CN112463937A - Intelligent psychological service software system - Google Patents
Intelligent psychological service software system Download PDFInfo
- Publication number
- CN112463937A CN112463937A CN202011205048.8A CN202011205048A CN112463937A CN 112463937 A CN112463937 A CN 112463937A CN 202011205048 A CN202011205048 A CN 202011205048A CN 112463937 A CN112463937 A CN 112463937A
- Authority
- CN
- China
- Prior art keywords
- question
- user
- psychological
- content
- answer
- 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.)
- Granted
Links
Images
Classifications
-
- 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
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation or dialogue systems
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/16—Devices for psychotechnics; Testing reaction times ; Devices for evaluating the psychological state
- A61B5/165—Evaluating the state of mind, e.g. depression, anxiety
-
- 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
-
- 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
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Psychiatry (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Social Psychology (AREA)
- Psychology (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Child & Adolescent Psychology (AREA)
- Medical Informatics (AREA)
- Developmental Disabilities (AREA)
- Hospice & Palliative Care (AREA)
- Animal Behavior & Ethology (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Molecular Biology (AREA)
- Veterinary Medicine (AREA)
- Surgery (AREA)
- Educational Technology (AREA)
- Biophysics (AREA)
- Heart & Thoracic Surgery (AREA)
- Artificial Intelligence (AREA)
- Human Computer Interaction (AREA)
- Biomedical Technology (AREA)
- Pathology (AREA)
- Epidemiology (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides an intelligent psychological service software system, which relates to the field of psychological construction and comprises the following components: the accompany support module is used for realizing man-machine conversation with the user according to the short text question-answer data; the psychological intervention module is used for pushing preset psychological intervention contents related to the user question contents to the user according to the user question contents in the man-machine conversation process; the mental health screening module is used for carrying out psychological evaluation on the user to obtain an evaluation result and providing psychological guidance for the user according to the evaluation result; the growth training module is used for providing at least one mental training toolkit for the user to perform mental elasticity training; and the psychological file module is used for summarizing all data in the man-machine conversation process and establishing a personal psychological file of the user. According to the invention, services of accompany support, psychological intervention, psychological health screening, growth training and psychological archive establishment are provided for the user through the intelligent service software, so that interactive communication between the intelligent service software and the user is realized, and the psychological health is guaranteed.
Description
Technical Field
The invention relates to the field of psychological construction, in particular to an intelligent psychological service software system.
Background
The maritime special task team is deployed overseas for a long time, heavy workload and pressure and monotonous life on the sea easily cause the psychological problems of team members, so that the design requirements of the maritime special task team on the maintenance and excitation of fighting morale are particularly urgent. At present, the means for managing and leading the mental states of team members on the maritime special task team is single, and the technology for providing software service for the mental health of the team members by combining an artificial intelligence technology and a psychological research concept does not exist in the prior technical scheme for a while.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides an intelligent psychological service software system, which specifically comprises:
the accompany support module is used for realizing a man-machine conversation process with a user according to prestored short text question-answer data;
the psychological intervention module is used for pushing preset psychological intervention contents related to the user question contents to the user according to the user question contents in the man-machine conversation process;
the mental health screening module is used for carrying out psychological evaluation on the user to obtain an evaluation result and providing corresponding psychological guidance for the user according to the evaluation result;
the growth training module is used for providing at least one mental training toolkit for the user to perform mental elasticity training;
and the psychological file module is respectively connected with the accompany support module, the psychological intervention module, the psychological health screening module and the growth training module and is used for summarizing all data in the man-machine conversation process and establishing the personal psychological file of the user.
The big data collecting sub-module is used for collecting political education content data, and/or navy team member psychology big data, and/or question-answer big data and/or mental health big data as the short text question-answer data;
the big data management submodule is connected with the big data collecting submodule and is used for processing the short text question-answer data collected by the big data collecting unit;
and the intelligent question-answer module is connected with the big data management module and is used for realizing a man-machine conversation process with a user according to the processed short text question-answer data.
Preferably, the intelligent question-answering module adopts a preset analysis algorithm to realize a man-machine conversation process with the user, and the analysis algorithm comprises: a psychological problem classification algorithm, and/or an emotion analysis algorithm, and/or a keyword extraction algorithm and/or a question-answer association deep learning algorithm.
