CN109979592A - Mental health method for early warning, user terminal, server and system - Google Patents
Mental health method for early warning, user terminal, server and system Download PDFInfo
- Publication number
- CN109979592A CN109979592A CN201910231911.8A CN201910231911A CN109979592A CN 109979592 A CN109979592 A CN 109979592A CN 201910231911 A CN201910231911 A CN 201910231911A CN 109979592 A CN109979592 A CN 109979592A
- Authority
- CN
- China
- Prior art keywords
- mental health
- content
- sentiment analysis
- early warning
- level
- 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
Classifications
-
- 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
- G06F40/216—Parsing using statistical methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/284—Lexical analysis, e.g. tokenisation or collocates
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H20/00—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance
- G16H20/70—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to mental therapies, e.g. psychological therapy or autogenous training
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- 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
- G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- Public Health (AREA)
- Medical Informatics (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Epidemiology (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Primary Health Care (AREA)
- Computational Linguistics (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Pathology (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Hospice & Palliative Care (AREA)
- Psychiatry (AREA)
- Developmental Disabilities (AREA)
- Psychology (AREA)
- Social Psychology (AREA)
- Child & Adolescent Psychology (AREA)
- Probability & Statistics with Applications (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The present invention provides a kind of mental health method for early warning, which comprises detects the input state or browse state of user terminal；Obtain the content for inputting or browsing；Sentiment analysis is carried out to the content；Server is sent by the result for carrying out sentiment analysis to the content, by the server judgement and to issue early warning according to the level in mental health of user.The present invention also provides a kind of mental health early warning user terminal, server and systems, solve the problems, such as that psychological health states can not be tracked in time in the prior art, can be realized by sentiment analysis and track and alert in time.
Description
Technical field
The present invention relates to Psychological Evaluation technical fields, whole in particular to a kind of mental health method for early warning, user
End, server and system.
Background technique
Due to life stress increase etc., mental diseases patient's sustainable growth such as depression.Traditional mental health water
It is flat to need to be helped face-to-face by seeking doctor to psychological hospital by patient, then assessed by doctor.And actually many trouble
Person does not go to see a doctor actively natively, it is difficult to track level in mental health in time.In addition, even with Internet technology
It realizes online interrogation, similarly needs to assess the psychological health states of patient by the way of one-to-one independent interrogation.?
In the case that mental health medical resource is deficient, the patient numbers that shrink can receive are very limited, it is difficult to realize big rule
The level in mental health of mould tracks and management.The personnel not good enough for psychological health states, family's adjuvant treatment especially lack.The heart
Reason Disease major part rehabilitation life is tided in the family, and the mental health of patient can be improved in scientific residential care
Level carries out psychological intervention to patient in time.However, general family is difficult to make for the level in mental health of patients with depression
Professional judgement, nor have professional knowledge, psychology auxiliary shield can not be taken for the level in mental health of patients with depression
Reason and intervention.
It is above-mentioned can not track psychological health states in time in the prior art aiming at the problem that, not yet propose effective solution at present
Certainly scheme.
Summary of the invention
The present invention is directed to solve above-mentioned technical problem at least to a certain extent.
The embodiment of the invention provides a kind of mental health method for early warning, user terminal, server and systems, existing to solve
There is the problem of technology can not track psychological health states in time.
According to an aspect of an embodiment of the present invention, a kind of mental health method for early warning is provided, which comprises
Detect the input state or browse state of user terminal；
Obtain the content for inputting or browsing；
Sentiment analysis is carried out to the content；
Server is sent by the result for carrying out sentiment analysis to the content, by the server judgement and basis
The level in mental health of user issues early warning.
According to an aspect of an embodiment of the present invention, a kind of mental health method for early warning is provided, which comprises
Receive to the content progress sentiment analysis of user terminal as a result, the content is that user terminal acquires user's input
The interior current browsed web content of perhaps user inputted in method；
Level in mental health is judged according to the sentiment analysis result；
Warning information is issued according to level in mental health.
Another aspect according to an embodiment of the present invention provides a kind of mental health early warning user terminal, the user
Terminal includes:
Second detecting module, for detecting the input state or browse state of user terminal；
Module is obtained, for obtaining the content for inputting or browsing；
Analysis module, for carrying out sentiment analysis to the content；
Second sending module, for sending server for the result for carrying out sentiment analysis to the content, by described
Server judgement and according to the level in mental health of user issue early warning.
Another aspect according to an embodiment of the present invention provides a kind of mental health Warning Service device, the server
Include:
Receiving module, for receiving the content of user terminal, the content is defeated in user terminal acquisition user's input method
The interior current browsed web content of perhaps user entered；
Analysis module, for carrying out sentiment analysis to the content；
Judgment module, for judging level in mental health according to sentiment analysis result；
Warning module, for issuing warning information to doctor terminal according to level in mental health.
Other side according to an embodiment of the present invention, provides a kind of mental health Warning Service device, and the doctor is whole
End includes:
First receiving module, for receiving to the content progress sentiment analysis of user terminal as a result, the content is use
The interior current browsed web content of perhaps user inputted in family terminal acquisition user's input method；
First judgment module, for judging level in mental health according to the sentiment analysis result；
Warning module, for issuing warning information according to level in mental health.
The present invention solves the problems, such as that psychological health states can not be tracked in time in the prior art, and feelings can be passed through by providing one kind
Mental health method for early warning, user terminal, server and the system for tracking and alerting in time are realized in sense analysis.
Detailed description of the invention
Fig. 1 is a kind of mental health method for early warning flow chart of the embodiment of the present invention.
Fig. 2 is that a kind of mental health method for early warning of the embodiment of the present invention judges mental health water according to sentiment analysis result
Flat flow chart of steps.
Fig. 3 is that a kind of mental health method for early warning of the embodiment of the present invention issues warning information step according to level in mental health
Rapid flow chart.
Fig. 4 is a kind of mental health method for early warning flow chart of further embodiment of this invention.
Fig. 5 is that a kind of mental health method for early warning of the embodiment of the present invention carries out sentiment analysis steps flow chart to the content
Figure.
Fig. 6 is that a kind of mental health method for early warning of further embodiment of this invention carries out sentiment analysis step to the content
Flow chart.
Fig. 7 is the corresponding character types of selection described in a kind of mental health method for early warning of another embodiment of the present invention
Sentiment analysis tool analyze the emotion flow chart of steps of the content.
Fig. 8 is a kind of mental health method for early warning flow chart of yet another embodiment of the invention.
Fig. 9 is a kind of mental health method for early warning flow chart of another embodiment of the present invention.
Figure 10 is a kind of mental health method for early warning flow chart of yet another embodiment of the invention.
Figure 11 is a kind of mental health Warning Service device structural schematic diagram of the embodiment of the present invention.
Figure 12 is a kind of mental health early warning user terminal structural schematic diagram of the embodiment of the present invention.
Figure 13 is a kind of mental health early warning doctor terminal structural schematic diagram of the embodiment of the present invention.
Figure 14 is a kind of mental health early warning doctor terminal structural schematic diagram of the embodiment of the present invention.
Figure 15 is a kind of mental health early warning system structural schematic diagram of the embodiment of the present invention.
