1. An intelligent intent recognition analysis system based on blockchain technology, comprising:

a data collection module: the data collection module collects voice data, health data and user voiceprint data through intelligent equipment and an application program;

a data extraction module: the data extraction module preprocesses the collected data to obtain preprocessed instruction data, voice voiceprint data, health data and user voiceprint data;

an information collection module: the information collection module uploads the collected health data and user voiceprint data to the block chain, matches the voice voiceprint data with the user voiceprint data on the block chain, and binds and uploads the current voiceprint data and the instruction to the block chain when the extracted voice voiceprint data is matched with the user voiceprint data on the block chain; when the extracted voice voiceprint data are not matched with the user voiceprint data on the block chain, marking the current instruction;

an intention analysis module: and the intention analysis module classifies the matched instruction data and correspondingly processes the instruction data based on the block chain data according to intention classification.

2. The intelligent blockchain-based intent recognition analysis system of claim 1, wherein the data collection module comprises:

the data collection module M1: inputting initial data including user voiceprint data and health data through intelligent equipment, and collecting voice data when the intelligent equipment is used;

the data collection module M2: uploading the health data and the user voiceprint data to a blockchain through an application program;

the data collection module M3: health data are collected through a monitoring function on the intelligent device.

3. The intelligent block chain technology-based intention recognition and analysis system of claim 1, wherein said data extraction module comprises:

the data extraction module M4: converting voice data collected by intelligent equipment into instruction text and voice voiceprint data;

the data extraction module M5: filtering health data collected by intelligent equipment;

the data extraction module M6: and filtering the health data and the user voiceprint data collected by the application program.

4. The intelligent blockchain-based intent recognition analysis system of claim 1, wherein in the intent analysis module: classifying the matched instruction data, wherein the instruction data comprises a functional intention, a potential health intention and a meal ordering intention, and performing corresponding processing based on the block chain data according to intention classification;

when the matched instruction text is a potential health intention, comparing the matched instruction text with the existing health data on the block chain, comparing the existing suggestions on the block chain with the health data of the current user to obtain a recommendation message, and feeding the recommendation message back to an intention analysis module, wherein the intention analysis module feeds back the recommendation message through a loudspeaker;

when the matched instruction text is the food ordering intention, transmitting the current instruction text to the block chain, selecting the current edible diet recommendation according to the existing food rule, time and physical data analysis on the block chain, and feeding back the current edible diet recommendation to the intention analysis module, wherein the intention analysis module feeds back the current edible diet recommendation through a loudspeaker;

when the instruction text is the functional intention, the direct feedback is stated through the intelligent equipment voice.

5. The system of claim 4, wherein the matched command text is uploaded to the blockchain when the matched command text is judged to be health related data by the intent analysis module in the intent analysis module.

6. The intelligent intent recognition and analysis system based on blockchain technology of claim 5, wherein the data analysis module analyzes the executed instructions;

when the current instruction is executed, comparing the health data on the block chain with the executed instruction, generating a health report, encrypting the health report and uploading the health report to the chain; when the current instruction cannot be executed, the data analysis module judges the current data as a dangerous signal, pushes the dangerous signal to a preset object through the timely warning module, and records the chain.

7. The intelligent blockchain-based intent recognition analysis system of claim 6, wherein in the data analysis module: combining the health data uploaded by the intelligent equipment on the block chain with the health related data stored in the block chain to generate a health report;

and classifying the instructions to be completed in the future, and submitting the instructions to a module to be fed back.

8. The intelligent block chain technology-based intention recognition and analysis system of claim 4, wherein the intelligent device comprises a smart speaker and a smart bracelet;

the intelligent sound box registers a personal account in a voiceprint and public and private key symmetric mode, and binds the identity of an intelligent sound box owner;

the intelligent sound box records voiceprint and health data when recording initial data, collects sound data during daily communication and collects sound data when using a sound box network for communication;

health data including heart rate and blood oxygen of a user are collected through a monitoring function on the smart bracelet, and the collected data are uploaded to a block chain;

inputting personal information, health data and associated person information in the application program; entering personal information includes: user name, nickname, and voiceprint.

9. An intelligent intention recognition analysis method based on a block chain technology is characterized by comprising the following steps:

step S1: collecting voice data, health data and user voiceprint data by using intelligent equipment and an application program;

step S2: preprocessing the collected data to obtain preprocessed instruction data, voice print data, health data and user voice print data;

step S3: uploading the collected health data and user voiceprint data to a block chain, matching the voice voiceprint data with the user voiceprint data on the block chain, and binding and uploading the current voiceprint data and instructions to the block chain when the extracted voice voiceprint data is matched with the user voiceprint data on the block chain; when the extracted voice voiceprint data are not matched with the user voiceprint data on the block chain, marking the current instruction;

step S4: and classifying the preprocessed instruction data, and performing corresponding processing based on the block chain data according to intention classification.

10. The intelligent intent recognition analysis method according to claim 9, wherein said step S1 comprises:

step S1.1: inputting initial data including user voiceprint data and health data through intelligent equipment, and collecting voice data when the intelligent equipment is used;

step S1.2: uploading the health data and the user voiceprint data to a blockchain through an application program;

step S1.3: collecting health data through a monitoring function on intelligent equipment;

the step S2 includes:

step S2.1: converting voice data collected by intelligent equipment into instruction text and voice voiceprint data;

step S2.2: filtering health data collected by intelligent equipment;

step S2.3: filtering the health data and the user voiceprint data collected by the application program;

in step S3, performing classification processing on the matched instruction data, including functional intent, potential health intent and meal ordering intent, and performing corresponding processing based on the blockchain data according to intent classification;