Preferably, the intelligent question answering module comprises:
the dialogue starting unit is used for outputting a preset comprehensive welcome language as a dialogue starting language when the user is judged to be in the first dialogue and outputting a welcome introduction language as a dialogue starting language when the user is judged not to be in the first dialogue;
the dialogue counting unit is connected with the dialogue starting unit and used for receiving and counting the current speaking times and the current speaking word number corresponding to the user questioning content of the user after the dialogue starting language is given and when the user is judged not to leave;
the intelligent question-answering unit is connected with the dialogue counting unit and used for giving a question answer corresponding to the user question content when the current speaking word number is not more than a word number threshold;
the professional content recommending unit is connected with the dialogue counting unit and used for giving out professional recommended content corresponding to the user question content when the current speaking word number is larger than the word number threshold;
and the user information collection unit is connected with the conversation statistical unit and used for collecting the user information when the current speaking times in a preset time period is greater than a threshold value of a number of times.
Preferably, the intelligent question and answer unit comprises:
the first judging subunit is used for giving a pre-configured question answer associated with the common word when the user question content is judged to be the common word, and giving a first judging result when the user question content is judged not to be the common word;
the first processing subunit is connected with the first judging subunit and used for extracting keywords from the user questioning content according to the first judging result, giving a first prompt message when the keywords are not extracted, and giving a first processing result when the keywords are extracted;
the second judgment subunit is connected with the first processing subunit and used for giving a second judgment result when the keyword is judged to be in a preset white list according to the first processing result and giving a third judgment result when the keyword is not in the white list;
the second processing subunit is connected with the second judging subunit and used for giving a second prompt message according to the second judgment result when the question answer associated with the keyword is not matched and giving the question answer when the question answer associated with the keyword is matched;
and the third processing subunit is connected with the second judging subunit and is used for giving a third prompt message when the question answer associated with the keyword is not matched according to the third judging result and giving a fourth prompt message and the question answer when the question answer associated with the keyword is matched.
Preferably, the user information collecting unit includes:
the collecting subunit is used for collecting user information according to a preset information research table, and the information research table comprises a plurality of research problems;
and the third judging subunit is connected with the collecting subunit and is used for outputting a welcome language to inquire whether the user wants to perform research or not when judging that all the research questions are not collected completely, and continuously collecting the research questions one by one when the user wants to perform research until all the research questions are collected completely.
Preferably, the psychological intervention module analyzes the user question content according to a pre-generated offline construction model, and pushes preset psychological intervention content related to the user question content to the user according to an analysis result.
Preferably, the offline construction model comprises a word vector learning database, a recommendation content database, a psychological problem knowledge map, a cure content knowledge map and a psychological question-answer knowledge map.
Preferably, the psychological intervention module comprises:
the keyword extraction unit is used for extracting keywords from the user question content to obtain question keywords;
and the content recommendation unit is connected with the keyword extraction unit and used for expanding the question keywords according to the psychological question knowledge graph to obtain question expansion keywords, processing the question expansion keywords according to the psychological question-answer knowledge graph to obtain answer keywords related to the question expansion keywords, expanding the answer keywords according to the cure content knowledge graph to obtain answer expansion keywords, performing semantic quantization on the answer expansion keywords according to the word vector learning database, and matching the semantic quantization results in the recommended content database to obtain the psychological intervention content to be pushed to the user.
The technical scheme has the following advantages or beneficial effects:
according to the technical scheme, the intelligent service software provides services of accompanying support, psychological intervention, psychological health screening, growth training and psychological archive establishment for the user, interactive communication between the intelligent service software and the user is achieved, the psychological problem of the user is effectively relieved, and the psychological health of the user is guaranteed.
Drawings
FIG. 1 is a schematic diagram of an intelligent psychological service software system according to a preferred embodiment of the invention.
Detailed Description
The invention is described in detail below with reference to the figures and specific embodiments. The present invention is not limited to the embodiment, and other embodiments may be included in the scope of the present invention as long as the gist of the present invention is satisfied.
In a preferred embodiment of the present invention, based on the above problems in the prior art, there is provided an intelligent psychological service software system, as shown in fig. 1, specifically including:
the accompany support module 1 is used for realizing a man-machine conversation process with a user according to prestored short text question-answer data;
the psychological intervention module 2 is used for pushing preset psychological intervention contents related to the user question contents to the user according to the user question contents in the man-machine conversation process;
the mental health screening module 3 is used for performing mental evaluation on the user to obtain an evaluation result and providing corresponding mental guidance for the user according to the evaluation result;
the growth training module 4 is used for providing at least one mental training toolkit for the user to perform mental elasticity training;
and the psychological file module 5 is respectively connected with the accompany support module 1, the psychological intervention module 2, the mental health screening module 3 and the growth training module 4, and is used for summarizing all data in the man-machine conversation process and establishing a personal psychological file of the user.
