CN112735564A - Mental health state prediction method, mental health state prediction apparatus, mental health state prediction medium, and computer program product - Google Patents
Mental health state prediction method, mental health state prediction apparatus, mental health state prediction medium, and computer program product Download PDFInfo
- Publication number
- CN112735564A CN112735564A CN202110128781.2A CN202110128781A CN112735564A CN 112735564 A CN112735564 A CN 112735564A CN 202110128781 A CN202110128781 A CN 202110128781A CN 112735564 A CN112735564 A CN 112735564A
- Authority
- CN
- China
- Prior art keywords
- text
- mental health
- health state
- emotion
- type
- 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
- 230000004630 mental health Effects 0.000 title claims abstract description 162
- 238000000034 method Methods 0.000 title claims abstract description 54
- 238000004590 computer program Methods 0.000 title claims abstract description 14
- 230000008451 emotion Effects 0.000 claims abstract description 255
- 230000003993 interaction Effects 0.000 claims abstract description 91
- 238000003058 natural language processing Methods 0.000 claims abstract description 45
- 238000013145 classification model Methods 0.000 claims abstract description 37
- 239000013598 vector Substances 0.000 claims description 174
- 238000004458 analytical method Methods 0.000 claims description 66
- 239000013604 expression vector Substances 0.000 claims description 33
- 230000002452 interceptive effect Effects 0.000 claims description 19
- 230000007935 neutral effect Effects 0.000 claims description 12
- 239000011159 matrix material Substances 0.000 description 32
- 238000000605 extraction Methods 0.000 description 26
- 238000005516 engineering process Methods 0.000 description 7
- 238000010801 machine learning Methods 0.000 description 7
- 238000004891 communication Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 6
- 201000010099 disease Diseases 0.000 description 4
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 4
- 238000013507 mapping Methods 0.000 description 4
- 230000009466 transformation Effects 0.000 description 4
- 238000013473 artificial intelligence Methods 0.000 description 3
- 230000001419 dependent effect Effects 0.000 description 3
- PXFBZOLANLWPMH-UHFFFAOYSA-N 16-Epiaffinine Natural products C1C(C2=CC=CC=C2N2)=C2C(=O)CC2C(=CC)CN(C)C1C2CO PXFBZOLANLWPMH-UHFFFAOYSA-N 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 230000009323 psychological health Effects 0.000 description 2
- 230000004931 aggregating effect Effects 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 208000020016 psychiatric disease Diseases 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
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
- G06F16/355—Class or cluster creation or modification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- 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
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Medical Informatics (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Primary Health Care (AREA)
- Evolutionary Computation (AREA)
- Databases & Information Systems (AREA)
- Epidemiology (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Mathematical Physics (AREA)
- Developmental Disabilities (AREA)
- Biomedical Technology (AREA)
- Pathology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Child & Adolescent Psychology (AREA)
- Computing Systems (AREA)
- Hospice & Palliative Care (AREA)
- Psychiatry (AREA)
- Psychology (AREA)
- Social Psychology (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- Machine Translation (AREA)
Abstract
The application discloses a mental health state prediction method, equipment, a medium and a computer program product, wherein the mental health state prediction method comprises the following steps: the method comprises the steps of taking a mental health state information text of a target user, and carrying out natural language processing on the mental health state information text to obtain a natural language processing result; based on a preset emotion classification model, carrying out emotion classification on each text word in the natural language processing result and the interaction relation among the text words to obtain an emotion classification result; and predicting the mental health state of the target user based on the emotion classification result to obtain a target prediction result. The method and the device solve the technical problem of low accuracy of mental health state prediction.
Description
Technical Field
The present application relates to the field of artificial intelligence in financial technology (Fintech), and in particular, to a mental health state prediction method, apparatus, medium, and computer program product.
Background
With the continuous development of financial science and technology, especially internet science and technology, more and more technologies (such as distributed technology, artificial intelligence and the like) are applied to the financial field, but the financial industry also puts higher requirements on the technologies, for example, higher requirements on the distribution of backlog in the financial industry are also put forward.
With the continuous development of computer technology, artificial intelligence is also more and more widely applied, currently, many studies show that a dream generally reflects daily activities of people, a psychologist generally predicts the mental health state of a psychological disease patient according to dream analysis information of the psychological disease patient, for example, the mental health state of a dream owner is poorer when the score is lower when the mental health state of the psychological disease patient is analyzed in a scoring manner, but the mental health cases are fewer and mastered by the psychologist, the dreams are often strange and many in variety, and the dream analysis information acquired by the psychologist is too thin and simple, so that the accuracy of mental health state prediction is lower.
Disclosure of Invention
The present application mainly aims to provide a mental health state prediction method, a device, a medium and a computer program product, and aims to solve the technical problem of low mental health state prediction accuracy in the prior art.
To achieve the above object, the present application provides a mental health state prediction method applied to a mental health state prediction apparatus, the mental health state prediction method including:
acquiring a mental health state information text of a target user, and performing natural language processing on the mental health state information text to obtain a natural language processing result;
based on a preset emotion classification model, carrying out emotion classification on each text word in the natural language processing result and the interaction relation among the text words to obtain an emotion classification result;
and predicting the mental health state of the target user based on the emotion classification result to obtain a target prediction result.
The present application also provides a mental health state prediction apparatus, the mental health state prediction apparatus is a virtual apparatus, and the mental health state prediction apparatus is applied to a mental health state prediction device, the mental health state prediction apparatus includes:
the natural language processing module is used for acquiring a mental health state information text of a target user, and performing natural language processing on the mental health state information text to acquire a natural language processing result;
the emotion classification module is used for carrying out emotion classification on each text word in the natural language processing result and the interaction relation among the text words based on a preset emotion classification model to obtain an emotion classification result;
and the prediction module is used for predicting the mental health state of the target user based on the emotion classification result to obtain a target prediction result.
The present application also provides a mental health state prediction apparatus, which is an entity apparatus, the mental health state prediction apparatus including: a memory, a processor and a program of the mental health state prediction method stored on the memory and executable on the processor, the program of the mental health state prediction method when executed by the processor being capable of implementing the steps of the mental health state prediction method as described above.
The present application also provides a medium which is a readable storage medium having stored thereon a program for implementing the mental health state prediction method, the program implementing the mental health state prediction method as described above when executed by a processor.
The present application also provides a computer program product comprising a computer program which, when executed by a processor, performs the steps of the mental health state prediction method as described above.