when the matched instruction text is a potential health intention, comparing the matched instruction text with the existing health data on the block chain, comparing the existing suggestions on the block chain with the health data of the current user to obtain a recommendation message, and feeding the recommendation message back to an intention analysis module, wherein the intention analysis module feeds back the recommendation message through a loudspeaker;

when the matched instruction text is the food ordering intention, transmitting the current instruction text to the block chain, selecting the current edible diet recommendation according to the existing food rule, time and physical data analysis on the block chain, and feeding back the current edible diet recommendation to the intention analysis module, wherein the intention analysis module feeds back the current edible diet recommendation through a loudspeaker;

when the instruction text is the functional intention, directly feeding back through the voice statement of the intelligent equipment;

in step S3, the matched instruction text is uploaded to the block chain when judged to be health related data by the intention analysis module;

wherein, the executed instruction is analyzed;

when the current instruction is executed, comparing the health data on the block chain with the executed instruction, generating a health report, encrypting the health report and uploading the health report to the chain; when the current instruction cannot be executed, the data analysis module judges the current data as a dangerous signal, pushes the dangerous signal to a preset object through the timely warning module, and records the chain;

the health report is generated by combining health data uploaded by intelligent equipment on the block chain and health related data stored in the block chain;

classifying the instructions to be completed in the future, and submitting the instructions to a module to be fed back;

the intelligent equipment comprises an intelligent sound box and an intelligent bracelet;

the intelligent sound box registers a personal account in a voiceprint and public and private key symmetric mode, and binds the identity of an intelligent sound box owner;

the intelligent sound box records voiceprint and health data when recording initial data, collects sound data during daily communication and collects sound data when using a sound box network for communication;

health data including heart rate and blood oxygen of a user are collected through a monitoring function on the smart bracelet, and the collected data are uploaded to a block chain;

inputting personal information, health data and associated person information in the application program; entering personal information includes: user name, nickname, and voiceprint.

Background

Patent document CN111899878A (application number: 202010752335.4) discloses an elderly health detection system, which can be applied in a smart medical scene to promote the construction of a smart city. The system comprises: the user data acquisition module is used for acquiring body parameter data and voice data of a user; the organization module is used for acquiring service data generated when different service mechanisms serve the users; the application program module is used for acquiring the body parameter data, the voice data and the service data, and evaluating the body health condition of the user according to the body parameter data and the service data to generate a body health evaluation result; the automatic machine learning module is used for analyzing the voice data and the service data to evaluate the mental health condition of the user to generate a mental health evaluation result and feeding the mental health evaluation result back to the application program module; and the block chain module is used for synchronizing the body parameter data, the voice data and the service data to each node of the block chain, so that data sharing can be realized, and the physical and mental health of the old can be guaranteed comprehensively and actively. The system cannot integrate the life of the user, and can only respond when the user actively asks health problems or the bracelet detects. The function is single, and the use viscosity of a user is difficult to guarantee. The design idea provided by the invention can ensure the use frequency of the user, namely the viscosity of the user, by combining three systems of entertainment, communication and diet, thereby achieving the purpose of protecting the physical and mental health of the user unconsciously in daily use.

In the prior art, the intelligent sound box combines the technologies of voice recognition, a block chain, an intelligent bracelet, meal ordering, communication and the like. But the disadvantages of the prior art include: the function is single, and the viscosity of a user is difficult to ensure. In the prior art, feedback can be performed only when a user actively asks health problems or the bracelet detects that problems occur, and the single health problem communication hardly guarantees the use frequency and viscosity of the user. The data collection process is limited, and an effective solution cannot be provided while the result is fed back. The prior art can only solve the problem when the user proposes the problem, and can not propose solution proposal except reminding the doctor; the invention can judge the possible problems of the user at any time during the daily communication of the user, for example, the user asks a question at 0: 00: "it will not be cold when going out and running now". The intelligent intent recognition system will determine the current time, combine the current weather, combine the previous body condition history uploaded by the user to give feedback "running at the current time will affect XXX on your body, and suggest running again at XXX". The prior art has low identification degree, is easy to cause data judgment errors and has poor safety. Since a user using voice may be using more than one person, the prior art cannot distinguish whether the person who is currently asking for a health question is himself. Such as: the person who is currently asking a health question is the wife who is currently using the user. She would have a serious misjudgment problem after asking her questions if she entered the current data into the user's health data sheet and analyzed. The technology of the invention integrates voiceprint recognition (listening to the tone to judge the owner) and can be used for recording a plurality of users, and when the users do not use the system, the users can actively ask questions to judge whether the data are linked. The rigor of the data is increased while the use of multiple users is ensured. The safety of the prior art is not strong enough; since the user data is used as public analysis of medical records after uplink, security is an important part of the technology. The prior art uses the following modes: encryption, invalidation or deletion, masking, rounding of date offsets, etc. But such approaches, if learned about specific rules, can easily extract critical information. Thereby causing data security risks. The mode provided by the invention is a public and private key symmetric mode for registering an account number. In conjunction with voiceprint upload data, data is unlocked (user's timbre is uploaded synchronously during uplink). The security guarantee is enhanced on the basis of the original encryption. The health data is collected singly. Diet is an important component of health analysis. In the prior art, a user data acquisition module acquires data actively uploaded by an intelligent bracelet, voice and a user. The food ordering module is combined into the system, so that the food intake of the user is silently counted on the premise of helping the user to order food. And further, judgment basis is added to the health analysis of the user. Meanwhile, under the support of a specific intention recognition system, the part of daily communication chatting is analyzed, and if data of health basis exists, the data can be stored in a block chain as part of health data analysis. Meanwhile, the online communication function provided by the invention can monitor effective health information under the condition of user authorization and monitor the health problems of the user as explained by relatives. The health data information is collected through an effective channel.

Disclosure of Invention

Aiming at the defects in the prior art, the invention aims to provide an intelligent intention recognition and analysis system and method based on a block chain technology.