Specifically, in this embodiment, the user of the intelligent psychological service software system is a maritime task-specific team member. The accompanying support module 1 provides positive and logical emotion discourse and emotion accompanying service through one-to-one intelligent question and answer and infinite man-machine conversation aiming at the emotion, marriage, family, occupation planning, addiction and confusion of the team members based on massive short text consultation and answer data (mainly comprising common short text question and answer data, popular psychological short text question and answer data, team member psychological short text question and answer data and the like), effectively relieves the problems of the mental levels of the team members such as solitary, distress, depression, sensitivity, anxiety, depression and the like, prevents the problems from deteriorating, and guides the team members to walk out of a psychological error zone.
The psychological intervention module 2 can quickly capture the individualized demands of the team members according to the chat contents in the human-computer interaction process, intelligently push professional contents such as professional consulting jargon, science popularization articles, audio and video, meditation music and the like which are pre-arranged by psychological experts to different team members, and realize professional psychological intervention while meeting the individualized mental demands of the team members. The professional consultation talk starts from a specific problem of a team member (such as love marriage, family relationship, social interpersonal, workplace psychology, emotional stress, mental health and the like), carries out psychological consultation on the team member, is supplemented with a science popularization article, audio and video, meditation music and the like corresponding to psychology, relieves the emotion of the team member in man-machine interaction, obtains knowledge and help, and further diagnoses and dredges the psychological problem. The psychological intervention module 2 focuses on realizing the intelligent recommendation of big data of the robot, is realized based on deep learning and big data analysis technology, and provides psychological content recommendation service for team members. Extracting keywords from the man-machine interaction content by the content recommendation service, constructing a psychological knowledge map, finding out psychological adjusting keywords corresponding to the keywords from the map, performing synonym expansion on the psychological adjusting keywords, and constructing a synonym library; and searching and sequencing in the recommended content database based on the synonym library, and outputting the most targeted recommended content.
The mental health screening module 3 carefully selects various professional psychological evaluation and interesting psychological test contents suitable for the team members, and corresponding evaluation methods and scoring standards, so that the team members can be helped to deeply understand the psychological states of the team members, correct guidance and solution ways are provided for psychological confusion, the system can be helped to further obtain the personalized psychological needs of the team members, the psychological confusion and problems of the team members are made clear, and data support is provided for intelligent content recommendation, psychological counseling and extreme emotion early warning.
The growth training module 4 provides a plurality of psychological training kits including a new team member adaptation training kit, a love growth training kit, an anger management training kit, and other customized training kits suitable for the group specificity of team members. The toolkits are developed based on powerful and meticulous psychology background logic and abundant expert experience, have good actual training effects, and provide abundant and diverse training means for the psychological growth of players.
The psychological file module 5 builds a personal psychological file for the team members by summarizing and analyzing the man-machine conversation content, and realizes the summarization of relevant data including but not limited to a personal emotion score list, an emotion fluctuation curve, an evaluation test result, a team member self-evaluation result in the psychological intervention process, the behavior of whether the team member accepts or not and the like.
In a preferred embodiment of the present invention, the companion support module 1 includes:
the big data collecting submodule 11 is used for collecting political education content data and/or navy team member psychology big data and/or question and answer big data and/or mental health big data as short text question and answer data;
the big data management submodule 12 is connected with the big data collecting submodule 11 and is used for processing the short text question-answer data collected by the big data collecting unit;
and the intelligent question-answer module 13 is connected with the big data management module 12 and is used for realizing a man-machine conversation process with a naval crew according to the processed short text question-answer data.
Specifically, in the present embodiment, the big data gathering sub-module 11 gathers political education content data, and/or team member psychology big data, and/or question and answer big data, and/or mental health big data from a plurality of channels. The political education content data comprises political theory learning content and the like; the team member psychology big data comprises but is not limited to self-confidence, pressure response, will power, sleep problems, fighting psychology, interpersonal relationship, team cohesion and the like; the big question-answer data mainly comprises massive short text consultation question-answer data obtained from psychologists, Internet software and the like; the mental health big data mainly comprises public mental health data, and is automatically collected through a crawler system.
The big data management submodule 12 is implemented based on a full text search tool and a psychological security knowledge map content management tool, and the big data management submodule 12 performs operations such as editing, labeling and inquiring on data acquired from multiple channels and provides support for the intelligent question and answer submodule 13.
The intelligent question-answer module 13 mainly includes analysis algorithms such as a psychological question classification algorithm, an emotion analysis algorithm, a keyword extraction algorithm, a question-answer association deep learning algorithm and the like, and the intelligent question-answer module 13 provides psychological support for team members through the analysis algorithms.
In a preferred embodiment of the present invention, the intelligent question and answer module 13 implements a human-computer interaction process with the user by using a preset analysis algorithm, and the analysis algorithm includes: a psychological problem classification algorithm, and/or an emotion analysis algorithm, and/or a keyword extraction algorithm and/or a question-answer association deep learning algorithm.