Wherein: 100, server；101, the first receiving module；102, first judgment module；103, warning module；104,
One statistical module；105, enquiry module；106, the first sending module；200, user terminal；201, the second detecting module；202, it obtains
Modulus block；203, analysis module；204, the second sending module；205, the second judgment module；206, selecting module；207, the second system
Count module；208, the second computing module；209, setup module；300, doctor terminal；301, third receiving module；302, third is raw
At module；303, third detecting module；304, third display module；400, family members' terminal；401, the 4th receiving module；402,
Four generation modules；403, the 4th detecting module；404, the 4th display module.
Specific embodiment
The attached figures are only used for illustrative purposes and cannot be understood as limitating the patent.
The following further describes the technical solution of the present invention with reference to the accompanying drawings and examples.
Embodiment 1
As shown in Figure 1, according to an aspect of an embodiment of the present invention, a kind of mental health method for early warning is provided, it is described
Method includes:
S110: receive to the content progress sentiment analysis of user terminal as a result, the content is that user terminal acquisition is used
The interior current browsed web content of perhaps user inputted in the input method of family；
S120: level in mental health is judged according to the sentiment analysis result；
S130: warning information is issued according to level in mental health.
In the specific implementation process, server is received to the content progress sentiment analysis of user terminal as a result, in described
Hold is that user terminal acquires the current browsed web content server of interior perhaps user inputted in user's input method according to emotion point
Analysis result judges level in mental health；Server issues warning information to doctor terminal according to level in mental health.Server root
Warning information is issued to family members' terminal according to level in mental health.
The sentiment analysis for the content that acquisition user inputs or browsed in the user terminal that it is held, according to sentiment analysis knot
Fruit judges the level in mental health of user, can be with real-time tracking user's heart in the case where not needing user's interrogation at hospital
Levels of mental health.
It just will appreciate that use due to by the way of active analysis level in mental health, solving the one-to-one artificial interrogation of tradition
The family psychological condition problem in short supply so as to cause psychological medical resource, so as to realize so that shrink services one-to-manyly
Large-scale level in mental health tracking and management.
The doctor terminal that server is held Xiang doctor according to level in mental health issues warning information, can make doctor and
When psychological monitoring and intervention are carried out to user, be conducive to the negative emotions for alleviating user.
Server issues warning information according to family members' terminal that level in mental health is held to family members, can be timely with family members
Psychological adjuvant treatment is carried out, the negative emotions for alleviating user are conducive to.
As shown in Fig. 2, in the specific implementation process, described the step of level in mental health is judged according to sentiment analysis result
Further include:
S121: statistical history sentiment analysis result；
S122: according to the diversity judgement level in mental health of history sentiment analysis result and current sentiment analysis result.
Specifically, described the step of level in mental health is judged according to sentiment analysis result further include: server statistics are gone through
History sentiment analysis result；Server is according to the diversity judgement mental health of history sentiment analysis result and current sentiment analysis result
It is horizontal.Specifically, the history sentiment analysis result may include different cycles history sentiment analysis as a result, as the previous day,
The last week, preceding January, the preceding first quarter, the previous year or the history sentiment analysis result since use.
In the specific implementation process, the diversity judgement according to history sentiment analysis result and current sentiment analysis result
Level in mental health step includes:
The level in mental health calculation formula are as follows:
Wherein, β is level in mental health value；
For the average value of history sentiment analysis result；
For current sentiment analysis as a result, the average value of i.e. last m sentiment analysis result.
Specifically, the average value and current sentiment analysis for comparing history sentiment analysis result are as a result, emotional value can be monitored
The size variation of variable quantity.It is changed greatly in β emotional value, and when being positive value, β is bigger, then level in mental health is poorer, psychological shape
State is more passive.When β is negative value, show level in mental health height, psychological condition is more positive than history emotion.0 is more leveled off in β
When, show that psychology fluctuation is smaller.
In the specific implementation process, the level in mental health calculation formula are as follows:
Wherein, β is level in mental health value；
For the average value of history sentiment analysis result in i-th of period；
For current sentiment analysis as a result, the average value of i.e. last m sentiment analysis result.
KβiFor the time correlation coefficient in i-th of period, and it is described
Specifically, closer to the period of current time, time correlation coefficient is bigger.Time correlation coefficient and be 1.One
In kind specific embodiment, i-th of the cycle time related coefficient be can be set are as follows:
1st period was the previous day, the previous day time correlation coefficient Kβ1=0.4；
2nd period was the last week, the last week time correlation coefficient Kβ2=0.3；
3rd period was preceding January, previous month related coefficient Kβ3=0.15；
4th period was previous season, previous season time correlation coefficient Kβ4=0.10；
5th period was the previous year, the previous year time correlation coefficient Kβ5=0.05.
Specifically, apart from the more long history sentiment analysis of current time as a result, for current sentiment analysis result difference value
Point of reference it is smaller.Time correlation coefficient is added, can make the level in mental health value more can reflect the difference value of psychology variation.
It is changed greatly in β emotional value, and when being positive value, β is bigger, then level in mental health is poorer, and psychological condition is more passive.It is negative in β
When value, show level in mental health height, psychological condition is more positive than history emotion.β more level off to 0 when, show psychological fluctuation
It is smaller.
Level in mental health is judged by emotion variable quantity, can react the size of the recent psychology fluctuation of user, it can
To react the psychological health states of user in time, carried out when discovery has the variation tendency of a large amount of negative emotion in a short time pre-
It is alert.
In a kind of exemplary embodiment, level in mental health can be divided into very actively, actively, in, passive, serious passiveness
Deng five grades, level in mental health value is equal to:
β is level in mental health value.Mental health water is judged according to sentiment analysis result described in the level in mental health
Flat step can be implemented as, and server judges whether the sentiment analysis result is lower than severe threshold；If being lower than serious threshold
Value, then the level in mental health is serious passive.Such as, the severe threshold is preset as 0.2, if the sentiment analysis result
It is 0.1, then β=(1-0.1)=0.9, the level in mental health are serious passive.
Level in mental health is judged by sentiment analysis end value, can reflect the level in mental health situation of user.
As shown in figure 3, in the specific implementation process, also being wrapped after the sending warning information step according to level in mental health
It includes:
S140: according to the user information and level in mental health, matched family is inquired from family members' supplementary knowledge library
Belong to assisted care knowledge point, family members' supplementary knowledge library, which stores, helps the relevant knowledge point of family members' assisted care；
S150: the corresponding family members of the user are sent by family members' assisted care knowledge point.
In the specific implementation process, described that matched mental health knowledge step is inquired from family members' supplementary knowledge library also
Include:
Inquire the corresponding family members of the user and family members and the customer relationship；
Matched family members are inquired from family members' supplementary knowledge library according to the relationship of the family members and the user to assist protecting
Manage knowledge point.
In the specific implementation process, described that matched mental health knowledge step is inquired from family members' supplementary knowledge library also
Include:
Calculate the knowledge point in family members' supplementary knowledge library and the knowledge degree of correlation of the user；
N knowledge points before being selected from big to small according to the knowledge degree of correlation；
Family members' terminal is sent by the knowledge point and its knowledge degree of correlation, with related according to knowledge by family members' terminal
Degree shows the knowledge point.