The invention provides an intelligent intention recognition analysis system based on a block chain technology, which comprises:

a data collection module: the data collection module collects voice data, health data and user voiceprint data through intelligent equipment and an application program;

a data extraction module: the data extraction module preprocesses the collected data to obtain preprocessed instruction data, voice voiceprint data, health data and user voiceprint data;

an information collection module: the information collection module uploads the collected health data and user voiceprint data to the block chain, matches the voice voiceprint data with the user voiceprint data on the block chain, and binds and uploads the current voiceprint data and the instruction to the block chain when the extracted voice voiceprint data is matched with the user voiceprint data on the block chain; when the extracted voice voiceprint data are not matched with the user voiceprint data on the block chain, marking the current instruction;

an intention analysis module: and the intention analysis module classifies the matched instruction data and correspondingly processes the instruction data based on the block chain data according to intention classification.

Preferably, the data collection module comprises:

the data collection module M1: inputting initial data including user voiceprint data and health data through intelligent equipment, and collecting voice data when the intelligent equipment is used;

the data collection module M2: uploading the health data and the user voiceprint data to a blockchain through an application program;

the data collection module M3: health data are collected through a monitoring function on the intelligent device.

Preferably, the data extraction module comprises:

the data extraction module M4: converting voice data collected by intelligent equipment into instruction text and voice voiceprint data;

the data extraction module M5: filtering health data collected by intelligent equipment;

the data extraction module M6: and filtering the health data and the user voiceprint data collected by the application program.

Preferably, in the intent analysis module: classifying the matched instruction data, wherein the instruction data comprises a functional intention, a potential health intention and a meal ordering intention, and performing corresponding processing based on the block chain data according to intention classification;

when the matched instruction text is a potential health intention, comparing the matched instruction text with the existing health data on the block chain, comparing the existing suggestions on the block chain with the health data of the current user to obtain a recommendation message, and feeding the recommendation message back to an intention analysis module, wherein the intention analysis module feeds back the recommendation message through a loudspeaker;

when the matched instruction text is the food ordering intention, transmitting the current instruction text to the block chain, selecting the current edible diet recommendation according to the existing food rule, time and physical data analysis on the block chain, and feeding back the current edible diet recommendation to the intention analysis module, wherein the intention analysis module feeds back the current edible diet recommendation through a loudspeaker;

when the instruction text is the functional intention, the direct feedback is stated through the intelligent equipment voice.

Preferably, in the intention analysis module, the matched instruction text is uploaded to the block chain when judged as health-related data by the intention analysis module.

Preferably, the data analysis module analyzes the executed instruction;

when the current instruction is executed, comparing the health data on the block chain with the executed instruction, generating a health report, encrypting the health report and uploading the health report to the chain; when the current instruction cannot be executed, the data analysis module judges the current data as a dangerous signal, pushes the dangerous signal to a preset object through the timely warning module, and records the chain.

Preferably, in the data analysis module: combining the health data uploaded by the intelligent equipment on the block chain with the health related data stored in the block chain to generate a health report;

and classifying the instructions to be completed in the future, and submitting the instructions to a module to be fed back.

Preferably, the smart device comprises a smart speaker and a smart bracelet;

the intelligent sound box registers a personal account in a voiceprint and public and private key symmetric mode, and binds the identity of an intelligent sound box owner;

the intelligent sound box records voiceprint and health data when recording initial data, collects sound data during daily communication and collects sound data when using a sound box network for communication;

health data including heart rate and blood oxygen of a user are collected through a monitoring function on the smart bracelet, and the collected data are uploaded to a block chain;

inputting personal information, health data and associated person information in the application program; entering personal information includes: user name, nickname, and voiceprint.

The invention provides an intelligent intention recognition analysis method based on a block chain technology, which comprises the following steps:

step S1: collecting voice data, health data and user voiceprint data by using intelligent equipment and an application program;

step S2: preprocessing the collected data to obtain preprocessed instruction data, voice print data, health data and user voice print data;

step S3: uploading the collected health data and user voiceprint data to a block chain, matching the voice voiceprint data with the user voiceprint data on the block chain, and binding and uploading the current voiceprint data and instructions to the block chain when the extracted voice voiceprint data is matched with the user voiceprint data on the block chain; when the extracted voice voiceprint data are not matched with the user voiceprint data on the block chain, marking the current instruction;

step S4: and classifying the preprocessed instruction data, and performing corresponding processing based on the block chain data according to intention classification.

Preferably, the step S1 includes:

step S1.1: inputting initial data including user voiceprint data and health data through intelligent equipment, and collecting voice data when the intelligent equipment is used;

step S1.2: uploading the health data and the user voiceprint data to a blockchain through an application program;

step S1.3: collecting health data through a monitoring function on intelligent equipment;

the step S2 includes:

step S2.1: converting voice data collected by intelligent equipment into instruction text and voice voiceprint data;

step S2.2: filtering health data collected by intelligent equipment;

step S2.3: filtering the health data and the user voiceprint data collected by the application program;

in step S3, performing classification processing on the matched instruction data, including functional intent, potential health intent and meal ordering intent, and performing corresponding processing based on the blockchain data according to intent classification;

when the matched instruction text is a potential health intention, comparing the matched instruction text with the existing health data on the block chain, comparing the existing suggestions on the block chain with the health data of the current user to obtain a recommendation message, and feeding the recommendation message back to an intention analysis module, wherein the intention analysis module feeds back the recommendation message through a loudspeaker;

when the matched instruction text is the food ordering intention, transmitting the current instruction text to the block chain, selecting the current edible diet recommendation according to the existing food rule, time and physical data analysis on the block chain, and feeding back the current edible diet recommendation to the intention analysis module, wherein the intention analysis module feeds back the current edible diet recommendation through a loudspeaker;