Specifically, in this embodiment, the psychological problem classification algorithm performs multi-level and multi-dimensional classification on a large number of related materials, so as to search, recommend, analyze, and process the materials.
The emotion analysis algorithm is used for carrying out emotion analysis on subjective emotion information contained in sentences and chapters in the psychological library and summarizing viewpoints contained in the contents; and performing emotional analysis on the contents of the psychological and public opinions to acquire the subjective attitude of the team members. According to the subjective attitude of the team member, the content and the question and answer recommendation can be more humanized, and the requirements of the team member are better met. Emotion analysis can classify the input text into negative (derogatory), positive (recognition), neutral, etc. Evaluation words and combination evaluation units, word position characteristics, n-gram word characteristics, part of speech characteristics, upper and lower sentence emotion categories and the like appearing in the text are comprehensively considered and used for a machine learning model. The evaluation words and the combined evaluation unit are also used for establishing rules, and the final emotion classification result is obtained by combining the results of the machine learning model. The corpus resources and dictionary resources required by emotion analysis can be obtained through public resources (such as an MPQA news comment corpus, a HowNet evaluation word dictionary and the like), and special knowledge in a specified field can be added for supplement.
The keyword extraction algorithm selects a plurality of representative words in the text to prompt the central thought of the full text. In the process of processing massive texts (such as question and answer texts), the keywords of each text are summarized to help the question and answer robot to accurately grasp the intention of the team member. And (3) grading the key degrees of all words appearing in the article by combining technologies such as text sequencing, reverse document word frequency (TF-IDF), part of speech tagging and the like, thereby extracting a plurality of most meaningful keywords to be used as tags for the subsequent processing of the article.
The question-answer association deep learning algorithm gives answers automatically or in an auxiliary mode through artificial intelligence. Modeling with questions and answers, two algorithms can be used to implement intelligent questions and answers based on question and answer pairs. One approach is to build a large and complete question-and-answer library to provide answers to the naval crew by searching for technical matching questions. Another approach is to use machine learning to learn the questions and answers individually as a sequence model while joining their sequence features together. The sequence Model can be learned by using a recurrent neural network, for example, splicing two sentences S and T, dividing the middle by a special separator EOS, constructing an input layer of the RNN, sleeving a bidirectional and deep network structure, sleeving an Attention Model layer on the output of the highest RNN layer, calculating the Attention weight of each node of the BLSTM, and then multiplying each node by the Attention weight to sum to obtain a vector representation.
In the preferred embodiment of the present invention, the intelligent question answering module 13 includes:
the dialogue starting unit 131 is used for outputting a preset comprehensive welcome language as a dialogue starting language when the user is judged to be in the first dialogue and outputting a welcome introduction language as a dialogue starting language when the user is judged not to be in the first dialogue;
the dialogue counting unit 132 is connected with the dialogue starting unit 131 and used for receiving and counting the current speaking times and the current speaking word number corresponding to the user questioning content of the user after the dialogue starting language is given and when the user is judged not to leave;
the intelligent question-answering unit 133 is connected with the dialogue counting unit 132 and used for giving a question answer corresponding to the question content of the user when the current spoken word number is not more than a word number threshold;
the professional content recommending unit 134 is connected with the dialogue counting unit 132 and is used for giving professional recommended content corresponding to the user question content when the current spoken word number is greater than the word number threshold;
the user information collecting unit 135 is connected to the session counting unit 132, and is configured to collect user information when the current speaking times in a preset time period is greater than a threshold value of a number of times.
Specifically, in this embodiment, the dialog of the intelligent question and answer module 13 starts from the dialog starting unit 131, determines whether the team member is in the first dialog, and sends a relatively comprehensive welcome language to introduce the functions and functions of the intelligent service software system when the team member is in the first dialog; and issuing a reputable introductory when the crew is not engaged in the first conversation. And judges whether the team member leaves after the dialogue is spoken, and receives and counts the current speaking times and the current speaking word number corresponding to the questioning content of the team member when the team member does not leave through the dialogue counting unit 132. In this real-time example, the word count threshold may be 100, the preset time period may be 24 hours, and the number threshold may be 5. The dialogue statistic unit 132 determines whether the current number of words spoken is greater than 100: if the number is more than 100, calling the professional content recommendation unit 134 to give professional recommendation content corresponding to the questioning content of the team member; if not, the dialog statistic unit 132 determines whether the player has a chat within 24 hours: if there is no chat, the intelligent question-answering module 13 sends a prompt utterance to attract team members to chat; if there is a chat, the dialog statistic unit 132 determines whether the current number of utterances is greater than 5: if the number of times is more than 5, calling the user information collection unit 135 to collect the crew information; if the number of the input times is not more than 5, the input of the crew is ready to be accepted.