In the specific implementation process, the age of the knowledge degree of correlation and user, gender, level in mental health, psychological disease
Sick type and family members are related to the relationship of user.The attribute of knowledge point is set in family members' supplementary knowledge library, and the attribute includes
Age of user, gender, level in mental health, the relationship of mental disease type and family members and user.It calculates separately user and knows
Know the degree of correlation of the correspondence attribute of point, and all properties are added, is i.e. the knowledge degree of correlation of user and knowledge point.User and knowledge
The degree of correlation of the correspondence attribute of point can pass through Semantic Similarity Measurement.
The knowledge degree of correlation formula are as follows:
N is the knowledge point degree of correlation；
KyiIt is the weight of the i-th attribute；
Kc (i) is the i-th attribute of knowledge point；
User (i) is the i-th attribute of user；
Yu [kc (i), user (i)] refers to the semantic similarity of kc (i) He user (i).
By the knowledge degree of correlation, find with the maximally related knowledge document of user, be pushed to family members, family members Geng You section can be made
It learns, targetedly carry out family members' assisted care.
The present invention solves the problems, such as that psychological health states can not be tracked in time in the prior art, and feelings can be passed through by providing one kind
The mental health method for early warning for tracking and alerting in time is realized in sense analysis.
Embodiment 2
As shown in figure 4, other side according to an embodiment of the present invention, provides a kind of mental health method for early warning, institute
The method of stating includes:
S210: the input state or browse state of user terminal are detected；
S220: the content for inputting or browsing is obtained；
S230: sentiment analysis is carried out to the content；
S240: sending server for the result for carrying out sentiment analysis to the content, by the server judgement
And early warning is issued according to the level in mental health of user.
In a kind of exemplary embodiment, the user terminal operations system is that android is obtained when the completion inputs
Take inputted content step to implement are as follows: InputMethodService function: the function starts for the first time in input method
When it is called, for doing the setting initialized；By calling onBindlnput interface function, in other client and
Input method connection；It calls InputMethodManager module as input method manager, manages the interaction of each section；Pass through
OnFinishlnputO function is used to obtain current end of input；Input method calls onDestroy () function when closing.
As shown in figure 5, in the specific implementation process, carrying out sentiment analysis step to the content includes:
S231: judge that the type of the content, the type include text, picture or voice；
S232: corresponding sentiment analysis mode is selected according to the type of the content, the sentiment analysis mode includes text
This analysis, picture analyzing and speech analysis.
Specifically, server judges that the type of the content, the type include text, picture or voice；Server root
Corresponding sentiment analysis mode is selected according to the type of the content, the sentiment analysis mode includes text analyzing, picture analyzing
And speech analysis.
Specifically, if the type is picture, server selects Image emotional semantic analytical technology to carry out the content
Affection recognition of image.Image emotional semantic analysis is to analyze and extract affective characteristics from image, use pattern identification and machine learning
Method calculating is executed to it, and then understand the emotion of people.Main mode identification technology includes: template in Image emotional semantic analysis
Match pattern identification, the pattern-recognition of Fuzzy Pattern Recognition, support vector machines, and the deep learning based on artificial neural network.
The specific method of Image emotional semantic analysis is the prior art, and this is not described in detail here.
Specifically, if the type is voice, server selects speech emotional analytical technology to carry out the content
Affection recognition of image.Speech emotional analysis is that voice signal is analyzed and handled, and obtains the affective state that people is in.It is main
The speech emotion recognition algorithm wanted includes gauss hybrid models, support vector machines, K arest neighbors, hidden Markov model, sonograph
+ convolution loop neural network, manual feature+convolution loop neural network.The specific method of speech emotional analysis is the prior art,
This is not described in detail here.
Some user's preferences select corresponding sentiment analysis mode in input voice or picture, according to the type of content, can
Such as to be equally able to carry out sentiment analysis when picture, voice content when user inputs rich-media content.
In the specific implementation process, according to the type of the content select corresponding sentiment analysis mode step include: as
Content described in fruit is text, then analyzes the emotion of the content.
Specifically, if the content is text, the emotion of content described in server analysis.Realize text emotion analysis
Method and algorithm include rule-based, automatic system and hybrid system.Rule-based method defines one group by script
Rule, for identification subjectivity, polarity or opinion main body.Various inputs can be used in rule.For example, classical NLP technology, such as
Stem, symbol, part-of-speech tagging and parsing.In addition, rule can also use dictionary (i.e. word and expression way list).Based on rule
The key step of algorithm then include: define two polarization word lists (for example, the negation words such as poor, worst, ugly become reconciled, most preferably,
The fronts such as beauty word)；Calculate the positive word number occurred in text in the content.Calculate the negative word number occurred in text.Such as
The quantity that fruit front occurs is greater than the quantity that negative word occurs and then returns to positive mood, on the contrary, returning to negative emotions.Otherwise,
It returns neutral.Automatic method depends on machine learning.Sentiment analysis task is usually modeled as classification problem, the content
Then text input returns to corresponding classification to classifier, for example, just, bearing or neutral (if carrying out polarity check).
In a kind of exemplary embodiment, if the content is text, the algorithm packet of the emotion of the content is analyzed
Include training unit and analytical unit.The realization of training unit are as follows: use development set；With machine learning classification algorithm training the inside
Training set obtains machine learning model classifier；Classified with machine learning model classifier to development set, finally obtains text point
The result of class；Manual intervention is carried out using corpus, data mark is carried out to text, provide the accurate of Machine learning classifiers
Accuracy；Obtain algorithm and characteristic dimension；It obtains test set, the machine learning of foundation is tested.The realization of analytical unit
Are as follows: calling trains Bayes model；Save final model；Load final Bayesian model；Segment and go stop words
Operation；Read in active text and passive text；What is called is Bayes model training method；It calls in Sentiment class
Handle method；Call the classify method in Bayes class；Call the classify method in Bayes.
As shown in fig. 6, in the specific implementation process, it is described that emotion is carried out to the content if the content is text
Analytical procedure includes:
S233: judging that the character types of the content, the character types include Chinese, one in English and other characters
Kind or a variety of combinations；
S234: the sentiment analysis tool of the corresponding character types of selection analyze the emotion of the content.
Specifically, server judges that the character types of the content, the character types include Chinese, English and other words
One of symbol or a variety of combinations；The sentiment analysis tool of the corresponding character types of server selection analyze in described
The emotion of appearance.If the character types are English, server selects English text sentiment analysis tool, including Natural
Language Toolkit (NLTK), scikit-learn, SpaCy, Textacy, Tensorflow, Theano, fastText,
TextBlob.If if the character types are Chinese, server selects Chinese text sentiment analysis tool, including
SnowNLP, BosonNLP, Tencent's AI sentiment analysis.
By judging that character types select corresponding sentiment analysis tool, the use that can be applicable under multilingual environment.
As shown in fig. 7, in the specific implementation process, the sentiment analysis tool of the corresponding character types of selection carries out
The emotion step for analyzing the content includes:
S234a: the character types in the content are counted；
S234b: if including two or more language in the character types, it is corresponding that each character types are counted
Number of characters；
S234c: the ratio of the total text character number of number of characters Zhan of each character types is calculated；
S234d: the sentiment analysis result of the content is calculated according to the accounting of each character types.