when the instruction text is the functional intention, directly feeding back through the voice statement of the intelligent equipment;

in step S3, the matched instruction text is uploaded to the block chain when judged to be health related data by the intention analysis module;

wherein, the executed instruction is analyzed;

when the current instruction is executed, comparing the health data on the block chain with the executed instruction, generating a health report, encrypting the health report and uploading the health report to the chain; when the current instruction cannot be executed, the data analysis module judges the current data as a dangerous signal, pushes the dangerous signal to a preset object through the timely warning module, and records the chain;

the health report is generated by combining health data uploaded by intelligent equipment on the block chain and health related data stored in the block chain;

classifying the instructions to be completed in the future, and submitting the instructions to a module to be fed back;

the intelligent equipment comprises an intelligent sound box and an intelligent bracelet;

the intelligent sound box registers a personal account in a voiceprint and public and private key symmetric mode, and binds the identity of an intelligent sound box owner;

the intelligent sound box records voiceprint and health data when recording initial data, collects sound data during daily communication and collects sound data when using a sound box network for communication;

health data including heart rate and blood oxygen of a user are collected through a monitoring function on the smart bracelet, and the collected data are uploaded to a block chain;

inputting personal information, health data and associated person information in the application program; entering personal information includes: user name, nickname, and voiceprint.

Compared with the prior art, the invention has the following beneficial effects:

1. by adopting a block chain encryption and decryption technology, the invention solves the problems that the old user cannot evaluate the health of the old user by using voice equipment such as a sound box in life on the premise of protecting the privacy of the user, and cannot know and timely feed back the health condition of the old user in time without flexible voice control because the old user has high threshold for using a smart phone, is inconvenient and cannot timely know the physical condition and appeal of the old user when the old user is not around the old user;

2. according to the invention, through public and private key encryption and decryption technology of the block chain, the voiceprint recognition technology realizes the technical effect of enhancing the privacy security of the user; meanwhile, the technical effect that a plurality of users use the same intelligent equipment without mixing the health data judgment result is realized through the uniqueness of sound;

3. according to the invention, the collection of diet health data is realized through the meal ordering function, powerful data conditions are provided for health data analysis, and the technical effect of analyzing health data such as food intake, types and the like is realized;

4. the sound box communication technology has the characteristics of convenience and rapidness, and no manual operation is needed, so that the beneficial effects of convenience in telephone communication of the old people with visual impairment, inconvenient movement and the like, conversation health data analysis and the like are realized;

5. according to the invention, the block chain has the characteristics of 'decentralized', 'public transparency', 'difficult tampering' and the like, so that the health medical data are linked, the sharing of the health data is enhanced, and the real credible effect is achieved; meanwhile, by means of the characteristics of traceability and whole-course trace retention of the block chain, the user historical data, namely the health data, can be subjected to time axis analysis, so that the tracking analysis effect of the user data is ensured; the technical effect that a user, namely a guardian, can check the health data report by using the mobile phone app anytime and anywhere and provide health basis for a doctor is achieved.

Drawings

Other features, objects and advantages of the invention will become more apparent upon reading of the detailed description of non-limiting embodiments with reference to the following drawings:

FIG. 1 is a schematic diagram of an intelligent intent recognition analysis system based on blockchain technology;

FIG. 2 is a flow chart of an intelligent intent recognition analysis method based on a blockchain technique.

Detailed Description

The present invention will be described in detail with reference to specific examples. The following examples will assist those skilled in the art in further understanding the invention, but are not intended to limit the invention in any way. It should be noted that it would be obvious to those skilled in the art that various changes and modifications can be made without departing from the spirit of the invention. All falling within the scope of the present invention.

Example 1

The invention provides an intelligent intention recognition analysis system based on a block chain technology, which comprises:

a data collection module: the data collection module collects voice data, health data and user voiceprint data through intelligent equipment and an application program;

a data extraction module: the data extraction module preprocesses the collected data to obtain preprocessed instruction data, voice voiceprint data, health data and user voiceprint data;

an information collection module: the information collection module uploads the collected health data and user voiceprint data to the block chain, matches the voice voiceprint data with the user voiceprint data on the block chain, and binds and uploads the current voiceprint data and the instruction to the block chain when the extracted voice voiceprint data is matched with the user voiceprint data on the block chain; when the extracted voice voiceprint data are not matched with the user voiceprint data on the block chain, marking the current instruction;

an intention analysis module: and the intention analysis module classifies the matched instruction data and correspondingly processes the instruction data based on the block chain data according to intention classification.

Specifically, the data collection module includes:

the data collection module M1: inputting initial data including user voiceprint data and health data through intelligent equipment, and collecting voice data when the intelligent equipment is used;

the data collection module M2: uploading the health data and the user voiceprint data to a blockchain through an application program;

the data collection module M3: health data are collected through a monitoring function on the intelligent device.

Specifically, the data extraction module includes:

the data extraction module M4: converting voice data collected by intelligent equipment into instruction text and voice voiceprint data;

the data extraction module M5: filtering health data collected by intelligent equipment;

the data extraction module M6: and filtering the health data and the user voiceprint data collected by the application program.

Specifically, in the intent analysis module: classifying the matched instruction data, wherein the instruction data comprises a functional intention, a potential health intention and a meal ordering intention, and performing corresponding processing based on the block chain data according to intention classification;

when the matched instruction text is a potential health intention, comparing the matched instruction text with the existing health data on the block chain, comparing the existing suggestions on the block chain with the health data of the current user to obtain a recommendation message, and feeding the recommendation message back to an intention analysis module, wherein the intention analysis module feeds back the recommendation message through a loudspeaker;

when the matched instruction text is the food ordering intention, transmitting the current instruction text to the block chain, selecting the current edible diet recommendation according to the existing food rule, time and physical data analysis on the block chain, and feeding back the current edible diet recommendation to the intention analysis module, wherein the intention analysis module feeds back the current edible diet recommendation through a loudspeaker;

when the instruction text is the functional intention, the direct feedback is stated through the intelligent equipment voice.