In this embodiment, the core hardware used by the intelligent psychological service software system includes: the system comprises at least two servers for hot backup, at least one administrator personal computer, at least one conversation robot terminal device, at least one carrier personal computer, a plurality of intelligent mobile terminals, a plurality of matched earphones and at least one network switch. All the core hardware devices are connected to the public cloud server through the Internet. The server is respectively deployed with an intelligent question and answer service, an intelligent recommendation service, a conversation process service and a public cloud voice service, and all data are stored and managed. The intelligent question-answering service is a service executed by the intelligent question-answering unit 133, the intelligent recommendation service is a service executed by the professional content recommendation unit 134, and the conversation process service is a service executed by the conversation starting unit 131 and the conversation statistical unit 132. The team member personal computer is provided with client software, when the team member is inconvenient for voice conversation, man-machine conversation can be carried out with the system in a text input mode through the intelligent mobile terminal, and the network switch is used for providing more connection ports for the sub-network so as to be connected with more crew personal computers. The administrator personal computer manages the running state of the whole system and monitors the behavior of the team members in real time. The conversation robot terminal equipment deploys a conversation client to realize face-to-face voice communication with team members. In the aspect of voice service, a commercial mature cloud voice platform is selected for providing support, multiple voice services are provided for an application terminal in a cloud service mode, and service contents comprise:
voice transcription: the conversion from voice to text is realized, the content is accurate and quick, and the manual input time is saved;
voice quality inspection: intelligently detecting the input content of a naval crew and checking the content which does not meet the standard;
semantic understanding: semantic understanding is carried out on related input contents, and intelligent processing of languages is achieved;
and (3) voice synthesis: the character-to-voice is clear and natural, and the sound production effect is comparable to the human voice.
Dialogue robot terminal equipment chooses commercial ripe intelligent hardware equipment for use, reaches the communication through net twine and "public cloud", gathers team member's sound in real time and sends and handle for cloud speech service to handle cloud service and broadcast the sound that comes back to the transmission, supporting earphone can be used for receiving the sound that comes back.
In a preferred embodiment of the present invention, the intelligent question answering unit 133 includes:
a first judging subunit 136, configured to give a pre-configured question answer associated with the commonly used word when the user question content is judged to be the commonly used word, and give a first judgment result when the user question content is judged not to be the commonly used word;
the first processing sub-unit 137, connected to the first determining sub-unit 136, is configured to perform keyword extraction on the user question content according to the first determination result, and provide a first prompt message when the keyword is not extracted, and provide a first processing result when the keyword is extracted;
a second judging subunit 138, connected to the first processing subunit 137, for providing a second judging result when the keyword is judged to be in the preset white list according to the first processing result, and providing a third judging result when the keyword is not in the white list;
a second processing subunit 139, connected to the second judging subunit 138, configured to provide a second prompt message when the question answer of the associated keyword is not matched according to the second judgment result, and provide the question answer when the question answer of the associated keyword is matched;
the third processing subunit 140, connected to the second judging subunit 138, is configured to provide a third prompting message when the question answer of the associated keyword is not matched according to the third judgment result, and provide a fourth prompting message and the question answer when the question answer of the associated keyword is matched.
In a preferred embodiment of the present invention, the professional content recommending unit 134 comprises:
the recommendation subunit 141 is configured to invoke a pre-stored professional recommendation algorithm to perform processing according to the user question content, and output a professional recommendation content with the highest matching degree with the user question content when the processing result includes at least one professional recommendation content;
and the classification subunit 142, connected to the recommendation subunit 141, is configured to perform type classification on the professional recommendation content, and output a corresponding guidance phrase and the professional recommendation content according to a type classification result for a user to select.
Specifically, in this embodiment, the parameters of the pre-stored professional recommendation algorithm include a recommendation target, and/or a crew content, and/or a matching degree threshold. The classification subunit 142 classifies the common psychological problems of the team members, associates the political education content data and the short text consultation questions and answers, professional consultation techniques, psychological assessment, science library, and ideological and meditation audio and video database data in the psychological expert knowledge base with the common psychological problem classification system, and further realizes retrieval and recommendation of related data contents and analysis processing of corresponding data sets. At present, according to sources, heading words, text common words, text keywords and emotion words mentioned in the text, based on expert rules and machine learning, and combined with a logistic regression and naive Bayes model, when massive contents are classified, 88% accuracy and 89% recall rate can be realized.
