspeaker unit, so that image, video and audio replies are carried out, a mobile App is generated through an android studio, the voice authority of a user is obtained, and the man-machine conversation function of the user and the psychological consultation and healing system is completed.
The invention has the beneficial effects that: the method has the advantages that the psychological therapy call-healing operation can be crawled through the web crawler, the database is adaptively updated in real time according to the time period and the relativity information, the reliability of the score under the condition of multi-label multi-feature values is increased by combining a multi-criterion decision and a TOPSIS algorithm, an ideal solution and a non-ideal solution are constructed, the Euclidean distance of the ideal solution and the non-ideal solution is calculated, the reliable score can be obtained, the weight of each different feature is given to the different feature, the optimal value of the objective function is calculated under the constraint condition, the psychological therapy call-healing operation with lower utility value in the database can be effectively relieved, the cache is released, and the service quality of users can be increased. In order to extract the characteristics of speech recognition, a hierarchical digital dynamic network is adopted, compared with a static network, a plurality of layers can be connected to m layers in the hierarchical digital dynamic network, some of the connections can form a reply connection through a tap delay line, the hierarchical digital dynamic network can also be provided with a plurality of input vectors, the input vectors can be connected to any layer of the network, in contrast, the static multilayer network is provided with only one input vector and is connected to the first layer, the layers are sequentially connected according to numerical order, in the hierarchical digital dynamic network, any layer can be connected with any layer, in the hierarchical digital dynamic network, the sensitivity value and the explicit derivative are obtained through iteration and weighting by utilizing the chain law in BP back propagation, and a feedback expression is obtained, the original hierarchical digital dynamic network is expanded into the hierarchical digital dynamic back propagation network, under the network, the characteristics of speech can be effectively identified, and finally whether the psychological therapy is replied to a user or not is determined by utilizing the pearson correlation coefficient and the threshold value comparison. The psychological consultation and treatment system based on the man-machine conversation and the core algorithm can realize online and offline man-machine conversation, judge the psychological state of the user according to the keywords extracted from the voice of the user, and further match the psychological treatment and treatment operation with the highest correlation, so that psychological dispersion is brought to the user.
Drawings
The invention will be further described with reference to the accompanying drawings, in which embodiments do not constitute any limitation on the invention, and other drawings can be obtained by one of ordinary skill in the art without undue effort from the following drawings.
Fig. 1 is a schematic diagram of the structure of the present invention.
Detailed Description
The invention will be further described with reference to the following examples.
Referring to fig. 1, a psychological consultation and healing system based on man-machine dialogue and core algorithm of the present embodiment includes a user side and a system terminal, wherein:
the user side comprises a right management module, a voice acquisition module, a voice sending module and an interface interaction module, wherein the right management module is used for acquiring the right of a user and comprises a right establishment unit and a right distribution unit, the right management is implemented by a system administrator, the system administrator establishes required right through the right establishment unit and comprises the steps of acquiring personal information of the user, starting the voice right and distributing the user right through the right distribution unit, after the voice right is started, the voice of the user is transmitted to the voice acquisition module, the voice acquisition module stores the voice of the user to a database, in order to ensure the privacy of the user, only the history record of the last use of the user is saved, if the user is not on line for more than 30 days, all the use records of the user are cleared, the voice of the user is sent to a system terminal through the voice sending module, and a psychological consultation and cure system is obtained through feedback of the system terminal, and the interface interaction module comprises an image display unit, a video display unit and a loudspeaker unit for replying images, videos and audios;
the system terminal comprises a heart physiotherapy call-healing collecting module, a voice recognition module, a psychological call-healing matching module and a psychological call-healing updating module, wherein the psychological call-healing collecting module comprises a crawling unit, a storage unit, a psychological call-healing feature extraction unit, a scoring unit and a screening unit, the crawling unit performs real-time crawling according to the existing psychological call-healing of a network and a database, a value MySQL database is stored by the storage unit, the light information processing of the heart physiotherapy call-healing feature extraction unit is carried out to lighten the storage pressure of the MySQL database, the extracted features need to be marked, scoring processing is carried out on each tag in the scoring unit according to the features in the tag, the psychological call-healing with lower score value is screened by the screening unit, so that the storage pressure of the MySQL database is further released, the voice recognition module is used for receiving the voice sent by the voice sending module at the user side, the voice recognition module comprises a voice-text conversion unit and a voice-text feature extraction unit, the voice-text conversion unit is used for converting the voice of the user into text, the voice-text feature extraction unit is used for extracting the feature information such as key words of the voice text of the user, and the like, the feature information is matched with the psychological therapy call with highest relativity in a MySQL database of the safe physical therapy call through a psychological therapy call matching module, the psychological therapy call matching module comprises a feature training unit, a testing unit and a verification unit, the training, the testing and the verification of the psychological therapy call are respectively completed, the psychological therapy call updating module periodically updates the MySQL database after marking the user with each service, the psychological therapy call with lower cleaning score is cleaned, and crawl the latest psychotherapy call as a replacement.
Specifically, the rights management module is used for acquiring the rights of the user, and comprises a rights establishment unit and a rights distribution unit, rights management is implemented as a system administrator, the system administrator establishes required rights through the rights establishment unit, the required rights comprise the acquisition of personal information of the user, the starting of voice rights, the personal information of the user comprises age, gender, occupation, marital and main appeal, the rights of the user are distributed through the rights distribution unit, so that rights management is refined, the rights setting granularity is improved, the administrator key is updated regularly, and the administrator rights list is maintained regularly.
Specifically, the user voice collected by the voice collection module is transmitted to the voice recognition module of the system terminal by wifi through the voice transmission module, and features in the user voice are recognized.
Specifically, the central physiotherapy and call-healing feature extraction unit of the psychological therapy and call-healing collection module performs light information processing, reduces the storage pressure of the MySQL database, and marks the extracted features, wherein the marks comprise whether the psychological therapy and call-healing is performed, the adaptation degree of the psychological therapy and call-healing, the professional word adaptation degree of the psychological therapy and call-healing, and the psychological therapy and call-healing source.
Specifically, psychological therapy calluses can be divided into three categories: developed psychology problem related vocalization, adaptive psychology problem related vocalization and obstructive psychology problem related vocalization.
Specifically, the feature extraction unit and scoring unit of the central physiotherapy and callout feature extraction module of the psychological therapy and callout collection module are used for extracting features of the label and scoring, if the crawling unit crawls I-line callout together, whether the I-line callout in the label is the feature value of the psychological therapy and callout is marked as alpha i Wherein I is more than or equal to 1 and less than or equal to I, I is an integer, and alpha i E {0,1}, if α i =0, then the speech is not a psychotheraphy call if α i =1, then the call is not a psycho-therapeutic call, and the fitness of the ith psycho-therapeutic call in the label is noted as β i, wherein βi ∈[0,1]When beta is i When the value of (2) is closer to 0, the adaptation degree of the ith psychology cure in the label is not obvious to the voice recovery effect of the user, when beta i When the value of (2) is closer to 1, the adaptation degree of the ith psychological therapy call-healing operation in the label is obvious in the voice reply effect of the user, and the matching degree of the ith psychological therapy call-healing operation professional word in the label is marked as gamma i, wherein γi ∈[0,1]When gamma is i When the value of (2) is closer to 0, the confidence of the matching degree of the ith psychology and guaranty operation professional word in the label is lower, and when gamma is i When the value of (2) is closer to 1, the confidence of the matching degree of the i-th psychology and callout professional word in the label is higher, and the quantized psychology and callout source in the label is marked as delta i And in order to ensure that the adaptation degree label of the psychology and the matching degree of the psychology professional word do not fall into a local solution interval, the source of the psychology after the expansion quantization converges to [0,100 ]]Within the interval, wherein delta i ∈[0,100]When delta i When the value of (2) is closer to 0, the matching degree of the ith psychology and callosity professional word in the label is lower in reliability, and when delta i When the value of (2) is closer to 100, the matching degree reliability of the i-th psychology and callout professional word in the label is higher, and the objective function and constraint conditions can be obtained according to the matching degree reliability:
max(β),max(γ),max(δ)
such that:
assuming whether the adaptation degree is the psychological therapy call-healing degree, the matching degree of the psychological therapy call-healing professional word and the weight of the quantized psychological therapy call-healing source are respectively w 1 ，w 2 ，w 3 Has w 1 +w 2 +w 3 =1, the weighted normalized label is as follows: e (E) β ＝w 1 *β，E γ ＝w 2 *γ，E δ ＝w 3 *δ
wherein ,Eβ For weighted adaptation of psychotherapy calluses E γ For weighted psychology therapy call-healing professional word matching degree E δ For weighted quantized heart physiotherapy callus sources, an ideal solution A of an objective function is obtained by utilizing a multi-criterion decision and TOPSIS algorithm * And non-ideal solution A of objective function - The following are provided:
wherein ,
ideal solution for psychological therapy call healing degree>
Ideal solution for word matching degree of psychology therapy and callosity>
For quantifying ideal solution of heart physiotherapy and callus technique source, < ->
Non-ideal solution for psychological therapy call-healing fit->
For the non-ideal solution of the matching degree of the psychological therapy and the medical words, the ++>
To quantify the non-ideal solution of the cardiophysiotherapy callus source, the Euclidean distance of the ideal solution and the non-ideal solution is as follows:
then calculate the psychological therapy call-healing as
The optimal solution can be written as:
F＝max{F * }
such that:
w 1 ,w 2 ,w 3 ∈[0,1]
w 1 +w 2 +w 3 ＝1
specifically, the voice recognition module adopts a layered digital dynamic network to recognize the voice input by the user, and presumes that the m-th layer net input n is in the layered digital dynamic network m (t) is noted as:
wherein ,
is a set of layer subscripts connected directly forward to the mth layer, DL m,l Is the set of all delays on the tapped delay line between the first and the mth layers, LW m,l Is the layer weight between the l input vectors and the mth layer, IW m,l Is the input weight between the first input vector and the mth layer, I m Is connected withSet under input vector to mth layer, DI m,l Is the set of all delays on the tapped delay line between the i-th input vector and the m-th layer, p l (t-d) is the first input vector at time t-d, b m Is the bias of the m-th layer, the output of the m-th layer is:
a m (t)＝f m (n m (t))
to train the network, the steepest descent method is used, using the sum of squares of the errors:
wherein, F is the sum of squares of the errors, e is the error of each iteration, Q is the total iteration number, τ is the target output, a is the actual output, and two components of the gradient are:
wherein ,lw1,1 Is the layer weight between 1 input vector and layer 1, iw 1,1 Is the input weight between the 1 st input vector and the 1 st layer, the derivative of the network output to the weight is:
the operation formula of the network is as follows: a (t) =lw 1,1 a(t-1)+iw 1,1 p (t), the derivative of the network output with respect to the weight can be written as:
specifically, the hierarchical digital dynamic network is improved by using a BP back propagation rule, and assuming U is a set of all output layer numbers, U epsilon U, set X is a set of all input layer numbers, X epsilon X, and X and performance index F are in one-to-one mapping relation, the gradient is calculated as follows by a chain rule:
for each iteration, the sum term contains the term corresponding to each output layer if the performance index F (x) is not a particular output a u An explicit function of (t), the explicit derivative is 0, wherein,
the chain rule can be utilized to further popularize:
for calculating->
Assume that the net input for layer x is +.>
There is->
wherein ,/>
Inputting weight value a for the kth of the x layer u (t-d) is the output of the layer u at the number of iterations of t-d, defining the sensitivity of the layer u as:
explicit derivatives can be found:
wherein 
The sensitivity of the kth weight of the nth layer to the ith weight of the mth layer is represented.
Specifically, the psychological therapy call-healing matching module heals the call through the psychological therapyI-string microphone crawled by the microphone collection module is taken as a sample I and is marked as I' = { I 1 ,i 2 ,…,i I The output layer generated by the voice recognition module outputs U user voice texts as a sample two, and the U 'is recorded as U' = { U 1 ,u 2 ,…,u U Then the overall average of sample one and sample two is noted as
The overall covariance is:
the overall pearson correlation coefficient can be written as:
defining a correlation coefficient threshold Tu, tu E [0,1 ]]If ρ I`,U` >Tu, then select the psychotherapy call for replying to the user.
Preferably, if max (ρ I`,U` )<Tu indicates that the timeliness of the crawled psychological therapy call-up operation can not meet the demands of users, and then the call-up operation with low long-term use rate of the MySQL database is required to be cleaned through the cardiac physiotherapy call-up operation updating module, and the crawl unit is restarted to obtain the latest psychological therapy call-up operation.
Preferably, in order to ensure the effectiveness of the psychoacoustic healing procedure returned to the user, tu is set to 0.8, i.e. when ρ I`,U` ∈[Tu,1]The more effective the psychological therapy of this man-machine conversation.
Specifically, the interface interaction module comprises an image display unit, a video display unit and a loudspeaker unit, so as to reply images, videos and audios, generate a mobile App through an android studio, acquire user voice authority and complete the man-machine conversation function of a user and a psychological consultation and healing system.
The invention has the beneficial effects that: the method has the advantages that the psychological therapy call-healing operation can be crawled through the web crawler, the database is adaptively updated in real time according to the time period and the relativity information, the reliability of the score under the condition of multi-label multi-feature values is increased by combining a multi-criterion decision and a TOPSIS algorithm, an ideal solution and a non-ideal solution are constructed, the Euclidean distance of the ideal solution and the non-ideal solution is calculated, the reliable score can be obtained, the weight of each different feature is given to the different feature, the optimal value of the objective function is calculated under the constraint condition, the psychological therapy call-healing operation with lower utility value in the database can be effectively relieved, the cache is released, and the service quality of users can be increased. In order to extract the characteristics of speech recognition, a hierarchical digital dynamic network is adopted, compared with a static network, a plurality of layers can be connected to m layers in the hierarchical digital dynamic network, some of the connections can form a reply connection through a tap delay line, the hierarchical digital dynamic network can also be provided with a plurality of input vectors, the input vectors can be connected to any layer of the network, in contrast, the static multilayer network is provided with only one input vector and is connected to the first layer, the layers are sequentially connected according to numerical order, in the hierarchical digital dynamic network, any layer can be connected with any layer, in the hierarchical digital dynamic network, the sensitivity value and the explicit derivative are obtained through iteration and weighting by utilizing the chain law in BP back propagation, and a feedback expression is obtained, the original hierarchical digital dynamic network is expanded into the hierarchical digital dynamic back propagation network, under the network, the characteristics of speech can be effectively identified, and finally whether the psychological therapy is replied to a user or not is determined by utilizing the pearson correlation coefficient and the threshold value comparison. The psychological consultation and treatment system based on the man-machine conversation and the core algorithm can realize online and offline man-machine conversation, judge the psychological state of the user according to the keywords extracted from the voice of the user, and further match the psychological treatment and treatment operation with the highest correlation, so that psychological dispersion is brought to the user.
Finally, it should be noted that the above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the scope of the present invention, and although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications or equivalent substitutions can be made to the technical solution of the present invention without departing from the spirit and scope of the technical solution of the present invention.
Claims (10)
1. A psychological consultation and healing system based on man-machine conversation and core algorithm comprises a user side and a system terminal, wherein:
the user side comprises a right management module, a voice acquisition module, a voice sending module and an interface interaction module, wherein the right management module is used for acquiring the right of a user and comprises a right establishment unit and a right distribution unit, the right management is implemented by a system administrator, the system administrator establishes required right through the right establishment unit and comprises the steps of acquiring personal information of the user, starting the voice right and distributing the user right through the right distribution unit, after the voice right is started, the voice of the user is transmitted to the voice acquisition module, the voice acquisition module stores the voice of the user to a database, in order to ensure the privacy of the user, only the history record of the last use of the user is saved, if the user is not on line for more than 30 days, all the use records of the user are cleared, the voice of the user is sent to a system terminal through the voice sending module, and a psychological consultation and cure system is obtained through feedback of the system terminal, and the interface interaction module comprises an image display unit, a video display unit and a loudspeaker unit for replying images, videos and audios;
the system terminal comprises a heart physiotherapy call-healing collecting module, a voice recognition module, a psychological call-healing matching module and a psychological call-healing updating module, wherein the psychological call-healing collecting module comprises a crawling unit, a storage unit, a psychological call-healing feature extraction unit, a scoring unit and a screening unit, the crawling unit performs real-time crawling according to the existing psychological call-healing of a network and a database, a value MySQL database is stored by the storage unit, the light information processing of the heart physiotherapy call-healing feature extraction unit is carried out to lighten the storage pressure of the MySQL database, the extracted features need to be marked, scoring processing is carried out on each tag in the scoring unit according to the features in the tag, the psychological call-healing with lower score value is screened by the screening unit, so that the storage pressure of the MySQL database is further released, the voice recognition module is used for receiving the voice sent by the voice sending module at the user side, the voice recognition module comprises a voice-text conversion unit and a voice-text feature extraction unit, the voice-text conversion unit is used for converting the voice of the user into text, the voice-text feature extraction unit is used for extracting the feature information such as key words of the voice text of the user, and the like, the feature information is matched with the psychological therapy call with highest relativity in a MySQL database of the safe physical therapy call through a psychological therapy call matching module, the psychological therapy call matching module comprises a feature training unit, a testing unit and a verification unit, the training, the testing and the verification of the psychological therapy call are respectively completed, the psychological therapy call updating module periodically updates the MySQL database after marking the user with each service, the psychological therapy call with lower cleaning score is cleaned, and crawl the latest psychotherapy call as a replacement.
2. The psychological consultation and healing system based on man-machine conversation and core algorithm according to claim 1, wherein the authority management module is used for obtaining the authority of the user, and comprises an authority establishment unit and an authority distribution unit, the authority management is implemented by a human system manager, the system manager establishes required authority through the authority establishment unit, the required authority comprises obtaining personal information of the user, starting voice authority, the personal information of the user comprises age, gender, occupation, marital and main appeal, the authority distribution unit distributes the authority of the user to refine the authority management, improve the authority setting granularity, update manager keys periodically and maintain manager authority lists periodically.
3. The psychological consultation and healing system based on man-machine conversation and core algorithm according to claim 1, wherein the user voice collected by the voice collection module is transmitted to the voice recognition module of the system terminal by wifi through the voice sending module, and features in the user voice are recognized.
4. The psychological consultation and healing system based on man-machine conversation and core algorithm according to claim 1, wherein the psychological healing collecting module central physical therapy healing feature extraction unit is light in weight information processing, the MySQL database storage pressure is relieved, the extracted features need to be labeled, and the labels comprise whether the psychological healing is performed, the adaptation degree of the psychological healing, the professional word adaptation degree of the psychological healing and the psychological healing source.
5. The psychological consultation and healing system based on man-machine conversation and core algorithm according to claim 4, wherein the feature extraction unit and scoring unit of the central physiotherapy healy feature collection module are used for extracting features of the label and scoring, if the crawling unit crawls the I-th speech, the feature value of whether the I-th speech is the psychological healy in the label is marked as alpha i Wherein I is more than or equal to 1 and less than or equal to I, I is an integer, and alpha i E {0,1}, if α i =0, then the speech is not a psychotheraphy call if α i =1, then the call is not a psycho-therapeutic call, and the fitness of the ith psycho-therapeutic call in the label is noted as β i, wherein βi ∈[0,1]When beta is i When the value of (2) is closer to 0, the adaptation degree of the ith psychology cure in the label is not obvious to the voice recovery effect of the user, when beta i When the value of (2) is closer to 1, the adaptation degree of the ith psychological therapy call-healing operation in the label is obvious in the voice reply effect of the user, and the matching degree of the ith psychological therapy call-healing operation professional word in the label is marked as gamma i, wherein γi ∈[0,1]When gamma is i When the value of (2) is closer to 0, the confidence of the matching degree of the ith psychology and guaranty operation professional word in the label is lower, and when gamma is i When the value of (2) is closer to 1, the confidence of the matching degree of the i-th psychology and callout professional word in the label is higher, and the quantized psychology and callout source in the label is marked as delta i And in order to ensure that the adaptation degree label of the psychology and the matching degree of the psychology professional word do not fall into a local solution interval, the source of the psychology after the expansion quantization converges to [0,100 ]]Within the interval, wherein delta i ∈[0,100]When the value of delta i is more close to 0, the matching degree reliability of the ith psychological therapy and callout professional word in the label is lower, and when delta i When the value of (2) is closer to 100, the matching degree reliability of the i-th psychology and callout professional word in the label is higher, and the objective function and constraint conditions can be obtained according to the matching degree reliability:
max(β)，max(γ)，max(δ)
such that:
assuming whether the adaptation degree is the psychological therapy call-healing degree, the matching degree of the psychological therapy call-healing professional word and the weight of the quantized psychological therapy call-healing source are respectively w 1 ，w 2 ，w 3 Has w 1 +w 2 +w 3 =1, the weighted normalized label is as follows: e (E) β ＝w 1 *β，E γ ＝w 2 *γ，E δ ＝w 3 *δ
wherein ,Eβ For weighted adaptation of psychotherapy calluses E γ For weighted psychology therapy call-healing professional word matching degree E δ For weighted quantized heart physiotherapy callus sources, an ideal solution A of an objective function is obtained by utilizing a multi-criterion decision and TOPSIS algorithm * And non-ideal solution A of objective function - The following are provided:
wherein ,
ideal solution for psychological therapy call healing degree>
Ideal solution for word matching degree of psychology therapy and callosity>
For quantifying ideal solution of heart physiotherapy and callus technique source, < ->
For a non-ideal solution of the psychological therapy call healing fit,
for the non-ideal solution of the matching degree of the psychological therapy and the medical words, the ++>
To quantify the non-ideal solution of the cardiophysiotherapy callus source, the Euclidean distance of the ideal solution and the non-ideal solution is as follows:
then calculate the psychological therapy call-healing as
The optimal solution can be written as:
F＝max{F * }
such that:
w 1 ，w 2 ，w 3 ∈[0，1]
w 1 +w 2 +w 3 ＝1
6. the psychological consulting and healing system based on man-machine conversation and core algorithm of claim 1, wherein the speech recognition module employs a hierarchical digital dynamic network for userInput speech recognition assuming m-th net input n in hierarchical digital dynamic networks m (t) is noted as:
wherein ,
is a set of layer subscripts connected directly forward to the mth layer, DL m，l Is the set of all delays on the tapped delay line between the first and the mth layers, LW m，l Is the layer weight between the l input vectors and the mth layer, IW m，l Is the input weight between the first input vector and the mth layer, I m Is the set under the input vector connected to the mth layer, DI m，l Is the set of all delays on the tapped delay line between the i-th input vector and the m-th layer, p l (t-d) is the first input vector at time t-d, b m Is the bias of the m-th layer, the output of the m-th layer is:
a m (t)＝f m (n m (t))
to train the network, the steepest descent method is used, using the sum of squares of the errors:
wherein, F is the sum of squares of the errors, e is the error of each iteration, Q is the total iteration number, τ is the target output, a is the actual output, and two components of the gradient are:
wherein ,lw1，1 is the layer weight between 1 input vector and layer 1, iw 1，1 Is the input weight between the 1 st input vector and the 1 st layer, the derivative of the network output to the weight is:
the operation formula of the network is as follows: a (t) =lw 1，1 a(t-1)+iw 1，1 p (t), the derivative of the network output with respect to the weight can be written as:
7. the psychological consulting and healing system based on human-computer conversations and core algorithms of claim 6, wherein the hierarchical digital dynamic network is improved by BP back propagation rules assuming U is a set of all output layer numbers, U e U, set X is a set of all input layer numbers, X e X, and X is a one-to-one mapping relationship with performance index F, the gradient is calculated by the chain law as follows:
for each iteration, the sum term contains the term corresponding to each output layer if the performance index F (x) is not a particular output a u An explicit function of (t), the explicit derivative is 0, wherein,
the chain rule can be utilized to further popularize:
for calculating->
Assume that the net input for layer x is +.>
There is->
wherein ,/>
Inputting weight value a for the kth of the x layer u (t-d) is the output of the layer u at the number of iterations of t-d, defining the sensitivity of the layer u as:
can calculate explicit derivative
wherein 
The sensitivity of the kth weight of the nth layer to the ith weight of the mth layer is represented.
8. The psychological consulting and healing system based on man-machine conversation and core algorithm of claim 1, wherein, the psychological therapy call-healing matching module takes I call-healing which is crawled by the heart physiotherapy call-healing collecting module as a sample I, and marks I '= { I' 1 ，i 2 ，...，i I The output layer generated by the voice recognition module outputs U user voice texts as a sample two, and the U 'is recorded as U' = { U 1 ，u 2 ，...，u U Then the overall average of sample one and sample two is noted as
The overall covariance is cov (I ', U'), then the overall pearson correlation coefficient can be written as:
defining a correlation coefficient threshold Tu，Tu∈[0，1]If ρ I`，U` > Tu, the psychotherapy call is selected for use in replying to the user.
9. The psychological consulting and healing system based on human-machine dialogue and core algorithms according to claim 8, characterized in that if max (ρ I｀，U` ) And (3) the time efficiency of the crawled psychological therapy call-healing operation is described to be less than the requirement of a user, and the call-healing operation with low long-term use rate of the MySQL database is required to be cleaned through a cardiac physiotherapy call-healing operation updating module at the moment, and the crawl unit is restarted to obtain the latest psychological therapy call-healing operation.
10. The psychological consultation and healing system based on the man-machine conversation and the core algorithm according to claim 1, wherein the interface interaction module comprises an image display unit, a video display unit and a loudspeaker unit, so that image, video and audio replies are carried out, a mobile App is generated through Androidstudio, and user voice authority is obtained, and the man-machine conversation function of a user and the psychological consultation and healing system is completed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310109277.7A CN116312968A (en) | 2023-02-09 | 2023-02-09 | Psychological consultation and healing system based on man-machine conversation and core algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310109277.7A CN116312968A (en) | 2023-02-09 | 2023-02-09 | Psychological consultation and healing system based on man-machine conversation and core algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116312968A true CN116312968A (en) | 2023-06-23 |
Family
ID=86835172
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310109277.7A Pending CN116312968A (en) | 2023-02-09 | 2023-02-09 | Psychological consultation and healing system based on man-machine conversation and core algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116312968A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111564202A (en) * | 2020-04-30 | 2020-08-21 | 深圳市镜象科技有限公司 | Psychological counseling method based on man-machine conversation, psychological counseling terminal and storage medium |
| US20200335096A1 (en) * | 2018-04-19 | 2020-10-22 | Boe Technology Group Co., Ltd. | Pinyin-based method and apparatus for semantic recognition, and system for human-machine dialog |
| CN112182197A (en) * | 2020-11-09 | 2021-01-05 | 北京明略软件系统有限公司 | Method, device and equipment for recommending dialect and computer readable medium |
| WO2021169263A1 (en) * | 2020-02-26 | 2021-09-02 | 平安科技（深圳）有限公司 | Semantic matching method and device based on internal adversarial mechanism, and storage medium |
| WO2022142006A1 (en) * | 2020-12-30 | 2022-07-07 | 平安科技（深圳）有限公司 | Semantic recognition-based verbal skill recommendation method and apparatus, device, and storage medium |
| CN115376656A (en) * | 2022-08-10 | 2022-11-22 | 重庆医科大学 | Intelligent interaction system for psychological consultation of children and teenagers |
-
2023
- 2023-02-09 CN CN202310109277.7A patent/CN116312968A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20200335096A1 (en) * | 2018-04-19 | 2020-10-22 | Boe Technology Group Co., Ltd. | Pinyin-based method and apparatus for semantic recognition, and system for human-machine dialog |
| WO2021169263A1 (en) * | 2020-02-26 | 2021-09-02 | 平安科技（深圳）有限公司 | Semantic matching method and device based on internal adversarial mechanism, and storage medium |
| CN111564202A (en) * | 2020-04-30 | 2020-08-21 | 深圳市镜象科技有限公司 | Psychological counseling method based on man-machine conversation, psychological counseling terminal and storage medium |
| CN112182197A (en) * | 2020-11-09 | 2021-01-05 | 北京明略软件系统有限公司 | Method, device and equipment for recommending dialect and computer readable medium |
| WO2022142006A1 (en) * | 2020-12-30 | 2022-07-07 | 平安科技（深圳）有限公司 | Semantic recognition-based verbal skill recommendation method and apparatus, device, and storage medium |
| CN115376656A (en) * | 2022-08-10 | 2022-11-22 | 重庆医科大学 | Intelligent interaction system for psychological consultation of children and teenagers |
Non-Patent Citations (1)
| Title |
|---|
| 施海虎;闫峰;李冬梅;胡英飞;: "基于人机会话技术的网上心理咨询专家系统", 计算机工程与设计, no. 12, pages 149 - 151 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Salekin et al. | A weakly supervised learning framework for detecting social anxiety and depression | |
| CN110222163B (en) | Intelligent question-answering method and system integrating CNN and bidirectional LSTM | |
| CN103810999B (en) | Language model training method based on Distributed Artificial Neural Network and system thereof | |
| Klaylat et al. | Emotion recognition in Arabic speech | |
| JP2022553749A (en) | Acoustic and Natural Language Processing Models for Velocity-Based Screening and Behavioral Health Monitoring | |
| CN111125525B (en) | Personalized transformation correction strategy recommendation system for prisoner and operation method thereof | |
| CN107526720A (en) | Meaning generation method, meaning generating means and program | |
| CN107230401A (en) | Utilize internet and the Teaching of Writing interactive system and implementation method of voice technology | |
| CN116542241B (en) | Matching method of emergency plan and emergency medical rescue cooperative command platform system | |
| CN115393933A (en) | Video face emotion recognition method based on frame attention mechanism | |
| CN102339606A (en) | Telephone automatic voice recognition screening system for depression | |
| CN113241178B (en) | Device for determining severity of depression of tested person | |
| Wang | [Retracted] Research on Open Oral English Scoring System Based on Neural Network | |
| CN118262874A (en) | Knowledge-graph-based traditional Chinese medicine diagnosis and treatment model data expansion system and method | |
| Gu et al. | Multimodal attention network for trauma activity recognition from spoken language and environmental sound | |
| CN116312968A (en) | Psychological consultation and healing system based on man-machine conversation and core algorithm | |
| CN115171870A (en) | Diagnosis guiding and prompting method and system based on m-BERT pre-training model | |
| JP2013117683A (en) | Voice recognizer, error tendency learning method and program | |
| Serban et al. | The octopus approach to the Alexa competition: A deep ensemble-based socialbot | |
| Banerjee et al. | Intelligent stuttering speech recognition: A succinct review | |
| Mnih et al. | Improving a statistical language model through non-linear prediction | |
| Abdulbaqi et al. | Speech-based activity recognition for trauma resuscitation | |
| Feng et al. | Estimating public speaking anxiety from speech signals using unsupervised transfer learning | |
| CN114333790A (en) | Data processing method, device, equipment, storage medium and program product | |
| Campbell et al. | Speech and Text Processing for Major Depressive Disorder Detection |
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 |