Specifically, in the intention analysis module, the matched instruction text is uploaded to the block chain when judged to be health-related data by the intention analysis module.

Specifically, the data analysis module analyzes the executed instruction;

when the current instruction is executed, comparing the health data on the block chain with the executed instruction, generating a health report, encrypting the health report and uploading the health report to the chain; when the current instruction cannot be executed, the data analysis module judges the current data as a dangerous signal, pushes the dangerous signal to a preset object through the timely warning module, and records the chain.

Specifically, in the data analysis module: combining the health data uploaded by the intelligent equipment on the block chain with the health related data stored in the block chain to generate a health report;

and classifying the instructions to be completed in the future, and submitting the instructions to a module to be fed back.

Specifically, the intelligent device comprises an intelligent sound box and an intelligent bracelet;

the intelligent sound box registers a personal account in a voiceprint and public and private key symmetric mode, and binds the identity of an intelligent sound box owner;

the intelligent sound box records voiceprint and health data when recording initial data, collects sound data during daily communication and collects sound data when using a sound box network for communication;

health data including heart rate and blood oxygen of a user are collected through a monitoring function on the smart bracelet, and the collected data are uploaded to a block chain;

inputting personal information, health data and associated person information in the application program; entering personal information includes: user name, nickname, and voiceprint.

The invention provides an intelligent intention recognition analysis method based on a block chain technology, which comprises the following steps:

step S1: collecting voice data, health data and user voiceprint data by using intelligent equipment and an application program;

step S2: preprocessing the collected data to obtain preprocessed instruction data, voice print data, health data and user voice print data;

step S3: uploading the collected health data and user voiceprint data to a block chain, matching the voice voiceprint data with the user voiceprint data on the block chain, and binding and uploading the current voiceprint data and instructions to the block chain when the extracted voice voiceprint data is matched with the user voiceprint data on the block chain; when the extracted voice voiceprint data are not matched with the user voiceprint data on the block chain, marking the current instruction;

step S4: and classifying the preprocessed instruction data, and performing corresponding processing based on the block chain data according to intention classification.

Specifically, the step S1 includes:

step S1.1: inputting initial data including user voiceprint data and health data through intelligent equipment, and collecting voice data when the intelligent equipment is used;

step S1.2: uploading the health data and the user voiceprint data to a blockchain through an application program;

step S1.3: collecting health data through a monitoring function on intelligent equipment;

the step S2 includes:

step S2.1: converting voice data collected by intelligent equipment into instruction text and voice voiceprint data;

step S2.2: filtering health data collected by intelligent equipment;

step S2.3: filtering the health data and the user voiceprint data collected by the application program;

in step S3, performing classification processing on the matched instruction data, including functional intent, potential health intent and meal ordering intent, and performing corresponding processing based on the blockchain data according to intent classification;

when the matched instruction text is a potential health intention, comparing the matched instruction text with the existing health data on the block chain, comparing the existing suggestions on the block chain with the health data of the current user to obtain a recommendation message, and feeding the recommendation message back to an intention analysis module, wherein the intention analysis module feeds back the recommendation message through a loudspeaker;

when the matched instruction text is the food ordering intention, transmitting the current instruction text to the block chain, selecting the current edible diet recommendation according to the existing food rule, time and physical data analysis on the block chain, and feeding back the current edible diet recommendation to the intention analysis module, wherein the intention analysis module feeds back the current edible diet recommendation through a loudspeaker;

when the instruction text is the functional intention, directly feeding back through the voice statement of the intelligent equipment;

in step S3, the matched instruction text is uploaded to the block chain when judged to be health related data by the intention analysis module;

wherein, the executed instruction is analyzed;

when the current instruction is executed, comparing the health data on the block chain with the executed instruction, generating a health report, encrypting the health report and uploading the health report to the chain; when the current instruction cannot be executed, the data analysis module judges the current data as a dangerous signal, pushes the dangerous signal to a preset object through the timely warning module, and records the chain;

the health report is generated by combining health data uploaded by intelligent equipment on the block chain and health related data stored in the block chain;

classifying the instructions to be completed in the future, and submitting the instructions to a module to be fed back;

the intelligent equipment comprises an intelligent sound box and an intelligent bracelet;

the intelligent sound box registers a personal account in a voiceprint and public and private key symmetric mode, and binds the identity of an intelligent sound box owner;

the intelligent sound box records voiceprint and health data when recording initial data, collects sound data during daily communication and collects sound data when using a sound box network for communication;

health data including heart rate and blood oxygen of a user are collected through a monitoring function on the smart bracelet, and the collected data are uploaded to a block chain;

inputting personal information, health data and associated person information in the application program; entering personal information includes: user name, nickname, and voiceprint.

Example 2

Example 2 is a preferred example of example 1

The invention combines with a voiceprint recognition system, binds the identity of the owner of the intelligent sound box according to the corresponding account number of the voiceprint, and simultaneously can analyze the user of the sound box when the data is linked up, synchronously uploads the voiceprint data of the user and binds the voiceprint data with the user authority generated in advance as the identifier of the user block chain, thereby providing double protection for the data and avoiding the problem of non-self questioning from mixing into the health data of the current owner when the data is linked up each time. When the data is called, the privacy of the current health, question data and the like can be protected doubly by combining a private key generated by the voiceprint.

The data of the user on the chain before is all plaintext display, the user data and the privacy data of the invention are all encrypted chains, and the user privacy is sufficiently protected.