In a preferred embodiment of the present invention, the user information collecting unit 135 includes:
the collecting subunit 143 is configured to collect user information according to a preset information research table, where the information research table includes a plurality of research questions;
the third judging subunit 144, connected to the collecting subunit 143, is configured to output a welcome phrase to ask the user whether the user wants to perform research when it is judged that the collection of all research questions is not completed, and continue to collect the research questions one by one when the user wants to perform research until the collection of all research questions is completed.
Specifically, in this embodiment, the information categories collected in the information research table include: the team members have an existing family, and/or the mood of the team members, and/or the economic level of the team members, and/or the growth of the individual team members.
In a preferred embodiment of the present invention, the psychological intervention module 2 analyzes the user question content according to a pre-generated offline construction model, and pushes the preset psychological intervention content associated with the user question content to the user according to the analysis result.
In a preferred embodiment of the present invention, the offline-structured model includes a word vector learning database, a recommended content database, a psychological problem knowledge map, a cure content knowledge map, and a psychological question-and-answer knowledge map.
Specifically, in this embodiment, the word vector learning database constructs a massive learning database through network resources to train a deep learning-based word vector model. The word vector model is a deep learning model, helps a system to evaluate distances between words and between sentences, and plays a key role in quantitative evaluation of semantics.
The recommended content database is created and reviewed by an expert team for a variety of content helpful to the vessel crew, including articles, assessments, audio, and the like. A content library is established by adopting MongoDB and elastic search, keywords of the content library are extracted by an unsupervised machine learning model, and the comprehensiveness of content description is increased by manual confirmation after the keywords are expanded by a word vector model.
The mental problem knowledge map is used for extracting key words aiming at mental layer problems, outputting synonym and synonym lists by a word vector-based method, and expanding the key words through manual confirmation.
The construction mode of the healing content knowledge graph is similar to that of a mental level knowledge graph, synonym and synonym lists are output for given keywords by a word vector-based method, and the synonym and synonym lists are expanded through manual confirmation.
The psychological question-answer knowledge graph is constructed based on big data analysis of the question-answer data of the team members corresponding to each keyword, and the healing words with the maximum relevance are obtained.
In a preferred embodiment of the present invention, the psychological intervention module 2 comprises:
a keyword extraction unit 21, configured to perform keyword extraction on user question content to obtain a question keyword;
the content recommending unit 22 is connected with the keyword extracting unit 21 and is used for expanding the question keywords according to the psychological question knowledge graph to obtain question expanded keywords, processing the question expanded keywords according to the psychological question-answer knowledge graph to obtain answer keywords related to the question expanded keywords, expanding the answer keywords according to the cure content knowledge graph to obtain answer expanded keywords, performing semantic quantization on the answer expanded keywords according to the word vector learning database, and matching the result of the semantic quantization in the recommended content database to obtain psychological intervention content to be pushed to the user.
Specifically, in this embodiment, the keyword extraction unit 21 extracts keywords from the questioning content of the team members to obtain questioning keywords, and then constructs a mental knowledge graph. The content recommendation unit 22 finds answer keywords associated with the question keywords from the psychological question knowledge graph, then the answer keywords are expanded through the cure content knowledge graph to obtain answer expanded keywords, then the answer expanded keywords are subjected to semantic quantization through the word vector learning database, finally the most targeted recommended content is searched and sequenced in the recommended content database according to semantic quantization results, and the recommended content is output and pushed to the naval staff.
While the invention has been described with reference to a preferred embodiment, it will be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the spirit and scope of the invention.
Claims (10)
1. An intelligent psychological service software system is characterized by specifically comprising:
the accompany support module is used for realizing a man-machine conversation process with a user according to prestored short text question-answer data;
the psychological intervention module is used for pushing preset psychological intervention contents related to the user question contents to the user according to the user question contents in the man-machine conversation process;
the mental health screening module is used for carrying out psychological evaluation on the user to obtain an evaluation result and providing corresponding psychological guidance for the user according to the evaluation result;
the growth training module is used for providing at least one mental training toolkit for the user to perform mental elasticity training;
and the psychological file module is respectively connected with the accompany support module, the psychological intervention module, the psychological health screening module and the growth training module and is used for summarizing all data in the man-machine conversation process and establishing the personal psychological file of the user.
2. The intelligent psychological service software system according to claim 1, wherein the companion support module comprises:
the big data collecting sub-module is used for collecting political education content data, and/or psychology big data, and/or question and answer big data and/or mental health big data as the short text question and answer data;
the big data management submodule is connected with the big data collecting submodule and is used for processing the short text question-answer data collected by the big data collecting unit;
and the intelligent question-answer module is connected with the big data management module and is used for realizing a man-machine conversation process with the user according to the processed short text question-answer data.