Specifically, the sentiment analysis tool of the corresponding character types of the selection analyze the emotion step of the content
It suddenly include: the character types in content described in server statistics；If in the server character types including two kinds or two kinds
Above language then counts the corresponding number of characters of each character types；Server calculates the total text of number of characters Zhan of each character types
The ratio of number of characters；Server calculates the sentiment analysis result of the content according to the accounting of each character types.
In the actual environment, some users like coming in same sentence using the sentence that multilingual, such as Chinese and English mix
Expression.It is mixed that how various language can be efficiently solved according to the sentiment analysis result that the accounting of each character types calculates the content
The miscellaneous sentiment analysis problem in same sentence.
In the specific implementation process, the accounting according to each character types calculates the sentiment analysis result step of the content
Suddenly include:
Calculate the sentiment analysis result formula of the content are as follows:
Wherein, PLiFor the sentiment analysis result of i-th kind of character；
KLiFor the ratio of the total text character number of i-th kind of number of characters Zhan
P is the sentiment analysis result of the content.
Specifically, number of the P between 0-1, more levels off to 1, then more positive, more approaches 0, then more passive.
Sentiment analysis is carried out in the user terminal, can be sent to server to avoid by the content in user terminal, be given up
The computing capability of sentiment analysis is distributed to user terminal for revealing the doubt of privacy by user, reduces the performance of server
Pressure.
In the specific implementation process, described the step of sentiment analysis is carried out to the content further include:
Obtained the level in mental health of a upper period；
Data were set according to the level in mental health of a upper period and analyze frequency, the data analysis frequency is described works as
The number of sentiment analysis and the ratio of user terminal input number are carried out in the preceding period；
Frequency is analyzed according to the data, and sentiment analysis is carried out to the content.
Specifically, the step of sentiment analysis is carried out to content realization are as follows: user terminal obtained a upper period
Level in mental health；User terminal was arranged data according to the level in mental health of a upper period and analyzes frequency, the data
Analyzing frequency is that the ratio of the number and user terminal input number of sentiment analysis is carried out in the current slot；User terminal
Frequency is analyzed according to the data, and sentiment analysis is carried out to the content.The data analysis frequency is bigger, then to user terminal
The analysis times of input are bigger.When 1 when data analysis frequency reaching maximum value, user terminal inputs every time can be by carry out emotion
Analysis.
Specifically, the number formula of sentiment analysis is in this period
UT=FT×IT
UTFor the number of sentiment analysis, i.e., the number of sentiment analysis is carried out in this period；
ITFor this period the number of user input；
FTAnalyzing frequency for data is that the number that sentiment analysis is carried out in current slot and user terminal input number
Ratio, its calculation formula is:
Wherein, FT-1Frequency is analyzed for previous time period data；
β is level in mental health value.
Specifically, F0 is set as 1, i.e., when user inputs every time or browses webpage, user terminal all can be to being inputted in
The webpage perhaps browsed carries out sentiment analysis and is synchronized to server.
According to level in mental health, different data analysis frequencies or data analysis times are set, β level off to 1 when, then
Psychology is more passive, and data analysis frequency is more frequent, increases monitoring dynamics, facilitates the problem of finding terminal user in time.It is small in β
When 0, psychology is more positive, reduces Monitoring frequency, advantageously reduces the energy consumption of user terminal, reduces to user terminal
The consumption of energy.
The present invention solves the problems, such as that psychological health states can not be tracked in time in the prior art, and feelings can be passed through by providing one kind
The mental health method for early warning for tracking and alerting in time is realized in sense analysis.
Embodiment 3
As shown in figure 8, another aspect according to an embodiment of the present invention, provides a kind of mental health method for early warning, institute
The method of stating includes:
S310: server is received according to the issued warning information of level in mental health；
S320: early warning interface is generated according to the warning information, the early warning interface passes through text, image, audio or view
One or more combinations of frequency show warning information.
Specifically, doctor terminal receives server according to the issued warning information of level in mental health；Doctor terminal according to
The warning information generates early warning interface, one or more groups that the early warning interface passes through text, image, audio or video
It closes and shows warning information.
Specifically, family members' terminal receives server according to the issued warning information of level in mental health；Family members' terminal according to
The warning information generates early warning interface, one or more groups that the early warning interface passes through text, image, audio or video
It closes and shows warning information.
As shown in figure 9, in the specific implementation process, the early warning interface includes first information control, described according to
After warning information generates early warning interface step further include:
S330: the event in detecting and early warning interface；
S340: the selection event in response to being directed to the first information control shows User Detail, and the user is detailed
Thin information includes history level in mental health, current level in mental health.
Specifically, the event in doctor terminal detecting and early warning interface；Doctor terminal is in response to being directed to the first information control
The selection event of part shows User Detail, and the User Detail includes history level in mental health, currently psychology is strong
Kang Shuiping.
Specifically, the event in family members' terminal detecting and early warning interface；Family members' terminal is in response to being directed to the first information control
The selection event of part shows User Detail, and the User Detail includes history level in mental health, currently psychology is strong
Kang Shuiping.
As shown in Figure 10, in the specific implementation process, the early warning interface includes the second information control, described according to
After warning information generates early warning interface step further include:
S330: the event in detecting and early warning interface；
S350: the selection event in response to being directed to second information control shows the knowledge list of family members' assisted care, institute
It states family members' assisted care knowledge list and shows the family members' assisted care knowledge point being adapted with current level in mental health.
Specifically, the event in family members' terminal detecting and early warning interface；Family members' terminal is in response to being directed to the second information control
The selection event of part shows that the knowledge list of family members' assisted care, family members' assisted care knowledge list are shown and current psychology
The adaptable family members' assisted care knowledge point of the general level of the health.
In the specific implementation process, after the displaying family members assisted care knowledge listings step further include:
Detect the event in the knowledge list of family members' assisted care；
In response to the selection event for the knowledge point in family members' assisted care knowledge list, show that corresponding family members are auxiliary
Help nursing knowledge point.
Specifically, the event in family members' terminal detecting family members' assisted care knowledge list；Family members' terminal is in response to being directed to institute
The selection event of the knowledge point in family members' assisted care knowledge list is stated, shows corresponding family members' assisted care knowledge point.
In a kind of exemplary embodiment, the mental health method for early warning can be realized are as follows:
The input state or browse state of user terminal detecting user terminal；
User terminal obtains the content for inputting or browsing；Wherein, the step of obtaining inputted content realization are as follows: defeated
Enter when method starts for the first time and call InputMethodService function, does the setting initialized；Pass through calling
OnBindlnput interface function is connected with the input method；Call InputMethodManager module as input method manager,
Manage the interaction of each section；It is used to obtain current end of input by onFinishlnputO function；Input method close when
Call onDestroy () function；
User terminal user terminal carries out sentiment analysis to the content；Wherein, user terminal carries out feelings to the content
Feeling analytical procedure includes: to judge the type of the content, and the type includes text, picture or voice；According to the content
Type selects corresponding sentiment analysis mode, and the sentiment analysis mode includes text analyzing, picture analyzing and speech analysis；Such as
Type described in fruit is text, then judges that the character types of the content, the character types include Chinese, English and other characters
One of or a variety of combinations；The sentiment analysis tool of the corresponding character types of selection analyze the feelings of the content
Sense；Count the character types in the content；If including two or more language in the character types, count
The corresponding number of characters of each character types；Calculate the ratio of the total text character number of number of characters Zhan of each character types；According to each character
The accounting of type calculates the sentiment analysis result of the content；
The result for carrying out sentiment analysis to the content is sent server by user terminal；
Server judges level in mental health according to sentiment analysis result；It is described strong according to sentiment analysis result judgement psychology
The step of Kang Shuiping further include: statistical history sentiment analysis result；According to history sentiment analysis result and current sentiment analysis knot
The diversity judgement level in mental health of fruit；The history sentiment analysis result may include the history sentiment analysis knot of different cycles
Fruit；
Server issues warning information to doctor terminal according to level in mental health；
Doctor terminal receives server according to the issued warning information of level in mental health；
Doctor terminal according to the warning information generate early warning interface, the early warning interface by text, image, audio or
One or more combinations of video show warning information；
Event in doctor terminal detecting and early warning interface, the early warning interface include first information control；
Doctor terminal shows User Detail, the use in response to the selection event for the first information control
Family details include history level in mental health, current level in mental health；
Server inquires the corresponding family members of the user and family members and the customer relationship according to the user information；
Server issues warning information to family members' terminal according to level in mental health；
Family members' terminal receives server according to the issued warning information of level in mental health；
Family members' terminal according to the warning information generate early warning interface, the early warning interface by text, image, audio or
One or more combinations of video show warning information；
Event in doctor terminal detecting and early warning interface, the early warning interface include first information control；
Doctor terminal shows User Detail, the use in response to the selection event for the first information control
Family details include history level in mental health, current level in mental health；
Server inquires matched family members' assisted care knowledge point from family members' supplementary knowledge library, and family members' auxiliary is known
Know library and stores the relevant knowledge point of help family members' assisted care；Wherein, the age of the knowledge degree of correlation and user, gender,
Level in mental health, mental disease type and family members are related to the relationship of user；Knowledge point is set in family members' supplementary knowledge library
Attribute, the attribute includes the pass of age of user, gender, level in mental health, mental disease type and family members and user
System；The degree of correlation of user with the corresponding attribute of knowledge point are calculated separately, and all properties are added, is i.e. user and knowledge point knows
Know the degree of correlation；User can be by Semantic Similarity Measurement from family members' supplementary knowledge library with the degree of correlation of the corresponding attribute of knowledge point
Inquire matched family members' assisted care knowledge point step further include: the knowledge point in calculating family members' supplementary knowledge library and the user
The knowledge degree of correlation；N knowledge points before being selected from big to small according to the knowledge degree of correlation；The knowledge point and its knowledge is related
Degree is sent to doctor terminal, to show the knowledge point according to the knowledge degree of correlation by the doctor terminal；
Family members' assisted care knowledge point is sent the corresponding family members of the user by server；
The early warning interface further includes the second information control, and family members' terminal is in response to the choosing for second information control
Event is selected, shows that the knowledge list of family members' assisted care, family members' assisted care knowledge list are shown and current mental health water
Put down adaptable family members' assisted care knowledge point；
Family members' terminal detects the event in the knowledge list of family members' assisted care；
Family members' terminal is in response to the selection event for the knowledge point in family members' assisted care knowledge list, displaying pair
Answer family members' assisted care knowledge point
The present invention solves the problems, such as that psychological health states can not be tracked in time in the prior art, and feelings can be passed through by providing one kind
The mental health method for early warning for tracking and alerting in time is realized in sense analysis.
Embodiment 4
As shown in figure 11, according to an aspect of an embodiment of the present invention, a kind of mental health Warning Service device is provided
100, the server 100 includes:
First receiving module 101, for receiving to the content progress sentiment analysis of user terminal as a result, the content is
The interior current browsed web content of perhaps user inputted in user terminal acquisition user's input method；
First judgment module 102, for judging level in mental health according to the sentiment analysis result；
Warning module 103, for issuing warning information according to level in mental health.
In the specific implementation process, server is received to the content progress sentiment analysis of user terminal as a result, in described
Hold is that user terminal acquires the current browsed web content server of interior perhaps user inputted in user's input method according to emotion point
Analysis result judges level in mental health；Server issues warning information to doctor terminal according to level in mental health.Server root
Warning information is issued to family members' terminal according to level in mental health.
Actively acquire user inputted in the user terminal that it is held in perhaps browsing content mode, then to described interior
Hold and carry out sentiment analysis, the level in mental health of user is judged according to sentiment analysis result, is asked not needing user at hospital
It, can be with the real-time tracking user psychology general level of the health in the case where examining.
Due to by the way of active analysis user input content or browsing content, solving the one-to-one artificial interrogation of tradition
Just will appreciate that the user psychology problem in short supply so as to cause psychological medical resource, so as to so that shrink one-to-manyly
Service.
The doctor terminal that server is held Xiang doctor according to level in mental health issues warning information, can make doctor and
When psychological monitoring and intervention are carried out to user, be conducive to the negative emotions for alleviating user.
Server issues warning information according to family members' terminal that level in mental health is held to family members, can be timely with family members
Psychological adjuvant treatment is carried out, the negative emotions for alleviating user are conducive to.
In the specific implementation process, the server 100 further include:
First statistical module 104 is used for statistical history sentiment analysis result；
The first judgment module 102 is also used to the difference according to history sentiment analysis result and current sentiment analysis result
Different judgement level in mental health.
Specifically, the history sentiment analysis result may include the history sentiment analysis of different cycles as a result, as previous
It, the last week, preceding January, the preceding first quarter, the previous year or the history sentiment analysis result since use.
In the specific implementation process, the server 100 further include:
Enquiry module 105, for being inquired from family members' supplementary knowledge library according to the user information and level in mental health
To matched family members' assisted care knowledge point, family members' supplementary knowledge library, which stores, helps the relevant knowledge of family members' assisted care
Point；
First sending module 106, for sending the corresponding family members of the user for family members' assisted care knowledge point.
In the specific implementation process, the server 100 further include:
The enquiry module 105 is also used to inquire the corresponding family members of the user and family members and the customer relationship；
The enquiry module 105 is also used to according to the relationship of the family members and the user from family members' supplementary knowledge library
Inquire matched family members' assisted care knowledge point.
In the specific implementation process, the server 100 further include:
First computing module 107, for calculating the knowledge point in family members' supplementary knowledge library and the knowledge degree of correlation of the user；
The selecting module 105 is also used to N before selecting from big to small according to knowledge degree of correlation knowledge points；
First sending module 106 is also used to send family members' terminal for the knowledge point and its knowledge degree of correlation, with
The knowledge point is shown according to the knowledge degree of correlation by family members' terminal.
In the specific implementation process, the age of the knowledge degree of correlation and user, gender, level in mental health, psychological disease
Sick type and family members are related to the relationship of user.The attribute of knowledge point is set in family members' supplementary knowledge library, and the attribute includes
Age of user, gender, level in mental health, the relationship of mental disease type and family members and user.It calculates separately user and knows
Know the degree of correlation of the correspondence attribute of point, and all properties are added, is i.e. the knowledge degree of correlation of user and knowledge point.User and knowledge
The degree of correlation of the correspondence attribute of point can pass through Semantic Similarity Measurement.
By the knowledge degree of correlation, find with the immediate document of user, be pushed to family members, family members can be made more have science,
Targetedly carry out family members' assisted care.
The present invention solves the problems, such as that psychological health states can not be tracked in time in the prior art, and feelings can be passed through by providing one kind
The mental health method for early warning for tracking and alerting in time is realized in sense analysis.
Embodiment 5
As shown in figure 12, it is whole to provide a kind of mental health early warning user for other side according to an embodiment of the present invention
End 200, the user terminal 200 include:
Second detecting module 201, for detecting the input state or browse state of user terminal；
Module 202 is obtained, for obtaining the content for inputting or browsing；
Analysis module 203, for carrying out sentiment analysis to the content；
Second sending module 204, for sending server for the result for carrying out sentiment analysis to the content, by institute
The server stated judges and issues early warning according to the level in mental health of user.
In a kind of exemplary embodiment, the user terminal operations system is that android is obtained when the completion inputs
Take inputted content step to implement are as follows: InputMethodService function: the function starts for the first time in input method
When it is called, for doing the setting initialized；By calling onBindlnput interface function, in other client and
Input method connection；It calls InputMethodManager module as input method manager, manages the interaction of each section；Pass through
OnFinishlnputO function is used to obtain current end of input；Input method calls onDestroy () function when closing.
As shown in Fig. 2, in the specific implementation process, the user terminal 200 includes:
Second judgment module 205, for judging that the type of the content, the type include text, picture or voice；
Selecting module 206, for selecting corresponding sentiment analysis mode, the sentiment analysis according to the type of the content
Mode includes text analyzing, picture analyzing and speech analysis.
As shown in figure 3, in the specific implementation process, the user terminal 200 includes:
Second judgment module 205, is also used to judge the character types of the content, during the character types include
Text, one of English and other characters or a variety of combinations；
The selecting module 206, be also used to select to correspond to the character types sentiment analysis tool analyzed described in
The emotion of content.
Specifically, server judges that the character types of the content, the character types include Chinese, English and other words
One of symbol or a variety of combinations；The sentiment analysis tool of the corresponding character types of server selection analyze in described
The emotion of appearance.If the character types are English, server selects English text sentiment analysis tool, including Natural
Language Toolkit (NLTK), scikit-learn, SpaCy, Textacy, Tensorflow, Theano, fastText,
TextBlob.If if the character types are Chinese, server selects Chinese text sentiment analysis tool, including
SnowNLP, BosonNLP, Tencent's AI sentiment analysis.
By judging that character types select corresponding sentiment analysis tool, the use that can be applicable under multilingual environment.
In the specific implementation process, the user terminal 200 further include:
Second statistical module 207, for counting the character types in the content；
Second statistical module 207, if being also used to the language in the character types comprising two or more,
Then count the corresponding number of characters of each character types；
Second computing module 208, the ratio of the total text character number of number of characters Zhan for calculating each character types；
Second computing module 208 is also used to calculate the sentiment analysis of the content according to the accounting of each character types
As a result.
In the specific implementation process, the user terminal 200 further include:
The acquisition module 202, for obtaining the level in mental health of a upper period；
Setup module 209 analyzes frequency, the data for data to be arranged according to the level in mental health of a upper period
Analyzing frequency is that the ratio of the number and user terminal input number of sentiment analysis is carried out in the current slot；
The analysis module 203 is also used to analyze frequency according to the data to content progress sentiment analysis.
Specifically, the step of sentiment analysis is carried out to content realization are as follows: user terminal obtained a upper period
Level in mental health；User terminal was arranged data according to the level in mental health of a upper period and analyzes frequency, the data
Analyzing frequency is that the ratio of the number and user terminal input number of sentiment analysis is carried out in the current slot；User terminal
Frequency is analyzed according to the data, and sentiment analysis is carried out to the content.The data analysis frequency is bigger, then to user terminal
The analysis times of input are bigger.When 1 when data analysis frequency reaching maximum value, user terminal inputs every time can be by carry out emotion
Analysis.
According to level in mental health, different data analysis frequencies or data analysis times are set, β level off to 1 when, then
Psychology is more passive, and data analysis frequency is more frequent, increases monitoring dynamics, facilitates the problem of finding terminal user in time.It is small in β
When 0, psychology is more positive, reduces Monitoring frequency, advantageously reduces the energy consumption of user terminal, reduces to user terminal
The consumption of energy.
The present invention solves the problems, such as that psychological health states can not be tracked in time in the prior art, and feelings can be passed through by providing one kind
The mental health early warning user terminal for tracking and alerting in time is realized in sense analysis.
Embodiment 6
As shown in figure 13, it is whole to provide a kind of mental health early warning doctor for another aspect according to an embodiment of the present invention
End 300, the doctor terminal 300 include:
Third receiving module 301, for receiving server according to the issued warning information of level in mental health；
Third generation module 302, for generating early warning interface according to the warning information, the early warning interface passes through text
One or more combinations of word, image, audio or video show warning information.
In the specific implementation process, the doctor terminal 300 further include:
Third detecting module 303, for the event in detecting and early warning interface；
Third display module 304 shows that user is detailed for the selection event in response to being directed to the first information control
Information, the User Detail include history level in mental health, current level in mental health.
In the specific implementation process, the doctor terminal 300 further include:
The third display module 304 is also used to the selection event in response to being directed to second information control, shows house
Belong to the list of assisted care knowledge, family members' assisted care knowledge list shows the family members being adapted with current level in mental health
Assisted care knowledge point.
In the specific implementation process, the doctor terminal 300 further include:
The third detecting module 303 is also used to detect the event in family members' assisted care knowledge list；
The third display module 304 is also used in response to for the knowledge in family members' assisted care knowledge list
The selection event of point shows corresponding family members' assisted care knowledge point.
The present invention solves the problems, such as that psychological health states can not be tracked in time in the prior art, and feelings can be passed through by providing one kind
The mental health early warning doctor terminal for tracking and alerting in time is realized in sense analysis.
Embodiment 7
As shown in figure 14, it is whole to provide a kind of mental health early warning family members for another aspect according to an embodiment of the present invention
End 400, family members' terminal 400 include:
4th receiving module 401, for receiving server according to the issued warning information of level in mental health；
4th generation module 402, for generating early warning interface according to the warning information, the early warning interface passes through text
One or more combinations of word, image, audio or video show warning information.
In the specific implementation process, family members' terminal 300 further include:
4th detecting module 403, for the event in detecting and early warning interface；
4th display module 404 shows that user is detailed for the selection event in response to being directed to the first information control
Information, the User Detail include history level in mental health, current level in mental health.
In the specific implementation process, family members' terminal 400 further include:
4th display module 404 is also used to the selection event in response to being directed to second information control, shows house
Belong to the list of assisted care knowledge, family members' assisted care knowledge list shows the family members being adapted with current level in mental health
Assisted care knowledge point.
In the specific implementation process, family members' terminal 400 further include:
4th detecting module 403 is also used to detect the event in family members' assisted care knowledge list；
4th display module 404 is also used in response to for the knowledge in family members' assisted care knowledge list
The selection event of point shows corresponding family members' assisted care knowledge point.
The present invention solves the problems, such as that psychological health states can not be tracked in time in the prior art, and feelings can be passed through by providing one kind
The mental health early warning doctor terminal for tracking and alerting in time is realized in sense analysis.
Embodiment 8
As shown in figure 15, another aspect according to an embodiment of the present invention provides a kind of mental health early warning system, institute
The system of stating includes mental health Warning Service device 100, user terminal 200, doctor terminal 300 and family members' terminal 400.The user
Terminal 200 and the server 100 communicate to connect.The doctor terminal 300 is communicated to connect with the server 100.The family
Belong to terminal 400 and the server 100 communicates to connect.
The server 100 includes that memory and the processor for being coupled to the memory, processor are configured as being based on
Instruction stored in memory executes the mental health method for early warning in the disclosure in specific embodiment 1.Memory for example may be used
To include system storage, fixed non-volatile memory medium etc..System storage is for example stored with operating system, using journey
Sequence, Boot loader (Boot Loader), database and other programs etc..
The user terminal 200 includes that memory and the processor for being coupled to the memory, processor are configured as base
In instruction stored in memory, the mental health method for early warning in disclosure specific embodiment 2 is executed.Memory for example may be used
To include system storage, fixed non-volatile memory medium etc..System storage is for example stored with operating system, using journey
Sequence, Boot loader (Boot Loader), database and other programs etc..
The doctor terminal 300 or family members' terminal 400 include memory and the processor for being coupled to the memory, processing
Device is configured as executing the mental health method for early warning in disclosure specific embodiment 3 based on instruction stored in memory.
Memory for example may include system storage, fixed non-volatile memory medium etc..System storage is for example stored with operation
System, application program, Boot loader (Boot Loader), database and other programs etc..
The present invention solves the problems, such as that psychological health states can not be tracked in time in the prior art, and feelings can be passed through by providing one kind
The mental health early warning system for tracking and alerting in time is realized in sense analysis.
The disclosure also provides a kind of computer readable storage medium, is stored thereon with computer program, and the program is processed
The step of device realizes the mental health method for early warning of any one aforementioned embodiment when executing.
Those skilled in the art should be understood that embodiment of the disclosure can provide as method, system or computer journey
Sequence product.Therefore, complete hardware embodiment, complete software embodiment or combining software and hardware aspects can be used in the disclosure
The form of embodiment.Moreover, it wherein includes the calculating of computer usable program code that the disclosure, which can be used in one or more,
Machine can use the meter implemented in non-transient storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.)
The form of calculation machine program product.
So far, be described in detail according to the mental health method for early warning, user terminal, server of the disclosure be
System.In order to avoid covering the design of the disclosure, some details known in the field are not described.Those skilled in the art according to
Above description, completely it can be appreciated how implementing technical solution disclosed herein.
Obviously, the above embodiment of the present invention be only to clearly illustrate example of the present invention, and not be pair
The restriction of embodiments of the present invention.For those of ordinary skill in the art, may be used also on the basis of the above description
To make other variations or changes in different ways.There is no necessity and possibility to exhaust all the enbodiments.It is all this
Made any modifications, equivalent replacements, and improvements etc., should be included in the claims in the present invention within the spirit and principle of invention
Protection scope within.
Claims (10)
1. a kind of mental health method for early warning, which is characterized in that the described method includes:
Detect the input state or browse state of user terminal；
Obtain the content for inputting or browsing；
Sentiment analysis is carried out to the content；
Server is sent by the result for carrying out sentiment analysis to the content, by the server judgement and according to user
Level in mental health issue early warning.
2. mental health method for early warning according to claim 1, which is characterized in that described to carry out emotion point to the content
The step of analysis further include:
Obtained the level in mental health of a upper period；
Data were set according to the level in mental health of a upper period and analyze frequency, when the data analysis frequency is described current
Between the number of sentiment analysis is carried out in section and user terminal inputs the ratio of number；
Frequency is analyzed according to the data, and sentiment analysis is carried out to the content.
3. mental health method for early warning according to claim 2, which is characterized in that carry out sentiment analysis step to the content
Suddenly include:
Judge that the type of the content, the type include text, picture or voice；
Corresponding sentiment analysis mode is selected according to the type of the content, the sentiment analysis mode includes text analyzing, figure
Piece analysis and speech analysis.
4. mental health method for early warning according to claim 3, which is characterized in that described if the content is text
Carrying out sentiment analysis step to the content includes:
Judge that the character types of the content, the character types include Chinese, one of English and other characters or a variety of
Combination；
The sentiment analysis tool of the corresponding character types of selection analyze the emotion of the content.
5. a kind of mental health method for early warning, which is characterized in that the described method includes:
It receives to the content progress sentiment analysis of user terminal as a result, the content is that user terminal acquires in user's input method
The interior current browsed web content of perhaps user of input；
Level in mental health is judged according to the sentiment analysis result；
Warning information is issued according to level in mental health.
6. mental health method for early warning according to claim 5, which is characterized in that described to be judged according to sentiment analysis result
The step of level in mental health further include:
Statistical history sentiment analysis result；
According to the diversity judgement level in mental health of history sentiment analysis result and current sentiment analysis result.
7. a kind of mental health early warning user terminal, which is characterized in that the user terminal further include:
Second detecting module, for detecting the input state or browse state of user terminal；
Module is obtained, for obtaining the content for inputting or browsing；
Analysis module, for carrying out sentiment analysis to the content；
Second sending module, for sending server for the result for carrying out sentiment analysis to the content, by the clothes
The judgement of business device simultaneously issues early warning according to the level in mental health of user.
8. a kind of mental health Warning Service device, which is characterized in that the server includes:
Receiving module, for receiving the content of user terminal, the content is to input in user terminal acquisition user's input method
The inside perhaps current browsed web content of user；
Analysis module, for carrying out sentiment analysis to the content；
Judgment module, for judging level in mental health according to sentiment analysis result；
Warning module, for issuing warning information to doctor terminal according to level in mental health.
9. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the program is held by processor
Mental health method for early warning described in one of claim 1 to 6 is realized when row.
10. a kind of mental health early warning system, which is characterized in that the system comprises:
User terminal, server according to claim 8 and family members' terminal according to claim 7；The user
Terminal is connect with the server communication；Family members' terminal is connect with the server communication；
Wherein, family members' terminal includes:
4th receiving module, for receiving server according to the issued warning information of level in mental health；
4th generation module, for according to the warning information generate early warning interface, the early warning interface by text, image,
One or more combinations of audio or video show warning information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910231911.8A CN109979592A (en) | 2019-03-25 | 2019-03-25 | Mental health method for early warning, user terminal, server and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910231911.8A CN109979592A (en) | 2019-03-25 | 2019-03-25 | Mental health method for early warning, user terminal, server and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109979592A true CN109979592A (en) | 2019-07-05 |
Family
ID=67080606
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910231911.8A Pending CN109979592A (en) | 2019-03-25 | 2019-03-25 | Mental health method for early warning, user terminal, server and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109979592A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111370113A (en) * | 2020-03-03 | 2020-07-03 | 淮安信息职业技术学院 | Remote psychological counseling system and method based on Internet of things cloud |
| CN112071386A (en) * | 2020-08-19 | 2020-12-11 | 深圳市医贝科技有限公司 | Active mental health early warning method and system |
| CN113055746A (en) * | 2021-04-19 | 2021-06-29 | 广州欢网科技有限责任公司 | Old people monitoring method, device, server and system based on television watching behaviors |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2012160082A (en) * | 2011-02-01 | 2012-08-23 | Bond:Kk | Input support device, input support method, and program |
| CN103493040A (en) * | 2011-04-21 | 2014-01-01 | 索尼公司 | A method for determining a sentiment from a text |
| CN103955614A (en) * | 2014-04-29 | 2014-07-30 | 北京盛世光明软件股份有限公司 | Method and system for predicting psychological crisis |
| CN104794211A (en) * | 2015-04-24 | 2015-07-22 | 清华大学 | Method and system for extracting sentiment inducements and analyzing inducement elements based on microblog text |
| CN105740224A (en) * | 2014-12-11 | 2016-07-06 | 仲恺农业工程学院 | Text analysis based user psychology early warning method and apparatus |
| CN106708805A (en) * | 2016-12-30 | 2017-05-24 | 深圳天珑无线科技有限公司 | Text statistics-based psychoanalysis method and device |
| CN107024996A (en) * | 2017-06-06 | 2017-08-08 | 温州云航信息科技有限公司 | Emotive advisory equipment |
| CN107220293A (en) * | 2017-04-26 | 2017-09-29 | 天津大学 | File classification method based on mood |
| TW201807602A (en) * | 2016-08-24 | 2018-03-01 | 慧科訊業有限公司 | System, apparatus and method for monitoring internet media events based on a constructed industry knowledge graph database |
| CN107832301A (en) * | 2017-11-22 | 2018-03-23 | 北京百度网讯科技有限公司 | Participle processing method, device, mobile terminal and computer-readable recording medium |
| CN108717406A (en) * | 2018-05-10 | 2018-10-30 | 平安科技（深圳）有限公司 | Text mood analysis method, device and storage medium |
| TWM573465U (en) * | 2018-07-18 | 2019-01-21 | 太米股份有限公司 | Metadata generation system |
| CN109472207A (en) * | 2018-10-11 | 2019-03-15 | 平安科技（深圳）有限公司 | Emotion identification method, apparatus, equipment and storage medium |
-
2019
- 2019-03-25 CN CN201910231911.8A patent/CN109979592A/en active Pending
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2012160082A (en) * | 2011-02-01 | 2012-08-23 | Bond:Kk | Input support device, input support method, and program |
| CN103493040A (en) * | 2011-04-21 | 2014-01-01 | 索尼公司 | A method for determining a sentiment from a text |
| CN103955614A (en) * | 2014-04-29 | 2014-07-30 | 北京盛世光明软件股份有限公司 | Method and system for predicting psychological crisis |
| CN105740224A (en) * | 2014-12-11 | 2016-07-06 | 仲恺农业工程学院 | Text analysis based user psychology early warning method and apparatus |
| CN104794211A (en) * | 2015-04-24 | 2015-07-22 | 清华大学 | Method and system for extracting sentiment inducements and analyzing inducement elements based on microblog text |
| TW201807602A (en) * | 2016-08-24 | 2018-03-01 | 慧科訊業有限公司 | System, apparatus and method for monitoring internet media events based on a constructed industry knowledge graph database |
| CN106708805A (en) * | 2016-12-30 | 2017-05-24 | 深圳天珑无线科技有限公司 | Text statistics-based psychoanalysis method and device |
| CN107220293A (en) * | 2017-04-26 | 2017-09-29 | 天津大学 | File classification method based on mood |
| CN107024996A (en) * | 2017-06-06 | 2017-08-08 | 温州云航信息科技有限公司 | Emotive advisory equipment |
| CN107832301A (en) * | 2017-11-22 | 2018-03-23 | 北京百度网讯科技有限公司 | Participle processing method, device, mobile terminal and computer-readable recording medium |
| CN108717406A (en) * | 2018-05-10 | 2018-10-30 | 平安科技（深圳）有限公司 | Text mood analysis method, device and storage medium |
| TWM573465U (en) * | 2018-07-18 | 2019-01-21 | 太米股份有限公司 | Metadata generation system |
| CN109472207A (en) * | 2018-10-11 | 2019-03-15 | 平安科技（深圳）有限公司 | Emotion identification method, apparatus, equipment and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| WEIXIN_30546189: ""情感分析snownlp包部分核⼼代码理解"", 《CSDN论坛》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111370113A (en) * | 2020-03-03 | 2020-07-03 | 淮安信息职业技术学院 | Remote psychological counseling system and method based on Internet of things cloud |
| CN112071386A (en) * | 2020-08-19 | 2020-12-11 | 深圳市医贝科技有限公司 | Active mental health early warning method and system |
| CN113055746A (en) * | 2021-04-19 | 2021-06-29 | 广州欢网科技有限责任公司 | Old people monitoring method, device, server and system based on television watching behaviors |
| CN113055746B (en) * | 2021-04-19 | 2022-09-30 | 广州欢网科技有限责任公司 | Old people monitoring method, device, server and system based on television watching behaviors |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Yazdavar et al. | Semi-supervised approach to monitoring clinical depressive symptoms in social media | |
| Rosa et al. | A knowledge-based recommendation system that includes sentiment analysis and deep learning | |
| Nahin et al. | Identifying emotion by keystroke dynamics and text pattern analysis | |
| US11625615B2 (en) | Artificial intelligence advisory systems and methods for behavioral pattern matching and language generation | |
| Marín et al. | On generating linguistic descriptions of time series | |
| CN109979568A (en) | Mental health method for early warning, server, family members' terminal and system | |
| El Morr et al. | Descriptive, predictive, and prescriptive analytics | |
| CN109979592A (en) | Mental health method for early warning, user terminal, server and system | |
| Jain et al. | Tweet recommender model using adaptive neuro-fuzzy inference system | |
| Chan et al. | Question-answering dialogue system for emergency operations | |
| Ambica et al. | An efficient expert system for diabetes by naive Bayesian classifier | |
| Zhang et al. | Multi-modal interactive fusion method for detecting teenagers’ psychological stress | |
| Gkatzia | Content selection in data-to-text systems: A survey | |
| Pachouly et al. | Depression detection on social media network (Twitter) using sentiment analysis | |
| CN109994207A (en) | Mental health method for early warning, server and system | |
| Li | Synesthetic design of digital elderly products based on big data | |
| Wandabwa et al. | Multi-interest semantic changes over time in short-text microblogs | |
| Ramaswamy | Visualization of the Sentiment of the Tweets. | |
| CN111339252B (en) | Searching method, searching device and storage medium | |
| White | Using topic models to detect behaviour patterns for healthcare monitoring | |
| CN114496231A (en) | Constitution identification method, apparatus, equipment and storage medium based on knowledge graph | |
| Taghvaei et al. | Analytical framework for mental health feature extraction methods in social networks | |
| Luo et al. | Shyness prediction and language style model construction of elementary school students | |
| Wickramasinghe et al. | A Framework for Tracking And Summarizing Daily Activities with Mobile Phone for Healthy Life of a User | |
| Vijayakumar et al. | User's learning capability aware E-content recommendation system for enhanced learning experience |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190705 |
|
| RJ01 | Rejection of invention patent application after publication |