The intelligent intention recognition system can be combined with city meal ordering service, can analyze according to specific body data collected in advance and on-chain medical health cloud data when a user orders a meal, recommends proper food, and can give corresponding warm prompt when the user orders a meal and is not suitable for body health. Ensuring the diet health of the user. Meanwhile, after the user orders, the user data can be encrypted and then linked up, food intake and nutrition intake are counted, and powerful help is provided for analyzing the health of the user in the future.

In order to enable the intelligent intention recognition system to be better integrated into life, the system supports the function of chatting, and a user can realize functions of chatting, booking an alarm clock and the like with the intelligent loudspeaker box. When the user chats with the intelligent loudspeaker box, the loudspeaker box can encrypt the chatting content and carry out chaining processing. The intention recognition system can perform psychological analysis according to keywords of the current user and the health cloud data on the chain while performing chain linking processing, and can report the caretaker through the collected chat information if the user has a psychological health problem, so that early warning is timely given to the caretaker. Meanwhile, the current user intention can be judged according to different psychological states and linked data, and measures such as different chatting modes, recommendation of relevant light music, books, news, jokes and the like are adopted according to the intention to actively assist the psychological construction of the user. Thereby achieving the effects of timely feedback and pacification of the user and the like.

The intelligent intention identification system can feed back information such as user behavior, diet, intention and the like to the uplink in time. Besides the intention actively proposed by the user, the user can synchronize the health data (heart rate, steps, blood oxygen and the like) of the intelligent bracelet worn on the person into the intention identification system. The system compares publicly-approved medical data and real data of some patients with real-time data of a current user at the cloud end, and links up and seals health data after analyzing, processing and encrypting the health data while generating a corresponding scheme, so that data with reference values, such as daily health reports, food intake and the like, are counted. The method is favorable for medical health preservation in the future.

In addition to timely accessible transactions, future transactions where the user speaks a desired item that cannot be completed may also be recorded by the intent recognition system, such as where the user desires a child to accompany, wishes to own those items, wishes to travel out, and so forth. And after recording, the data can be fed back to children viewed from the far end in time, so that the children can be satisfied in the future.

The invention provides an intelligent intention recognition analysis method based on a block chain technology, as shown in fig. 2, comprising the following steps:

step 1: and collecting user instructions.

The user finishes command collection through the intelligent sound box, and the intelligent sound box can receive the voice data of the user and splits the voice data into voiceprint and text in the background. Finally, the specific text content (keywords in a sentence) will be defined as instructions by the speech recognition system. The content intended to be identified is the "instruction".

The collected speech is processed into valid "instructions" that are sent to the intent recognition system. Such as: the absence of keyword instructions is directly filtered out by the collection system because such data is not within the processing scope of the system. This avoids wasting resources by sending invalid data into the intent recognition system.

A voiceprint is calculated from the collected speech and the speech is converted to text.

Each person's voice is unique, and the instruction collection system can extract a set of acoustic description parameters (vectors) from the voice signal and algorithmically generate and extract voice features. If the user is not used by the user, the mark can be carried out, the mark is transmitted to the intention recognition system along with a text instruction, and when the subsequent health problem needs to be subjected to uplink analysis, the intention recognition system can ask a question according to the current 'non-self mark': "ask about whether the health question is a question that occurs on the user of the user? "

The step 1 comprises the following steps:

step 1.1: and registering the personal account in a voiceprint and public and private key symmetric mode.

Step 1.2: the user can directly upload the health data to the chain through the application program, directly upload the data through the intelligent bracelet monitoring, and upload the data through the voice command.

The application program is used by the user to directly upload the physical examination report or the self disease history to the chain by using the app. Bracelet monitoring is a concrete proof that can directly embody the physical condition of a user, such as: heart rate, blood oxygen, etc. Or directly chaining the chain and using the chain for later health analysis of users. The voice command is uploaded to the chain only after being judged as health related data by the intention recognition system. When the user has obstacles or inconvenience in operating the app, the user can select to ask a question by the sound box, and the uplink is recorded in a mode expressed by the user.

Step 1.3: when the voice command uploads data, the voice command firstly passes through the sound box to carry out command collection and voiceprint splitting. The extracted voiceprint can be matched according to the data on the chain, and when the user uses the voiceprint, the voiceprint data and the command can be bound and uploaded to the chain. If the voiceprint matching fails to prove that the user is not personally used, the collection system asks the user, for example, the user is helped to ask the user himself and the part of the content is marked when the health problem is involved, and the part of the content is used as one of the health data analysis elements after the user needs to chain.

The chain voiceprint data is the encrypted voiceprint result retained on the block chain when the user registers, and the method can judge whether the sound is made by the user.

Step 2: and after the user command collection is completed, entering an intention recognition system.

After entering the intention recognition system, the intention recognition system first classifies the intentions into three categories, namely, functional intentions, potential health intentions and meal ordering intentions. After classification is completed, the intention recognition system assigns different processing modes to the user according to different intentions. Such as: and if the user wants to order, directly calling the ordering module, and performing health analysis and chain linking at the later stage. Such as: the user does not invoke the health analysis and chaining functions only by considering alarm clock reminders.

The step 2 comprises the following steps:

step 2.1: the intent recognition system classifies the collected instructions as functional intent, potential health intent, and meal order intent, respectively. The functional intentions are simple intentions of ordering alarm clocks, reminding items and the like without chaining health analysis. The potential health intentions are questions of the user, such as: poor mood, a painful belly, what is meaningful to leave in the world, etc., these potential health intentions are marked specifically and retain voiceprints. Order intent refers to a user's conventional order instructions.

Step 2.2: the functional intentions are simple intentions of ordering alarm clocks, reminding items and the like without chaining health analysis. The intention is only to process data which does not need to be compared, and tasks which need to be executed are stored on a temporary database. And directly feeding back to the user in a voice statement mode of the sound box. And the task is destroyed after being executed. And the waste of the memory is reduced.

Step 2.3: the potential health intentions are questions of the user, such as: poor mood, a painful belly, what is meaningful to leave in the world, etc., these potential health intentions are marked specifically and retain voiceprints.

Step 2.4: the order intention refers to a conventional order instruction of the user, and the intention and the voice print are recorded when the user orders.

And step 3: the intent recognition system, after processing the timely task, passes data related to the health analysis down to the blockchain portion for analysis.

The step 3 comprises the following steps:

step 3.1: the problem of potential health intention can be compared with existing health cloud data on the block chain, when the data belong to a mental health problem, the block chain in the physical health problem can be compared with the health data of the current user according to existing suggestions on the chain, the best recommendation information is compared and fed back to the intention identification system, and the intention identification system feeds back to a loudspeaker and directly feeds back to the user. Such as: when a user inquires about the belly pain, the intention recognition system can be combined with the diet at noon, the health data uploaded to the chain by the user before shows that the user has symptoms of chronic enteritis and the like, and the comprehensive consideration is fed back to the user, the symptoms are probably caused by enteritis and spicy food at noon, and the user is advised to drink more water, keep sports and reduce spicy food. If the patient feels severe discomfort, the patient should seek medical advice in time. When the user asks what meaning the user lives in, the intention recognition system can judge that the user possibly has psychological problems at present according to the big data support on the chain, and then finds out the most suitable contents such as relaxing, cheering music, entertainment news and the like from the chain to actively propose, suggest and push the contents to the user.

Step 3.2: when the food ordering intention is transmitted to the block chain system, the system can select the most suitable food recommendation for the current user to eat according to the existing data analysis of the user food rule, time, user physique and the like on the chain and feed back the most suitable food recommendation to the intention recognition system, and the intention recognition system feeds back the most suitable food recommendation to the loudspeaker and directly feeds back the most suitable food recommendation to the user. Ensuring the diet health of the user.

Step 3.3: the intention recognition system can directly use the pre-bound network telephone and the specific user to realize the network call when recognizing that the user intention is the call intention.

And 4, step 4: the intention recognition system delivers the valid health intention analyzed from the blockchain to the data analysis system, and the step 4 comprises the following steps:

if the block chain analysis indicates that the intention is health-related and the health analysis of the user is influenced, the data are transmitted to the data analysis system, and the data analysis system compares the health-related problems with the existing health data to generate a health report. If the intention is not related to the health data of the user and is not the reference condition of the health report of the user, the intention is filtered out and does not participate in the health report analysis of the user.

Step 4.1: the data after being fed back by the block chain and the data after being fed back by the user. The intention recognition system can combine the voiceprint and the health instruction collected in the step 2 and submit the combination to a data analysis system, the data analysis system can integrate the current instruction and the feedback data of the blockchain health cloud, and the necessary health data of the instruction execution success are combined into a book. Meanwhile, effective pathological data can pass through desensitization cochain, and medical staff can read the disease causes and treatment methods for the effective pathological data and then use the effective pathological data as other big data for reference.

Step 4.2: if the ordering data of the user is successfully ordered finally, the intention recognition system feeds back the food information to the block chain system again, then the cloud data support of the block chain system feeds back the food information to the data analysis system, and the data analysis system sorts the nutrition intake into a book according to what kind of influence is generated on the body.

Step 4.3: health information timely transmitted by the wristband can be directly arranged into a health manual through the information identification system

And 5: the data analysis system delivers the registered contents back to the blockchain system for uplink

The step 5 comprises the following steps:

step 5.1: the data analysis system can directly link the sorted health report, and the partial data can be checked and extracted by the user and the apps or the sound boxes in the later period of the children.

Step 5.2: the data analysis system will sort out the emergency information and the information that can not be completed at present. The part of data is directly pushed to the guardian through the information transmission module after the uplink. And performing chain linking processing to ensure timely effectiveness of the record.

Step 5.3: the information transmitted by the bracelet can also be directly sorted into the post-uplink by the information identification system. For later viewing of records.

According to the intelligent intention recognition analysis system based on the block chain technology provided by the invention, as shown in fig. 1, the intelligent intention recognition analysis system comprises:

the voice data collection module: can realize collecting user's speech data by the microphone of intelligent audio amplifier, include:

1. the user enters a voiceprint upon registration.

2. The user enters voiceprint and health data when entering the initial data.

3. The user collects voice data during daily communication with the sound box.

4. And collecting voice data when the speaker network is used for communication.

Bracelet data collection module: health data such as the user's rhythm of the heart, blood oxygen can be collected by the monitoring function on the intelligent bracelet. This partial data can be uploaded directly onto the chain for later health data analysis.

An application collection module: the module provides a data entry system for collecting apps, comprising:

1. when the user registers, personal information such as a user name, a nickname, a voiceprint and the like is input through app registration.

2. The user inputs the basic health information after the registration, and inputs the basic health data, the previous medical history, the height and the weight and the like.

3. And the guardian registers the relevant information such as a mobile phone number, an emergency contact person and the like when registering the account number.

A data extraction module: the data extraction module is the core of the whole instruction collection system. The mobile phone is responsible for filtering a voice data collection module, a bracelet data collection module and an application program collection module. The specific mode is as follows:

1. and converting the voice and tone collected by the voice data collection module into a text keyword instruction and a voiceprint.

2. And filtering data of partial abnormal data (such as in a non-wearing state) of the bracelet data collection module.

3. And filtering the data which is filled in by the application program collection module and has no reference value or is not in accordance with the specification.

An intention analysis module: the intention analysis module is a core module of the system, and can be combined with other modules to process and feed back all data, and the specific method is as follows:

1. and the instructions are classified by combining with a data extraction module, and the classified instructions are functional intention, potential health intention, meal ordering intention and conversation intention.

The functional intentions are as follows: the intention of timely finishing such as playing music, setting an alarm clock, reminding items and the like.

Potential health intents are as follows: i today have stomach discomfort, what sense to live, how to handle a leg break, etc.

The ordering intentions are as follows: to help me order a meat package in a pot.

The conversation intentions are as follows: call XXX.

2. And combining the block chain module to obtain a solution of the user intention.

Such as: the intent recognition system compares the health data of the current instruction on the blockchain to find the best disease cause and treatment plan suggestion on the chain.

The function intention can be combined with a database to realize functions of alarm clock customization, music recommendation, weather inquiry and the like for timely inquiry and feedback.

When the user orders the food, the intention recognition system compares the food with the user health data and the big data diet health map on the block chain, when the user clicks food harmful to the body (if the user clicks a pachyrhizus angulatus with diabetes), the intention recognition system prepares relevant instructions of 'the food is harmful to your body', and the like, and the relevant instructions are fed back to the user through the timely feedback module.

The intention recognition system can directly dial the call when the call is intended, and can set in advance whether to allow the call record to be monitored, if the call is allowed, the intention recognition system can start the voice data collection module and the data extraction module, so that the two modules intervene to jointly extract the health related content to be transmitted to the data analysis module in the later period, and if the health analysis is facilitated, the health report and the chain can be inserted for the later-period user to check.

3. The resulting solution is fed back to the customer in conjunction with the immediate feedback module and the blockchain module.

4. And selecting a proper diet for the user by combining the intelligent ordering module.

After the intention feedback system is combined with the block chain to obtain the diet desired by the user, the intelligent ordering module can be communicated with a third-party ordering platform or other catering units to help the user to order the diet.

A timely feedback module: the immediate feedback module can be understood as a speaker that broadcasts the solutions to the instructions summarized by the feedback system to the user.

Such as: the potential health intention and the meal ordering intention can be completely solved after being compared by combining the block chain, and after the solutions are obtained, the intention analysis module can directly submit the solutions to the instant feedback module, and the instant feedback solution is broadcasted to the user in a voice mode. The intelligent ordering module: the intelligent ordering module is responsible for ordering and paying for the diets confirmed by the user.

A block chain module: the blockchain module utilizes the characteristics of 'decentralized', 'open transparency', 'difficult to tamper', 'traceability' and the like of the blockchain to combine with the intention analysis module to complete a plurality of functions:

storage of user health data: health data uploaded during user registration is encrypted and stored on the chain

Storage of basic medical data: recognized medical data is stored on the chain

Health data storage daily used by a user: statement of health data generated by the user during daily use is stored in the chain, and can provide medical data reference for other users and also can be used for inquiring the statement of health for the user in the future.

And (3) carrying out data comparison analysis by combining an intention feedback system: the user's instructions are compared with the medical data on the blockchain to provide the best solution for the intention recognition system.

And finishing the health data uplink by combining an information collection module: the user links the health data when the user completes registration and fills in the basic health information. The health report form analysis method is used for health report form analysis.

And (3) completing a health data report by combining a data analysis module: the data analysis module compares the health related content of the current instruction with the on-chain data, and accordingly generates a health report.

Collect the food information of intelligence module of ordering and accomplish the cochain: and linking the collected information for confirming the ordering by the user, and generating a health report by combining a data analysis module.

A data analysis module: the data analysis module is mainly responsible for sorting and analyzing the instructions finished by the user. The concrete contents are as follows:

and comparing the health data on the blockchain with the instructions executed by the user, generating a specific health report which is convenient to view, encrypting the health report and uploading the health report to the chain. This portion of data is available for viewing by the user or other authorized user at any time.

When the instructions which cannot be processed are met, such as abnormal heart rate change of the user, the data can be judged to be dangerous signals by the data analysis module, and the dangerous signals are directly pushed to a guardian through the timely alarm module and are recorded for chain winding.

Health data uploaded to the block chain by the bracelet are extracted by the data analysis module and combined with other health related data to generate a health report.

The future instructions to be completed are processed and classified by the data analysis module and delivered to the module to be fed back. Such as: "want to go to three travels". The command is received and transmitted to the system to be fed back.

And comparing the health data on the block chain with the health data uploaded by the user, generating a specific health report convenient to view, encrypting the health report and uploading the health report to the chain.

And a timely warning module: the module can directly inform the guardian of the 'dangerous data' which is transmitted by the data analysis module and needs to be informed immediately through the app or the internet phone. The aim of protecting the user from finding danger all the time is achieved.

A module to be fed back: the module can reserve the 'expectation' information which is transmitted by the data analysis module and does not need to inform the guardian immediately, the information is stored in the database, and the guardian can check the appeal through the app at any time. Allowing the user's expectations to be met in the future.

Those skilled in the art will appreciate that, in addition to implementing the systems, apparatus, and various modules thereof provided by the present invention in purely computer readable program code, the same procedures can be implemented entirely by logically programming method steps such that the systems, apparatus, and various modules thereof are provided in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers and the like. Therefore, the system, the device and the modules thereof provided by the present invention can be considered as a hardware component, and the modules included in the system, the device and the modules thereof for implementing various programs can also be considered as structures in the hardware component; modules for performing various functions may also be considered to be both software programs for performing the methods and structures within hardware components.

The foregoing description of specific embodiments of the present invention has been presented. It is to be understood that the present invention is not limited to the specific embodiments described above, and that various changes or modifications may be made by one skilled in the art within the scope of the appended claims without departing from the spirit of the invention. The embodiments and features of the embodiments of the present application may be combined with each other arbitrarily without conflict.