3. The intelligent psychological service software system according to claim 2, wherein the intelligent question-answering module implements a human-machine interaction process with the user by using a preset analysis algorithm, and the analysis algorithm comprises: a psychological problem classification algorithm, and/or an emotion analysis algorithm, and/or a keyword extraction algorithm and/or a question-answer association deep learning algorithm.
4. The intelligent psychological service software system according to claim 3, wherein the intelligent question-answering module includes:
the dialogue starting unit is used for outputting a preset comprehensive welcome language as a dialogue starting language when the user is judged to be in the first dialogue and outputting a welcome introduction language as a dialogue starting language when the user is judged not to be in the first dialogue;
the dialogue counting unit is connected with the dialogue starting unit and used for receiving and counting the current speaking times and the current speaking word number corresponding to the questioning content of the user after the dialogue starting language is given and when the user is judged not to leave;
the intelligent question-answering unit is connected with the dialogue counting unit and used for giving a question answer corresponding to the question content of the user when the current speaking word number is not more than a word number threshold;
the professional content recommending unit is connected with the dialogue counting unit and used for giving out professional recommended content corresponding to user question content when the current speaking word number is larger than the word number threshold;
and the user information collection unit is connected with the conversation statistical unit and used for collecting the user information when the current speaking times in a preset time period is greater than a threshold value of a number of times.
5. The intelligent psychological service software system according to claim 4, wherein the intelligent question-answering unit comprises:
the first judging subunit is used for giving a pre-configured question answer associated with the common word when judging that the user question content is the common word, and giving a first judging result when judging that the user question content is not the common word;
the first processing subunit is connected with the first judging subunit and used for extracting keywords from the user questioning content according to the first judging result, giving a first prompt message when the keywords are not extracted, and giving a first processing result when the keywords are extracted;
the second judgment subunit is connected with the first processing subunit and used for giving a second judgment result when the keyword is judged to be in a preset white list according to the first processing result and giving a third judgment result when the keyword is not in the white list;
the second processing subunit is connected with the second judging subunit and used for giving a second prompt message according to the second judgment result when the question answer associated with the keyword is not matched and giving the question answer when the question answer associated with the keyword is matched;
and the third processing subunit is connected with the second judging subunit and is used for giving a third prompt message when the question answer associated with the keyword is not matched according to the third judging result and giving a fourth prompt message and the question answer when the question answer associated with the keyword is matched.
6. The intelligent psychological service software system according to claim 4, wherein the professional content recommendation unit comprises:
the recommendation subunit is used for calling a pre-stored professional recommendation algorithm to process according to the user question content, and outputting the professional recommendation content with the highest matching degree with the user question content when the processing result contains at least one professional recommendation content;
and the classification subunit is connected with the recommendation subunit and is used for performing type division on the professional recommendation content and outputting corresponding guide words and the professional recommendation content according to a type division result for the user to select.
7. The intelligent psychological service software system according to claim 4, wherein the user information collecting unit comprises:
the collecting subunit is used for collecting user information according to a preset information research table, and the information research table comprises a plurality of research problems;
and the third judging subunit is connected with the collecting subunit and is used for outputting a welcome language to inquire whether the user wants to perform research or not when judging that all the research questions are not collected completely, and continuously collecting the research questions one by one when the user wants to perform research until all the research questions are collected completely.
8. The intelligent psychological service software system of claim 1 wherein the psychological intervention module analyzes the user question content according to a pre-generated offline construction model and pushes the preset psychological intervention content associated with the user question content to the user according to the analysis result.
9. The intelligent psychological service software system of claim 8 wherein the offline built models include a word vector learning database, a recommended content database, a psychological problem knowledge graph, a healing content knowledge graph, and a psychological question and answer knowledge graph.