Compared with the dream analysis result obtained by performing dream analysis in a grading mode adopted by the prior art and the technical means for predicting the mental health state, the mental health state information text is obtained after the mental health state information text of a target user is obtained, wherein the mental health state information text comprises a dream description text, natural language processing is performed on the mental health state information text to split the text into sentences, phrases, words and the like, the interactive relationship between the words and the words is determined to obtain a natural language processing result, and then based on a preset emotion classification model, emotion classification is automatically performed on each text word in the natural language processing result and the interactive relationship between each text word, so that emotion information corresponding to each text word in the mental health state information text is obtained, and emotion information corresponding to the interaction relationship among the text words, and further obtaining emotion classification results, wherein the purpose of performing emotion analysis of the text word dimension on the dream information is achieved, compared with a mode of grading the whole dream information, the dream analysis process performed by a preset emotion classification model which is a machine learning model and is constructed based on massive information is more detailed and comprehensive, the obtained dream analysis information (namely, emotion classification results) is more reliable, accurate and rich, and furthermore, based on the emotion classification results, the mental health state prediction is performed on the target user, so that the purpose of performing mental health state prediction based on more reliable, accurate and rich dream analysis information can be achieved, the target prediction results are obtained, and the problems that the mental health state is generally strange due to fewer cases of the dream mastered by psychologists are overcome, the variety is various, and then the dream analysis information that psychologist obtained is too thin and brief, and then leads to the lower technical defect of accuracy of mental health state prediction, has promoted mental health state prediction's accuracy.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and together with the description, serve to explain the principles of the application.
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly described below, and it is obvious for those skilled in the art to obtain other drawings without inventive exercise.
FIG. 1 is a flowchart illustrating a mental health status prediction method according to a first embodiment of the present application;
FIG. 2 is a flowchart illustrating a mental health status prediction method according to a second embodiment of the present application;
fig. 3 is a schematic device configuration diagram of a hardware operating environment related to a mental health state prediction method in an embodiment of the present application.
The objectives, features, and advantages of the present application will be further described with reference to the accompanying drawings.
Detailed Description
It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
In a first embodiment of the present invention, referring to fig. 1, a method for predicting a mental health state includes:
step S10, acquiring a mental health state information text of a target user, and performing natural language processing on the mental health state information text to obtain a natural language processing result;
in this embodiment, it should be noted that the mental health state information text is a dream description text of a target user and is used to describe the dream information of the target user, and the mental health state prediction method is applied to an intelligent dream psychological analysis platform, where the target user can input the dream description text and user characteristic information on the intelligent dream psychological analysis platform to further obtain a mental health state prediction result output by the intelligent dream psychological analysis platform, where the user characteristic information is associated information of user characteristics, and the user characteristics include user age, mental disease history, gender, and the like.
The method comprises the steps of obtaining a mental health state information text of a target user, carrying out natural language processing on the mental health state information text to obtain a natural language processing result, specifically, receiving the mental health state information text input by the target user, carrying out natural language processing on the mental health state information text to divide the mental health state information text into sentences from the text, dividing the sentences into phrases, dividing the phrases into text words, and keeping an interactive relation among the text words to obtain the natural language processing result, wherein the natural language processing result comprises the sentences, the phrases, the text words and the interactive relation among the text words corresponding to the mental health state information text, wherein the interactive relation is a connection relation among the text words which are adjacent to each other, and the type of the interactive relation comprises an offensive emotion type, Friendly emotion type and neutral emotion type.
Wherein the natural language processing result comprises a text word expression vector and an interaction relation expression vector,
the step of performing natural language processing on the mental health state information text to obtain a natural language processing result comprises the following steps:
step S11, constructing a text representation tree network corresponding to the mental health state information text, wherein the text representation tree network at least comprises a text word and an interaction relation between the text words;
in this embodiment, it should be noted that the text word expression vector is a word vector representing a text word, and the interaction relationship expression vector is an encoding vector representing an interaction relationship between text words.
Constructing a text representation tree network corresponding to the mental health state information text, wherein the text representation tree network at least comprises a text word and an interactive relationship between the text word, and specifically, the mental health state information text is divided into sentences from the text, the sentences are divided into phrases, the phrases are further divided into text words, the mental health state information text, the sentences corresponding to the mental health state information text, the phrases, the text words and the interactive relationship between the text words are further constructed into the text representation tree network, wherein a root node of the text representation tree network represents the mental health state information text, leaf nodes of the text representation tree network represent the text words, and other intermediate nodes represent the sentences corresponding to the mental health state information text and the corresponding phrases, and the child nodes of the text are sentences, the child nodes of the sentences are phrases, the child nodes of the phrases are text words, and the connection exists between the text words, which indicates that the interactive relationship exists between the two text words.
Step S12, vectorizing each text word to obtain each text word expression vector;
in this embodiment, vectorizing each text word to obtain each text word expression vector, specifically, mapping each text word to a preset word vector space, so as to convert each text word into a corresponding word vector, and obtain a text word expression vector corresponding to each text word.
In another embodiment, the vectorizing each text word to obtain a vector representing each text word includes:
generating word vectors, part-of-speech vectors and word position vectors corresponding to the text words, wherein the word vectors are encoding vectors representing the text words and are used for uniquely representing the text words, the part-of-speech vectors are encoding vectors representing the parts of speech of the text words, and the text word position vectors are encoding vectors representing the positions of the text words in the text sentences, and further generating text word representing vectors corresponding to the text words based on the text word vectors corresponding to the text words, the corresponding text part-of-speech vectors and the corresponding text word position vectors.
And step S13, splicing every two text word expression vectors with the interaction relationship respectively to obtain each interaction relationship expression vector.
In this embodiment, it should be noted that two text words corresponding to one interaction relationship are used.
And splicing every two text word expression vectors with the interaction relationship to obtain each interaction relationship expression vector, specifically, splicing the text word expression vectors corresponding to the two text words corresponding to each interaction relationship to obtain the interaction relationship expression vector corresponding to each interaction relationship, for example, assuming that the two words with the interaction relationship are word a and word B, the text word expression vector corresponding to word a is (a, B), and the text word expression vector corresponding to word B is (c, d), the interaction relationship expression vector is (a, B, c, d).
Step S20, based on a preset emotion classification model, carrying out emotion classification on each text word in the natural language processing result and the interaction relation among the text words to obtain an emotion classification result;
in this embodiment, it should be noted that the preset emotion classification model includes a text word emotion classification model and an interaction relationship emotion classification model, where the text word emotion classification model is a machine learning model for classifying text words, and the interaction relationship emotion classification model is a machine learning model for classifying interaction relationships.
Performing emotion classification on each text word in the natural language processing result and an interaction relation between the text words based on a preset emotion classification model to obtain an emotion classification result, specifically, performing emotion classification on each text word based on the text word emotion classification model respectively to judge a first emotion type corresponding to each text word to obtain a text word classification result corresponding to each text word, performing emotion classification on the interaction relation between the text words based on the interaction relation emotion classification model to judge a second emotion type corresponding to each interaction relation to obtain an interaction relation classification result corresponding to each interaction relation, and taking each text word classification result and each interaction relation classification result as the emotion classification result together.
Wherein the emotion classification result comprises a text word classification result and an interaction relation classification result, the preset emotion classification model comprises a text word emotion classification model and an interaction relation emotion classification model,
the method comprises the following steps of carrying out emotion classification on each text word in the natural language processing result and the interactive relation among the text words based on a preset emotion classification model to obtain an emotion classification result, wherein the emotion classification result comprises the following steps:
step S21, based on the text word emotion classification model, performing emotion classification on each text word respectively to judge a first emotion type corresponding to each text word and obtain a text word classification result;
in this embodiment, based on the text word emotion classification model, emotion classification is performed on each text word, so as to determine a first emotion type corresponding to each text word, and obtain a text word classification result, specifically, the following steps are performed for each text word: based on the text word emotion classification model, mapping the text word expression vector to a classification regression value, and further mapping the classification regression value to an emotion tag, for example, the classification regression value may be mapped to an emotion tag through a softmax function, where the emotion tag is a type identifier of an emotion type, and based on the emotion tag, an emotion type corresponding to a text word may be determined, and a text word classification result is obtained, where the first emotion type includes a positive emotion type and a negative emotion type.
Step S22, based on the interaction relation emotion classification model, performing emotion classification on the interaction relation between the text words to judge a second emotion type corresponding to each interaction relation, and obtaining an interaction relation classification result.
In this embodiment, based on the interaction relationship emotion classification model, emotion classification is performed on an interaction relationship between the text words to determine a second emotion type corresponding to each interaction relationship, so as to obtain an interaction relationship classification result, and specifically, the following steps are performed for each interaction relationship:
based on the interaction relation emotion classification model, mapping an interaction relation expression vector corresponding to the interaction relation into a classification probability vector, wherein the classification probability vector is a vector composed of classification probabilities that the interaction relation belongs to different second emotion types, for example, assuming that the classification probability vector is (a, B, C), wherein a is a probability that the interaction relation belongs to a second emotion type a, B is a probability that the interaction relation belongs to a second emotion type B, and C is a probability that the interaction relation belongs to a second emotion type C, and then determining that the interaction relation belongs to the second emotion type corresponding to the maximum classification probability based on each classification probability in the classification probability vector, and then obtaining the interaction relation classification result.
And step S30, based on the emotion classification result, predicting the mental health state of the target user to obtain a target prediction result.
In this embodiment, it should be noted that the emotion classification result includes a first type emotion classification tag corresponding to each text word and a second type emotion classification tag corresponding to each interaction relationship, where the first type emotion classification tag includes a negative emotion tag corresponding to a negative emotion type and a positive emotion tag corresponding to a positive emotion type, and the second type emotion classification tag includes an aggressive emotion tag corresponding to an aggressive emotion type, a friendly emotion tag corresponding to a friendly emotion type, and a neutral emotion tag corresponding to a neutral emotion type.
Performing mental health state prediction on the target user based on the emotion classification result to obtain a target prediction result, specifically, performing a first type emotion score on the target user based on a first type emotion classification tag corresponding to each text word to obtain a first type score, wherein the first type score is a score for evaluating whether the dream of the target user is biased to negative emotion or positive emotion, and performing a second type emotion score on the target user based on a second type emotion classification tag corresponding to each interaction relation to obtain a second type score, wherein the second type score is a score for evaluating whether the dream of the target user is biased to aggressive emotion or friendly emotion or neutral emotion, and further based on the first type emotion score and the second type emotion score, and predicting the mental health state of the target user to obtain a target prediction result, for example, in one embodiment, calculating a weighted average of the first type emotion score and the second type emotion score, and judging whether the mental health state of the target user is good or bad according to the size of the weighted average to obtain the target prediction result.
Wherein the emotion classification result at least comprises a first type emotion classification label corresponding to the text word and a second type emotion classification label corresponding to the interaction relation,
the step of predicting the mental health state of the target user based on the emotion classification result to obtain a target prediction result comprises the following steps:
step S31, generating text analysis information corresponding to the mental health state information text based on each first type emotion classification label and each second type emotion classification label;
in this embodiment, text analysis information corresponding to the mental health state information text is generated based on each first type emotion classification tag and each second type emotion classification tag, specifically, first type emotion scoring is performed on the target user based on the first type emotion classification tag corresponding to each text word to obtain first type scoring, second type emotion scoring is performed on the target user based on the second type emotion classification tag corresponding to each interaction relationship to obtain second type scoring, and the first type emotion scoring and the second type emotion scoring are combined to form a scoring vector, and the scoring vector is used as the text analysis information.
Wherein the first type of emotion classification tags include negative emotion tags and positive emotion tags, the second type of emotion classification tags include offensive emotion tags, friendly emotion tags, and neutral emotion tags,
the step of generating text analysis information corresponding to the mental health state information text based on each of the first type emotion classification tags and each of the second type emotion classification tags includes:
step S311, calculating a first type emotion score based on each of the negative emotion tags and each of the positive emotion tags;
in this embodiment, a first type emotion score is calculated based on each negative emotion tag and each positive emotion tag, specifically, a ratio of the negative emotion tag to the positive emotion tag in each negative emotion tag and each positive emotion tag is calculated, and then a first type score vector is generated as the first type emotion score based on the ratio of the negative emotion tag to the positive emotion tag, where the ratio of the negative emotion tag to the number of the negative emotion tags is a ratio of the positive emotion tags, and for example, if the ratio of the negative emotion tag to the number of the positive emotion tags is 20% and the ratio of the positive emotion to the positive emotion tags is 80%, the first type score vector is (20, 80).
Step S312, calculating a second type emotion score based on each offensive emotion label, each friendly emotion label and each neutral emotion label;
in this embodiment, a second type emotion score is calculated based on each of the offensive emotion tags, each of the friendly emotion tags, and each of the neutral emotion tags, and specifically, an offensive emotion tag proportion and a friendly emotion tag proportion in each of the offensive emotion tags, each of the friendly emotion tags, and each of the neutral emotion tags are calculated, and a second type score vector is generated as the second type emotion score based on the offensive emotion tag proportion and the friendly emotion tag proportion, wherein the offensive emotion tag proportion is a number of offensive emotion tags proportion, and the friendly emotion tag proportion is a number of friendly emotion tags proportion, for example, assuming that the number of offensive emotion tags is 40, the number of friendly emotion tags is 40, and the number of neutral emotion tags is 20, the offensive emotion label is 40%, the friendly emotion label is 40%, and the second type score vector is (40, 40).
Step 313, generating the text analysis information based on the first type emotion score and the second type emotion score.
In this embodiment, the text analysis information is generated based on the first type emotion score and the second type emotion score, specifically, the first type score vector and the second type score vector are spliced to obtain a spliced score vector, and the spliced score vector is used as the text analysis information
Step S32, obtaining user characteristic information of the target user, and performing mental health state prediction on the target user based on the text analysis information and the user characteristic information to obtain the target prediction result.
In this embodiment, it should be noted that the user characteristic information is associated information of user characteristics, where the user characteristics include age, gender, and history of psychological disease, and the text analysis information is an emotion score vector corresponding to a target user generated based on each of the first type emotion classification tags and each of the second type emotion classification tags, where the emotion score vector may be the concatenation score vector.
Acquiring user characteristic information of the target user, and based on the text analysis information and the user characteristic information, predicting the mental health state of the target user to obtain a target prediction result, specifically, receiving user characteristic information input by the target user, splicing the user characteristic representation vector corresponding to the user characteristic information with the emotion score vector to obtain a vector to be predicted, then inputting the vector to be predicted into a preset mental health state prediction vector to map the vector to be predicted into a model regression value, and further taking the model regression value as a mental health state score, the mental health state score is a score used for representing the quality degree of the mental health state of the target user, and the mental health state score is used as the target prediction result.
Compared with the technical means of carrying out mental health state prediction by obtaining a mental health state information text of a target user after a mental health state information text is obtained by carrying out dreams analysis in a grading mode in the prior art, the mental health state information text comprises a dreams description text, natural language processing is carried out on the mental health state information text to split the text into sentences, phrases, words and the like, the interactive relation between the words is determined to obtain a natural language processing result, emotion classification is automatically carried out on each text word in the natural language processing result and the interactive relation between each text word based on a preset emotion classification model, and emotion information corresponding to each text word in the mental health state information text is obtained, and emotion information corresponding to the interaction relationship among the text words, and further obtaining emotion classification results, wherein the purpose of performing emotion analysis of the text word dimension on the dream information is achieved, compared with a mode of grading the whole dream information, the dream analysis process performed by a preset emotion classification model which is a machine learning model and is constructed based on massive information is more detailed and comprehensive, the obtained dream analysis information (namely, emotion classification results) is more reliable, accurate and rich, and furthermore, based on the emotion classification results, the mental health state prediction is performed on the target user, so that the purpose of performing mental health state prediction based on more reliable, accurate and rich dream analysis information can be achieved, the target prediction results are obtained, and the problems that the mental health state is generally strange due to fewer cases of the dream mastered by psychologists are overcome, the variety is various, and then the dream analysis information that psychologist obtained is too thin and brief, and then leads to the lower technical defect of accuracy of mental health state prediction, has promoted mental health state prediction's accuracy.
Further, referring to fig. 2, in another embodiment of the present application, based on the first embodiment of the present application, the textual representation tree network includes at least one textual statement vector,
after the step of vectorizing each of the text words to obtain a vector representing each of the text words, the method for predicting mental health status further includes:
step A10, carrying out dependency syntax analysis on the text statement vector to obtain a dependency syntax analysis result;
in this embodiment, it should be noted that the process of dependency parsing is a process of parsing syntax information of a sentence, where the syntax information includes sentence pattern information and word component information, for example, assuming that a sentence is "who" i am ", after the dependency parsing, the sentence pattern information indicates that the sentence is a main-meaning object sentence, and the word component information indicates that" i "is a main-meaning," yes "is a predicate, and" who "is an object.
Performing dependency syntax analysis on the text statement vector to obtain a dependency syntax analysis result, specifically, inputting the text statement vector into a preset dependency syntax model, and performing dependency relationship judgment and dependency relationship type prediction on the text statement corresponding to the text statement vector, respectively, where it is to be noted that the preset dependency syntax model is a preset machine learning model for performing dependency syntax analysis, the dependency relationship judgment is performed to judge the dependency relationship between words, the dependency relationship type prediction is performed to predict the type of dependency relationship, for example, if there is a statement "ABC", where A, B and C are both words in a statement, after the dependency relationship judgment, B is determined to be dependent on a, C is determined to be dependent on B, and after the dependency relationship type prediction, the dependency relationship between a and B is determined to be the cardinal relation, and B, obtaining a dependency syntax analysis result by performing dependency syntax analysis on the text statement vector in an implementable manner, wherein the step of obtaining the dependency syntax analysis result comprises the following steps:
performing dependency relationship determination on a text sentence of the text sentence vector to obtain a dependency relationship determination result, performing dependency relationship type prediction on the text sentence to obtain a dependency relationship type prediction result, and further fusing the dependency relationship determination result and the dependency relationship type prediction result to obtain a dependency relationship type label between words in the text sentence, wherein the dependency relationship type label is an identifier of a dependency relationship type, and further based on the dependency relationship type label, period information and word component information of the text sentence can be determined to obtain the dependency syntax analysis result, wherein the dependency relationship determination result can be represented by a vector, the dependency relationship determination result in a vector form is a dependency relationship determination vector, and the dependency relationship type prediction result can be represented by a matrix, the matrix form corresponding to the dependency relationship type prediction result is a dependency relationship type prediction probability matrix, wherein the value at each bit in the dependency type prediction probability matrix is a dependency type tag probability prediction vector between one word and another word in the text sentence, wherein the value of each bit in the dependency type label probability prediction vector is a probability value of the preset dependency corresponding to the bit belonging to the dependency of one word and the other word in the text sentence, wherein the preset dependency relationship includes a predicate relationship, a move-guest relationship, etc., for example, assuming that the dependency relationship type label probability prediction vector between the word a and the word B is (0.1, 0.9), then 0.1 indicates a 10% probability of a dominance relationship between word a and word B and 0.9 indicates a 90% probability of a motile relationship between word a and word B.
In an implementation scheme, the step of fusing the dependency vector and the dependency type prediction probability matrix to obtain the dependency type labels from word to word in the text sentence includes:
and aggregating the dependency relationship vector and each dependency relationship type label probability prediction vector in the dependency relationship type prediction probability matrix to obtain an aggregated vector corresponding to each dependency relationship type label probability prediction vector, wherein the aggregation comprises weighted summation, splicing, summation and the like, then selecting a maximum bit value from bit values in the aggregated vector for each aggregated vector, and using a preset dependency relationship label corresponding to a bit corresponding to the maximum bit value as a dependency relationship type label corresponding to the dependency relationship type label probability prediction vector.
Wherein, the step of performing dependency syntax analysis on each text statement vector to obtain a dependency syntax analysis result comprises:
step A11, based on a preset dependency relationship discrimination model, performing dependency relationship discrimination on the text statement vector to obtain a dependency relationship discrimination result;
in this embodiment, it should be noted that the preset dependency syntax model includes a preset dependency relationship determination model, where the preset dependency relationship determination model is a machine learning model for determining whether there is a dependency relationship between words in the text sentence.
And judging the dependence relationship of the text statement vector based on a preset dependence relationship judging model to obtain a dependence relationship judging result, specifically, inputting the text statement vector into the preset dependence relationship judging model, and judging the dependence relationship of the text statement vector to judge whether the word in the text statement has the dependence relationship with the word to obtain the dependence relationship judging result.
Wherein the preset dependency relationship distinguishing model comprises a first feature extraction model, a first fully connected network, a second fully connected network and a first affine-doubly transformed network,
the step of judging the dependency relationship of the text statement vector based on the preset dependency relationship judging model to obtain a dependency relationship judging result comprises the following steps:
step A111, extracting features of the text statement vector based on the first feature extraction model to obtain a first feature extraction result;
in this embodiment, it should be noted that the first feature extraction model is a neural network that performs feature extraction on the text statement vector, and the first feature extraction model includes a Transformer model, an RNN network, a CNN network, and the like.
And performing feature extraction on the text statement vector based on the first feature extraction model to obtain a first feature extraction result, specifically, inputting the text statement vector into the first feature extraction model, performing feature extraction on the text statement vector to obtain a first feature extraction matrix, and taking the first feature extraction matrix as the first feature extraction result.
Step a112, based on the first fully connected network and the second fully connected network, fully connecting the first feature extraction results respectively to obtain a first sentence vector and a second sentence vector;
in this embodiment, the first feature extraction result is fully connected based on the first fully connected network and the second fully connected network, so as to obtain a first sentence vector and a second sentence vector, specifically, the first feature extraction matrix is input into the first fully connected network, the first feature extraction matrix is fully connected, so as to obtain a first sentence vector, the first feature extraction matrix is input into the second fully connected network, and the first feature extraction matrix is fully connected, so as to obtain a second sentence vector, where it is required to be noted that the first sentence vector includes at least one prefix vector for representing a representation vector of a word as a dependency in the dependency relationship, and the second sentence vector includes at least one suffix vector for representing a representation vector of a word as a dependency in the dependency relationship, for example, assuming that a word a is dependent on a word B, the word expression vector corresponding to the word B is a prefix vector, and the word expression vector corresponding to the word a is an end-of-word vector.
Step A113, based on the first affine-affine transformation network, performing affine-transformation on the first sentence vector and the second sentence vector to obtain a dependency relationship score matrix;
in this embodiment, based on the first affine-doubly-transformed network, the first sentence vector and the second sentence vector are subjected to affine-doubly-transformed to obtain a dependency score matrix, and specifically, the first sentence vector and the second sentence vector are input into the first affine-doubly-transformed network, and the first sentence vector and the second sentence vector are subjected to affine-doubly-transformed to calculate a probability score of a dependency relationship existing between each prefix vector in the first sentence vector and each suffix vector in the second sentence vector, and obtain a dependency score matrix, wherein the dependency score matrix is a score matrix composed of probability scores of a dependency relationship existing between each prefix vector and each suffix vector.
Step a114, determining the dependency relationship determination result based on the dependency relationship score matrix.
In this embodiment, the dependency relationship determination result is determined based on the dependency relationship score matrix, specifically, based on a preset maximal spanning tree algorithm, a maximal probability score sum satisfying a preset score selection condition is selected from the dependency relationship score matrix, and a dependency relationship vector composed of text word representation vectors corresponding to dependency relationships corresponding to the maximal probability score and corresponding target probability scores is used as the dependency relationship determination result, where the preset score selection condition includes that text words corresponding to the target probability scores are in one-to-one correspondence with text words in the text sentence, for example, each target probability score is assumed to be a and B, where a represents a probability score that a word B is attached to a word a, B represents a probability score that a word c is attached to a word B, and a word a corresponds to a vectorized word as a vector X, the word b corresponds to the vectorized word as a vector Y, the word c corresponds to the vectorized word as a vector Z, and the dependency relationship vector is a vector (X, 1, 0, 0, 1, Y, 1, 0, 0, 1, Z), where (1, 0, 0, 1) indicates that there is dependency relationship between words.
Step A12, based on the preset dependency relationship type prediction model and the dependency relationship determination result, performing dependency relationship type prediction on the text statement vector to obtain the dependency syntax analysis result.
In this embodiment, it should be noted that the preset dependency syntax model includes a preset dependency type prediction model, where the preset dependency type prediction model is a machine learning model for predicting a dependency type between words in a text sentence.
Performing dependency type prediction on the text statement vector based on a preset dependency type prediction model and the dependency discrimination result to obtain the dependency syntax analysis result, and specifically, performing dependency type prediction on the text statement vector based on the preset dependency type prediction model to obtain a dependency type probability score matrix, where it is to be noted that there is a dependency type probability score vector on each bit in the dependency type probability score matrix, where a value on each bit of the dependency type probability score vector is a probability score of a preset dependency type, for example, assuming that the dependency type probability score vector is (a, B), and a first predicate of the dependency type probability score vector corresponds to a principal relationship and a second predicate of the dependency type probability score vector corresponds to a guest relationship, and if the third bit corresponds to the parallel relationship, A is the probability score of the dominating relationship between the two words corresponding to the dependency relationship type probability score vector, B is the probability score of the motile relationship between the two words corresponding to the dependency relationship type probability score vector, then based on the dependency relationship determination result, each target dependency relationship type probability score vector is selected from the dependency relationship type probability score matrix, then the dependency relationship type corresponding to the maximum value in each target dependency relationship type probability score vector is used as the target dependency relationship type, further the dependency relationship type between the text statement words and words is obtained, and based on the dependency relationship type between the text statement words and words, the sentence formula of the text statement and the word component of each text word in the text statement are determined, further the dependency relationship type between the text statement words and words is determined, For example, if the text sentence is ABC, where the dependency relationship type between the word a and the word B is a predicate type, the dependency relationship type between the word B and the word C is a verb type, it may be determined that the sentence of the text sentence is a predicate, the word component corresponding to the word a is a subject, the word component corresponding to the word B is a predicate, and the word component corresponding to the word C is an object.
Wherein the dependency relationship determination result comprises a dependency relationship vector,
the step of performing dependency type prediction on the text statement vector based on a preset dependency type prediction model and the dependency discrimination result to obtain the dependency syntax analysis result includes:
step A121, based on the preset dependency relationship type prediction model, performing dependency relationship type prediction on the text statement vector to obtain a dependency relationship type probability score matrix;
in this embodiment, it should be noted that the preset dependency type prediction model includes a second feature extraction model, a third fully-connected network, a fourth fully-connected network, and a second doubly-affine transformation network.
Based on the preset dependency relationship type prediction model, performing dependency relationship type prediction on the text statement vector to obtain a dependency relationship type probability score matrix, specifically, inputting the text statement vector into a second feature extraction model, performing feature extraction on the text statement vector to obtain a second feature extraction matrix, inputting the second feature extraction matrix into a third full-connection network and a fourth full-connection network respectively to obtain a third sentence vector and a corresponding fourth sentence vector corresponding to the second feature extraction matrix, inputting the third sentence vector and the fourth sentence vector into a second double affine transformation network, and performing double affine transformation on the third sentence vector and the fourth sentence vector to obtain the dependency relationship type probability score matrix.
Step A122, the dependency relationship type probability score matrix and the dependency relationship vector are fused to obtain the dependency syntax analysis result.
In this embodiment, the dependency type probability score matrix and the dependency vector are fused to obtain the dependency syntax analysis result, and specifically, based on a preset fusion rule, each dependency type probability score vector in the dependency type probability score matrix is fused with the dependency vector to obtain a dependency type probability vector corresponding to each dependency type probability score vector, where the preset fusion rule includes weighted average, concatenation, summation, and the like, a value on each bit of the dependency type probability vector is a probability of a preset dependency type, the preset dependency type includes a predicate type, a move-guest type, a parallel relationship type, and the like, and then a maximum probability value is selected from each dependency type probability vector as a target dependency type probability, and determining the dependency relationship type corresponding to each maximum dependency relationship type probability which meets a preset probability selection condition in each target dependency relationship type probability, wherein the preset probability selection condition comprises that the text words corresponding to each selected maximum dependency relationship type probability correspond to each text word in the text sentence one by one.
Step A20, the dependency relationship type representation vector corresponding to each of the interaction relationships in the dependency syntax analysis result is spliced with the corresponding text word representation vector to obtain each of the interaction relationship representation vectors.
In this embodiment, the dependency relationship type representation vector corresponding to each of the interaction relationships in the dependency syntax analysis result is spliced with the corresponding text word representation vector to obtain each of the interaction relationship representation vectors, and specifically, the following steps are performed for each of the interaction relationships:
determining a dependency relationship type representation vector corresponding to the interaction relationship in the dependency syntax analysis result, and splicing the dependency relationship type representation vector with two text word representation vectors corresponding to the interaction relationship to obtain an interaction relationship representation vector corresponding to the interaction relationship, wherein the dependency relationship type representation vector is an encoding vector representing the dependency relationship type between the text words and the text words, and since the dependency relationship type is associated with semantic information of sentences, the purpose of generating the interaction relationship representation vector based on the semantic information of the text sentences is achieved, so that decision basis information when psychological health state prediction is performed based on the interaction relationship representation vector is richer, and the accuracy and reliability of the psychological health state prediction are improved.
The embodiment of the application provides a natural language processing method based on dependency syntax analysis, that is, dependency syntax analysis is performed on text statement vectors to obtain dependency syntax analysis results, and then corresponding dependency relationship type expression vectors and corresponding text word expression vectors of each interaction relationship in the dependency syntax analysis results are spliced to obtain each interaction relationship expression vector, so as to achieve the purpose of generating an interaction relationship expression vector with semantic information of a text statement, wherein a natural language processing result comprises each interaction relationship expression vector, and further emotion classification is performed on each text word and the interaction relationship between each text word in the natural language processing results based on a preset emotion classification model to obtain emotion classification results, that is, the interaction relationship expression vectors based on the semantic information of the text statement can be achieved, and performing emotion classification on the interaction relation among the text words in the natural language processing result, and then performing mental health state prediction on the target user based on the emotion classification result to obtain a target prediction result, so that decision basis information in the process of performing mental health state prediction based on the interaction relation expression vector is richer, and the accuracy and reliability of the mental health state prediction are improved.
Referring to fig. 3, fig. 3 is a schematic device structure diagram of a hardware operating environment according to an embodiment of the present application.
As shown in fig. 3, the mental health state prediction apparatus may include: a processor 1001, such as a CPU, a memory 1005, and a communication bus 1002. The communication bus 1002 is used for realizing connection communication between the processor 1001 and the memory 1005. The memory 1005 may be a high-speed RAM memory or a non-volatile memory (e.g., a magnetic disk memory). The memory 1005 may alternatively be a memory device separate from the processor 1001 described above.
Optionally, the mental health state prediction device may further include a rectangular user interface, a network interface, a camera, an RF (Radio Frequency) circuit, a sensor, an audio circuit, a WiFi module, and the like. The rectangular user interface may comprise a Display screen (Display), an input sub-module such as a Keyboard (Keyboard), and the optional rectangular user interface may also comprise a standard wired interface, a wireless interface. The network interface may optionally include a standard wired interface, a wireless interface (e.g., WI-FI interface).
Those skilled in the art will appreciate that the mental health state prediction apparatus configuration shown in fig. 3 does not constitute a limitation of the mental health state prediction apparatus, and may include more or fewer components than those shown, or some of the components may be combined, or a different arrangement of components.
As shown in fig. 3, the memory 1005, which is a kind of computer storage medium, may include an operating system, a network communication module, and a mental health state prediction program therein. The operating system is a program that manages and controls the mental health state predicting device hardware and software resources, and supports the operation of the mental health state predicting program as well as other software and/or programs. The network communication module is used for communication among the components in the memory 1005 and with other hardware and software in the mental health state prediction system.
In the mental health state prediction apparatus shown in fig. 3, the processor 1001 is configured to execute a mental health state prediction program stored in the memory 1005 to implement the steps of any one of the mental health state prediction methods described above.
The specific implementation of the mental health state prediction device of the present application is substantially the same as the embodiments of the mental health state prediction method described above, and details are not repeated here.
An embodiment of the present application further provides a mental health state prediction apparatus, where the mental health state prediction apparatus is applied to a mental health state prediction device, and the mental health state prediction apparatus includes:
the natural language processing module is used for acquiring a mental health state information text of a target user, and performing natural language processing on the mental health state information text to acquire a natural language processing result;
the emotion classification module is used for carrying out emotion classification on each text word in the natural language processing result and the interaction relation among the text words based on a preset emotion classification model to obtain an emotion classification result;
and the prediction module is used for predicting the mental health state of the target user based on the emotion classification result to obtain a target prediction result.
Optionally, the emotion classification module is further configured to:
based on the text word emotion classification model, performing emotion classification on each text word respectively to judge a first emotion type corresponding to each text word and obtain a text word classification result;
and classifying the emotion of the interaction relation between the text words based on the interaction relation emotion classification model so as to judge a second emotion type corresponding to each interaction relation and obtain an interaction relation classification result.
Optionally, the prediction module is further configured to:
generating text analysis information corresponding to the mental health state information text based on each first type emotion classification label and each second type emotion classification label;
and acquiring user characteristic information of the target user, and predicting the mental health state of the target user based on the text analysis information and the user characteristic information to obtain a target prediction result.
Optionally, the prediction module is further configured to:
calculating a first type sentiment score based on each of the negative sentiment tags and each of the positive sentiment tags;
calculating a second type emotion score based on each of the offensive emotion tags, each of the friendly emotion tags, and each of the neutral emotion tags;
generating the text analysis information based on the first type of sentiment score and the second type of sentiment score.
Optionally, the natural language processing module is further configured to:
constructing a text representation tree network corresponding to the mental health state information text, wherein the text representation tree network at least comprises a text word and an interactive relation between the text words;
vectorizing each text word to obtain a representation vector of each text word;
and respectively splicing every two text word expression vectors with the interaction relationship to obtain each interaction relationship expression vector.
Optionally, the mental health state prediction device is further configured to:
performing dependency syntax analysis on the text statement vector to obtain a dependency syntax analysis result;
and respectively splicing the dependency relationship type representation vector corresponding to each interactive relationship in the dependency syntax analysis result with the corresponding text word representation vector to obtain each interactive relationship representation vector.
Optionally, the mental health state prediction device is further configured to:
based on a preset dependency relationship judging model, judging the dependency relationship of the text statement vector to obtain a dependency relationship judging result;
and performing dependency relationship type prediction on the text statement vector based on a preset dependency relationship type prediction model and the dependency relationship judgment result to obtain a dependency syntax analysis result.
The detailed implementation of the mental health state prediction apparatus of the present application is substantially the same as the embodiments of the mental health state prediction method, and is not described herein again.
The present invention provides a medium, which is a readable storage medium, and the readable storage medium stores one or more programs, which are further executable by one or more processors for implementing the steps of any one of the above mental health state prediction methods.
The specific implementation of the readable storage medium of the present application is substantially the same as the embodiments of the mental health status prediction method, and is not described herein again.
The present application provides a computer program product, and the computer program product includes one or more computer programs, which can also be executed by one or more processors for implementing the steps of any one of the above mental health state prediction methods.
The specific implementation of the computer program product of the present application is substantially the same as the embodiments of the mental health status prediction method, and is not further described herein.
The above description is only a preferred embodiment of the present application, and not intended to limit the scope of the present application, and all modifications of equivalent structures and equivalent processes, which are made by the contents of the specification and the drawings, or which are directly or indirectly applied to other related technical fields, are included in the scope of the present application.
Claims (10)
1. A mental health state prediction method, characterized by comprising:
acquiring a mental health state information text of a target user, and performing natural language processing on the mental health state information text to obtain a natural language processing result;
based on a preset emotion classification model, carrying out emotion classification on each text word in the natural language processing result and the interaction relation among the text words to obtain an emotion classification result;
and predicting the mental health state of the target user based on the emotion classification result to obtain a target prediction result.
2. The mental health state prediction method of claim 1, wherein the emotion classification result includes a text word classification result and an interaction relationship classification result, the preset emotion classification model includes a text word emotion classification model and an interaction relationship emotion classification model,
the method comprises the following steps of carrying out emotion classification on each text word in the natural language processing result and the interactive relation among the text words based on a preset emotion classification model to obtain an emotion classification result, wherein the emotion classification result comprises the following steps:
based on the text word emotion classification model, performing emotion classification on each text word respectively to judge a first emotion type corresponding to each text word and obtain a text word classification result;
and classifying the emotion of the interaction relation between the text words based on the interaction relation emotion classification model so as to judge a second emotion type corresponding to each interaction relation and obtain an interaction relation classification result.
3. The mental health state prediction method of claim 1, wherein the emotion classification result comprises at least a first type emotion classification label corresponding to the text word and a second type emotion classification label corresponding to the interaction relationship,
the step of predicting the mental health state of the target user based on the emotion classification result to obtain a target prediction result comprises the following steps:
generating text analysis information corresponding to the mental health state information text based on each first type emotion classification label and each second type emotion classification label;
and acquiring user characteristic information of the target user, and predicting the mental health state of the target user based on the text analysis information and the user characteristic information to obtain a target prediction result.
4. The mental health state prediction method of claim 3, wherein the first type emotion classification tags include negative emotion tags and positive emotion tags, the second type emotion classification tags include offensive emotion tags, friendly emotion tags, and neutral emotion tags,
the step of generating text analysis information corresponding to the mental health state information text based on each of the first type emotion classification tags and each of the second type emotion classification tags includes:
calculating a first type sentiment score based on each of the negative sentiment tags and each of the positive sentiment tags;
calculating a second type emotion score based on each of the offensive emotion tags, each of the friendly emotion tags, and each of the neutral emotion tags;
generating the text analysis information based on the first type of sentiment score and the second type of sentiment score.
5. The mental health state prediction method of claim 1, wherein the natural language processing result includes a text word expression vector and an interaction relation expression vector,
the step of performing natural language processing on the mental health state information text to obtain a natural language processing result comprises the following steps:
constructing a text representation tree network corresponding to the mental health state information text, wherein the text representation tree network at least comprises a text word and an interactive relation between the text words;
vectorizing each text word to obtain a representation vector of each text word;
and respectively splicing every two text word expression vectors with the interaction relationship to obtain each interaction relationship expression vector.
6. The mental health state prediction method of claim 5 wherein the network of textual representation trees includes at least one textual statement vector,
after the step of vectorizing each of the text words to obtain a vector representing each of the text words, the method for predicting mental health status further includes:
performing dependency syntax analysis on the text statement vector to obtain a dependency syntax analysis result;
and respectively splicing the dependency relationship type representation vector corresponding to each interactive relationship in the dependency syntax analysis result with the corresponding text word representation vector to obtain each interactive relationship representation vector.
7. The mental health state prediction method of claim 6, wherein the step of performing dependency parsing on each text statement vector to obtain a dependency parsing result comprises:
based on a preset dependency relationship judging model, judging the dependency relationship of the text statement vector to obtain a dependency relationship judging result;
and performing dependency relationship type prediction on the text statement vector based on a preset dependency relationship type prediction model and the dependency relationship judgment result to obtain a dependency syntax analysis result.
8. A mental health state prediction apparatus characterized by comprising: a memory, a processor, and a program stored on the memory for implementing the mental health state prediction method,
the memory is used for storing a program for realizing the mental health state prediction method;
the processor is configured to execute a program implementing the mental health state prediction method to implement the steps of the mental health state prediction method according to any one of claims 1 to 7.
9. A medium, which is a readable storage medium, characterized in that the readable storage medium has stored thereon a program for implementing a mental health state prediction method, the program being executed by a processor to implement the steps of the mental health state prediction method according to any one of claims 1 to 7.
10. A computer program product comprising a computer program, wherein the computer program when executed by a processor implements the steps of the mental health state prediction method of any of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110128781.2A CN112735564A (en) | 2021-01-29 | 2021-01-29 | Mental health state prediction method, mental health state prediction apparatus, mental health state prediction medium, and computer program product |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110128781.2A CN112735564A (en) | 2021-01-29 | 2021-01-29 | Mental health state prediction method, mental health state prediction apparatus, mental health state prediction medium, and computer program product |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112735564A true CN112735564A (en) | 2021-04-30 |
Family
ID=75594820
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110128781.2A Pending CN112735564A (en) | 2021-01-29 | 2021-01-29 | Mental health state prediction method, mental health state prediction apparatus, mental health state prediction medium, and computer program product |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112735564A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113689936A (en) * | 2021-09-02 | 2021-11-23 | 广东厚阅科技有限公司 | Intelligent mental health state detection method and related device |
| CN113925376A (en) * | 2021-11-15 | 2022-01-14 | 杭州赛孝健康科技有限公司 | Intelligent toilet chair |
-
2021
- 2021-01-29 CN CN202110128781.2A patent/CN112735564A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113689936A (en) * | 2021-09-02 | 2021-11-23 | 广东厚阅科技有限公司 | Intelligent mental health state detection method and related device |
| CN113925376A (en) * | 2021-11-15 | 2022-01-14 | 杭州赛孝健康科技有限公司 | Intelligent toilet chair |
| CN113925376B (en) * | 2021-11-15 | 2022-11-15 | 杭州赛孝健康科技有限公司 | Intelligent potty chair |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111026842B (en) | Natural language processing method, natural language processing device and intelligent question-answering system | |
| US10649990B2 (en) | Linking ontologies to expand supported language | |
| CN109284399B (en) | Similarity prediction model training method and device and computer readable storage medium | |
| CN110727779A (en) | Question-answering method and system based on multi-model fusion | |
| CN108664599B (en) | Intelligent question-answering method and device, intelligent question-answering server and storage medium | |
| CN107807968B (en) | Question answering device and method based on Bayesian network and storage medium | |
| CN111680159A (en) | Data processing method and device and electronic equipment | |
| JP2020521210A (en) | Information processing method and terminal, computer storage medium | |
| Lin et al. | Deep structured scene parsing by learning with image descriptions | |
| CN113806550A (en) | Generation method and device of personalized knowledge graph and computer equipment | |
| CN112926308B (en) | Method, device, equipment, storage medium and program product for matching text | |
| CN112735564A (en) | Mental health state prediction method, mental health state prediction apparatus, mental health state prediction medium, and computer program product | |
| CN112069799A (en) | Dependency syntax based data enhancement method, apparatus and readable storage medium | |
| CN111242710A (en) | Business classification processing method and device, service platform and storage medium | |
| CN111475645B (en) | Knowledge point labeling method, knowledge point labeling device and computer readable storage medium | |
| CN116821373A (en) | Map-based prompt recommendation method, device, equipment and medium | |
| CN113868419B (en) | Text classification method, device, equipment and medium based on artificial intelligence | |
| CN115114974A (en) | Model distillation method, device, computer equipment and storage medium | |
| CN114840685A (en) | Emergency plan knowledge graph construction method | |
| CN114330704A (en) | Statement generation model updating method and device, computer equipment and storage medium | |
| CN113220854A (en) | Intelligent dialogue method and device for machine reading understanding | |
| CN113505786A (en) | Test question photographing and judging method and device and electronic equipment | |
| CN117112767A (en) | Question and answer result generation method, commercial query big model training method and device | |
| CN110232328A (en) | A kind of reference report analytic method, device and computer readable storage medium | |
| CN112529743B (en) | Contract element extraction method, device, electronic equipment and medium |
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 |