10. The intelligent psychological service software system of claim 8 wherein the psychological intervention module comprises:
the keyword extraction unit is used for extracting keywords from the questioning content of the user to obtain questioning keywords;
and the content recommendation unit is connected with the keyword extraction unit and used for expanding the question keywords according to the psychological question knowledge graph to obtain question expansion keywords, processing the question expansion keywords according to the psychological question-answer knowledge graph to obtain answer keywords related to the question expansion keywords, expanding the answer keywords according to the cure content knowledge graph to obtain answer expansion keywords, performing semantic quantization on the answer expansion keywords according to the word vector learning database, and matching the semantic quantization results in the recommended content database to obtain the psychological intervention content to be pushed to the user.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011205048.8A CN112463937B (en) | 2020-11-02 | 2020-11-02 | Intelligent psychological service software system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011205048.8A CN112463937B (en) | 2020-11-02 | 2020-11-02 | Intelligent psychological service software system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112463937A true CN112463937A (en) | 2021-03-09 |
| CN112463937B CN112463937B (en) | 2022-10-11 |
Family
ID=74834982
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011205048.8A Active CN112463937B (en) | 2020-11-02 | 2020-11-02 | Intelligent psychological service software system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112463937B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102760128A (en) * | 2011-04-26 | 2012-10-31 | 华东师范大学 | Telecommunication field package recommending method based on intelligent customer service robot interaction |
| WO2015109378A1 (en) * | 2014-01-23 | 2015-07-30 | Nobre Ismael | Integrated and automated method for obtaining, managing and making available multidimensional information for carrying out public perception and opinion surveys, and equipment for implementing the method |
| CN108121824A (en) * | 2018-01-12 | 2018-06-05 | 北京融快线科技有限公司 | A kind of chat robots and system towards financial service |
| CN109979450A (en) * | 2019-03-11 | 2019-07-05 | 青岛海信电器股份有限公司 | Information processing method, device and electronic equipment |
| CN111564202A (en) * | 2020-04-30 | 2020-08-21 | 深圳市镜象科技有限公司 | Psychological counseling method based on man-machine conversation, psychological counseling terminal and storage medium |
-
2020
- 2020-11-02 CN CN202011205048.8A patent/CN112463937B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102760128A (en) * | 2011-04-26 | 2012-10-31 | 华东师范大学 | Telecommunication field package recommending method based on intelligent customer service robot interaction |
| WO2015109378A1 (en) * | 2014-01-23 | 2015-07-30 | Nobre Ismael | Integrated and automated method for obtaining, managing and making available multidimensional information for carrying out public perception and opinion surveys, and equipment for implementing the method |
| CN108121824A (en) * | 2018-01-12 | 2018-06-05 | 北京融快线科技有限公司 | A kind of chat robots and system towards financial service |
| CN109979450A (en) * | 2019-03-11 | 2019-07-05 | 青岛海信电器股份有限公司 | Information processing method, device and electronic equipment |
| CN111564202A (en) * | 2020-04-30 | 2020-08-21 | 深圳市镜象科技有限公司 | Psychological counseling method based on man-machine conversation, psychological counseling terminal and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 左昕，彭李，汪金生，李敏，钟敏琴，杨安强，陈昊，于永菊: "心理弹性训练对水面舰艇军人自我意识和应对方式的影响", 《第三军医大学学报》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112463937B (en) | 2022-10-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10051122B2 (en) | Modeling voice calls to improve an outcome of a call between a representative and a customer | |
| RU2747425C2 (en) | Real-time answer system to questions from different fields of knowledge | |
| Hixon et al. | Learning knowledge graphs for question answering through conversational dialog | |
| JP5149737B2 (en) | Automatic conversation system and conversation scenario editing device | |
| KR102033388B1 (en) | Apparatus and method for question answering | |
| CN101799849A (en) | Method for realizing non-barrier automatic psychological consult by adopting computer | |
| CN110046230B (en) | Method for generating recommended speaking collection, and recommended speaking method and device | |
| JP2017215931A (en) | Conference support system, conference support device, conference support method, and program | |
| US9922644B2 (en) | Analysis of professional-client interactions | |
| CN109829042A (en) | A kind of knowledge Q-A system and method based on Biomedical literature | |
| CN111899140A (en) | Customer service training method and system based on dialect level improvement | |
| JP4383328B2 (en) | System and method for semantic shorthand | |
| US20160005421A1 (en) | Language analysis based on word-selection, and language analysis apparatus | |
| Tseng et al. | Approaching Human Performance in Behavior Estimation in Couples Therapy Using Deep Sentence Embeddings. | |
| CN112463937B (en) | Intelligent psychological service software system | |
| CN116881730A (en) | Chat scene matching system, method, equipment and storage medium based on context | |
| CN109086372B (en) | Artificial intelligence-based multi-round conversation method and system applied to legal consultation | |
| CN113505606B (en) | Training information acquisition method and device, electronic equipment and storage medium | |
| CN109726002A (en) | A kind of process flow method of adjustment and device | |
| US11837251B2 (en) | Virtual counseling system and counseling method using the same | |
| JP4451037B2 (en) | Information search system and information search method | |
| CN110858234A (en) | Method and device for pushing information according to human emotion | |
| CN114328867A (en) | Intelligent interruption method and device in man-machine conversation | |
| CN114661864A (en) | Psychological consultation method and device based on controlled text generation and terminal equipment | |
| CN112818090B (en) | Method and system for generating answer questions and questions based on harmonic words |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |