JP2017033581A - Health management server, health management server control method, and health management program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a health management server capable of creating an advice message which is more suitable for a client while achieving standardization and efficiency in the case of creating an advice message of an expert.SOLUTION: The health management server includes: a storage part for storing the health management target person information and question information and answer information of a health management target person who has a terminal; a reception part for receiving image information or message information transmitted from the terminal; an analysis part for analyzing language information on the basis of the received message information, and for acquiring the question information from the analyzed language information; a generation part for extracting the answer information corresponding to the acquired question information, and for generating a sentence example on the basis of the extracted answer information; an evaluation part for evaluating the extracted answer information on the basis of probability of the answer information; a correction part for correcting the sentence example and evaluation on the basis of the health management target person information; and a transmission part for transmitting the corrected sentence example and evaluation to the terminal.SELECTED DRAWING: Figure 2

Description

  The present invention relates to a health management server, a health management server control method, and a health management program, and more particularly to a health management server, a health management server control method, and a health management program that analyze language information.

  In recent years, health guidance services that receive expert guidance have become widespread in response to increasing health awareness such as diet and prevention of metabolic syndrome. In addition, with the shift to IT, there are increasing cases of receiving health guidance services on-line without direct contact between experts and clients.

  In such a health guidance service, the expert checks the meals taken by the client and the current health status from the online terminal, creates a message for guidance based on the confirmation result, and sends it to the client's online terminal. I need to send a message.

  As a technique for creating and transmitting such a message, for example, in Patent Document 1, in order for a doctor to send a health management advice message to a patient who is remote, the database server device receives biological information from the patient, A technique for generating a health management template message corresponding to received biological information is disclosed. Further, this document discloses that a doctor can acquire a generated template message and edit it as an advice message.

  Further, in Patent Document 2, the pointing device graphically displays any one of the measured biological signal, the feature pattern unit, the neutral net output, the movement direction estimation result, or the movement amount as the processing status of each unit during the point control. Discloses a technology including a feedback unit to be presented to the user.

JP 2004-321373 A JP 2005-011037 A

  However, in the invention described in Patent Document 1, the database server device needs to create an advice message from scratch for each patient by generating a template message for health management corresponding to biological information. Although standardization and efficiency have been achieved, template messages are generated from pre-saved message texts, so it is necessary for people to create and save template message texts for their variations in order to improve efficiency. There is an unsolved problem. In addition, the database device has a problem that it cannot generate a template message that is most suitable for the patient according to the characteristics of the biometric information for each patient and changes in the biometric information.

  In the invention described in Patent Document 2, it is possible to improve the direction estimation accuracy and the trajectory error by feeding back to the user, and thus to improve the operability. There is a problem that it does not contribute to the accuracy improvement of the device itself, and the usability of the pointing device itself does not change.

  Therefore, the present invention has been made in view of the above problems, and in creating and transmitting a message for guidance, it is possible to create an expert advice message from the client's biometric information, diet, health status, etc. Provide a server that can be standardized and streamlined, and that can provide advice that best suits the health status of the target client at the time, and improves the accuracy of the health guidance service itself through feedback. The purpose is to plan.

  A health management server according to the present invention is a health management server connected to a terminal via a network, and stores a health management target person information, question information, and answer information of a health management target person having the terminal; A receiving unit that receives image information or message information transmitted from a terminal, an analysis unit that analyzes language information based on the received message information, and acquires question information from the analyzed language information, and acquired question information And a generation unit that generates a sentence example based on the extracted answer information, an evaluation unit that evaluates the extracted answer information with a certainty factor indicating the certainty of the answer information, a sentence example, and A correction unit that corrects the evaluation based on the health management subject information, and a transmission unit that transmits the corrected sentence example and the evaluation to the terminal.

  Furthermore, the health management server according to the present invention may be connected to a terminal via a network, and the corrected sentence example and the evaluation may be displayed on the display unit of the terminal.

  Furthermore, in the health management server according to the present invention, the transmission unit transmits advice information including sentence examples and evaluations selected or selected and edited by the expert on the display unit to the terminal. Also good.

  Further, in the health management server according to the invention, the evaluation unit determines whether the evaluation confidence level is high or low, and corrects the sentence based on the determination result or generates an alert message to the expert. The correction unit corrects the sentence example based on the determination result and the answer information for avoidance when the evaluation unit selects to correct the sentence example. And a transmission part may notify an alert message with respect to an expert's terminal, when an evaluation part chooses notifying an alert message to an expert.

  Further, in the health management server according to the invention, the receiving unit receives expert sentence example selection information, health management target person behavior selection information, behavior execution information, and biological log information from the terminal, and the evaluation unit includes: The received sentence example selection information, action selection information, action execution information and biometric log information, or combinations based on the time stamps of these information, and the combined sentence example selection information, action selection information, action execution information and biometrics You may evaluate with respect to the reply information linked | related with log information.

  Further, in the health management server according to the invention, the image information is taken by a terminal, and the analysis unit calculates a feature of the taken image information, and based on the feature and the question information, language information May be analyzed.

  Furthermore, in the health management server according to the invention, the health management server includes a calculation unit that calculates the usage frequency of the response information for each health management subject, and the storage unit uses the past usage history and usage frequency of the response information of the health management subject. The history information may be stored, and the correction unit may correct the sentence example based on the usage history information.

  Furthermore, in the health management server according to the invention, the receiving unit receives image information or message information transmitted from a plurality of terminals grouped in advance, and the evaluation unit is the same as the sentence example transmitted to the terminal. Alternatively, a similar or opposite sentence example identifies a terminal in a group transmitted within a predetermined time, determines a health care target person who owns the identified terminal as a comparison target person, and the correction unit determines the sentence example and evaluation On the other hand, it may be corrected based on the comparison target person information.

  A health management server control method according to the present invention is a health management control method connected to a terminal via a network, and stores health management subject information, question information, and answer information of a health management subject having a terminal. A storage step; a reception step for receiving image information or message information transmitted from the terminal; an analysis step for analyzing language information based on the received message information and acquiring question information from the analyzed language information; A step of extracting answer information corresponding to the extracted question information, generating a sentence example based on the extracted answer information, and an evaluation step of evaluating the extracted answer information with a certainty factor indicating the certainty of the answer information A correction step for correcting the sentence example and the evaluation based on the health management subject information, and the corrected sentence example and the evaluation are It is intended to include a transmission step of transmitting relative.

  A health management program according to the present invention is a health management program connected to a terminal via a network, and stores in a computer the health management target person information, question information, and answer information of the health management target person having the terminal. A storage function, a reception function for receiving image information or message information transmitted from the terminal, an analysis function for analyzing language information based on the received message information, and acquiring question information from the analyzed language information, and acquisition A function for extracting answer information corresponding to the extracted question information, generating a sentence example based on the extracted answer information, and an evaluation function for evaluating the extracted answer information with a certainty factor indicating the certainty of the answer information And a correction function for correcting the sentence examples and evaluation based on the health management subject information, and transmitting the corrected sentence examples and evaluation to the terminal. That is intended to include a transmission function.

  The health management server, the health management server control method, and the health management program according to the present invention respond to inquiries from clients (health management subjects) or experts based on the accumulated best practices in creating expert advice messages. A sentence example to be answered can be generated, and the generated sentence example can be customized with the characteristics of the client (biological information, meal, health condition at the time, etc.) and transmitted to the expert. Therefore, it is possible to standardize and improve efficiency in creating and transmitting an advice message for an expert, and it is possible to provide guidance more suitable for the client.

  In addition, the health management server, health management server control method, and health management program according to the present invention can learn by feedback of selection results and execution results of the client's advice and changes in the living body, so that accuracy improvement is automatically performed. A health guidance service can be provided.

System diagram showing system configuration of health management system Block diagram showing functional configuration of health management server and terminal Schematic diagram for explaining the interaction between a health manager and a health management server or expert Schematic diagram for explaining feedback to the health management server Schematic diagram showing a combination example of action selection information, action execution information, and biolog information Data conceptual diagram showing an example of the data structure of health management subject information and learning information Schematic diagram showing an example of correction of evaluation of response information Schematic diagram showing an example of the relationship between message information, language information, question information, answer information, and sentence examples Flow chart showing server operation Schematic diagram showing an example of genetic testing algorithm Schematic diagram showing an example of blood test algorithm Schematic diagram showing an example of content distribution algorithm

Hereinafter, an embodiment of the present invention will be described with reference to the drawings.
<Overview>
FIG. 1 is a system diagram showing a system configuration of a health management system.

  As shown in FIG. 1, the health management system includes a health management server 100 and a plurality of user terminals 200 and 300. The health management server 100 is connected to the user terminal 200 and the user terminal 300 via the network 400. In FIG. 1, only two user terminals are shown for ease of explanation, but it goes without saying that there may be more user terminals. In addition, specific devices of both the user terminal 200 and the user terminal 300 are not limited to the portable terminal and the personal computer as illustrated, and may be, for example, a smartphone, a tablet terminal, a personal computer, and other electronic devices. Good. Furthermore, the specific device of the user terminal 200 may be a wearable device, and the wearable device stores the heart rate and pulse rate, blood pressure, number of steps and activity amount, posture, position and movement, and position information of the health care target person. Includes equipment to be measured.

  Here, for example, it is assumed that the user terminal 200 indicates a terminal held by a health care target person, and the user terminal 300 indicates a terminal held by an expert. Then, the health management target person himself / herself sends a question message to the health management server 100 inquiring about a meal to be ingested at present.

  When the server 100 receives the question message, the server 100 analyzes the question message, generates a sentence example from the corresponding answer information, and evaluates the answer information. The sentence example and the evaluation are directly transmitted to the user terminal 200 as an advice message or displayed on the user terminal 300. In the latter case, the expert selects a sentence example to be transmitted to the health care target person from the displayed sentence example and evaluation, and edits it as necessary. The server 100 transmits to the user terminal 300 the sentence example and evaluation selected or edited by the expert as an advice message.

  Accordingly, the server 100 transmits an advice message or an advice message selected and edited by the expert at the user terminal 300 to the user terminal 200 via the server 100, so that the health management target person can perform health management. Guidance. In addition, the said instruction | indication may be performed for every health manager individually, and health management object person may be grouped and you may instruct | indicate in the said group unit. Here, grouping refers to grouping for performing group guidance for a plurality of health care subjects.

<Configuration>
Hereinafter, configurations of the server 100, the user terminal 200, and the user terminal 300 will be described.
FIG. 2 is a block diagram illustrating functional configurations of the server 100, the user terminal 200, and the user terminal 300.
As illustrated in FIG. 2, the server 100 includes a reception unit 110, an analysis unit 120, a generation unit 130, an evaluation unit 140, a correction unit 150, a storage unit 160, and a transmission unit 170. The

  The receiving unit 110 receives image information and message information transmitted from the user terminal 200 or the user terminal 300 via the network 400. Communication in the transmission / reception may be either wired or wireless, and any communication protocol may be used as long as mutual communication can be performed.

  The analysis unit 120 analyzes the language information based on the question information stored in the storage unit 160 from the message information sent from the user terminal 200 or the user terminal 300, or the analyzed language information or the analyzed language It has a function to search and acquire corresponding question information using the information and the health management subject information of the target health management subject as a key. Here, the analysis of linguistic information specifically means that the text of the message information is converted into a more formal expression. For example, it is performed using natural language processing techniques such as morphological analysis and syntax analysis. Just do it. Here, the question information refers to a set of words or the like broken down into formal elements that can be recognized by a computer as input information for deriving answer information from language information as shown in FIG.

  The analysis unit 120 calculates not only the message information received by the reception unit 110 but also the characteristics of the received image information, and analyzes the language information based on the calculated characteristics, the question information, and the health management target person information. It also has a function.

  Furthermore, the analysis unit 120 has a function of analyzing the characteristics of the living body based on the biological information and the biological log information of the health care target person and calculating the characteristic value of the living body. Here, the biological feature value and the behavior tendency value are a value calculated by trend estimation or the like from biological information, biological log information and behavior information, and a self-reported biological feature value or behavior interviewed in a lifestyle questionnaire, respectively. It means the trend value. In the calculation, analysis methods such as trend estimation, factor analysis, correlation analysis, and statistical analysis may be used.

  The generation unit 130 extracts answer information that is a sentence example candidate using the language information analyzed by the analysis unit 120. Here, the answer information means, as a best practice corresponding to the question information, association with the question information, answer contents, and past evaluation of the answer information. In the extraction, for example, a data mining technique such as extraction of analog patterns and classification may be used. A hypothesis is generated by inference from the extracted answer information, and the hypothesis is verified from the question information and the answer information by finding a ground for supporting the hypothesis or a disproval for negating it. In the inference, for example, deductive reasoning, inductive reasoning, and stochastic reasoning may be used. Extraction of answer information as candidate sentence examples, generation and verification of hypotheses are simultaneously performed in parallel processing. Integrate verification results processed in parallel and generate example sentences

  The evaluation unit 140 evaluates the certainty of the answer information that is a candidate sentence example extracted by the generation unit 130 based on the evidence that the generation unit 130 finds and the past evaluation of the answer information. Here, confidence is the degree to which the sentence example is shown, and it is an index that indicates the importance of the association rule in data mining. When an event occurs in which question information is input, the corresponding answer information is selected. The rate at which events that occur or are acted on occur.

  In addition, as shown in FIG. 3, the evaluation unit 140 determines whether the certainty level is higher or lower than a predetermined level. Answer information is extracted, and the extracted answer information is overwritten and corrected, or an alert message including the sent message information, sentence examples, and evaluation is transmitted to the user terminal 200 of the expert.

  Here, reply information for avoidance is reply information including reply contents indicating that the message contents could not be understood or an appropriate reply was not found, regardless of the contents of the sent message information. Say. For example, “I ’m sorry. I ’m sorry to hear that there ’s no appropriate answer to the content of the inquiry, so please call me again.” Or “I ’m sorry. In order to understand the content of the inquiry accurately, Please refer to the content of the inquiry again in specific words as much as possible. Please contact us again. "

  Further, as shown in FIG. 4, the evaluation unit 140 is a health management target included in the expert sentence example selection information received from the user terminal 300 by the receiving unit 110 and the message information and image information received from the user terminal 200. The answer information is evaluated based on the person's action selection information, action execution information, and biological log information. Specifically, as shown in FIG. 5, for example, the evaluation unit 140 performs a combination based on sentence example selection information, action selection information, action execution information, biometric log information time stamps, and the like. , Evaluate. In addition to the combination, the information may be evaluated alone.

  The combination includes, for each expert and health care target person, action execution information that holds a time stamp within a predetermined time, based on the sentence selection information and the time stamp of action selection information corresponding to the sentence selection information. Biolog information can be combined. Here, the predetermined time may be any time such as 15 minutes, 30 minutes, 1 hour, 3 hours or 5 hours, but it should be set according to the lifestyle such as eating and exercising. Is preferred. At this time, if the corresponding action information and biological log information cannot be received, it can be handled as no evaluation or negative evaluation.

  The response information associated with the combined sentence example selection information, action selection information, action execution information, and biometric log information has a high degree of certainty initially evaluated by the evaluation unit 140, is being executed according to advice, or the living body is a curriculum Response information that has changed toward the set target in the information has a higher evaluation as being more suitable, and evaluation information that has a lower degree of certainty, is not being executed as advice, or does not change the living body, has a lower appropriateness Can be fed back based on the objective facts.

  For example, as shown in FIG. 5, as shown in FIG. 5, when the recommendation rank, that is, the advice with the highest certainty is selected, and the action is performed according to the advice, as a result, the living body changes toward the set target, Feedback with weight 1 as the evaluation of points, acted as advised, and as a result, there was a change in the living body toward the setting target, but if advice that is not the advice with the highest certainty is selected, compare with the former Feedback with a weight of 0.8 can be given as an evaluation of 80 points of low evaluation. In addition, what is necessary is just to set these combination patterns and the evaluation value for every said pattern suitably.

  Further, when the user terminal 200 of the health care target person is grouped, the evaluation unit 140 also performs the same grouping in which a sentence example that is the same as, similar to, or opposite to the sentence example transmitted to the user terminal 200 is transmitted within a predetermined time. A health care target person who owns a terminal that is used is determined as a comparison target person. Here, the comparison target person is another health management target person who belongs to the same group and needs the same or similar guidance for comparison as a rival, or a target. A person who is subject to health care who needs completely conflicting instruction for comparison as a person or a teacher. Specifically, another health care target person who transmits the same, similar, or opposite sentence examples within a predetermined time is determined as a comparison target person.

  The correction unit 150 corrects the ranking based on the certainty degree evaluated by the evaluation unit 140 with the health management target person information and the comparison target person information. Specifically, the evaluation is corrected by correction values based on biometric information, behavior information, biolog information, biometric feature values, behavior tendency values, etc. of the subject of health rights, and ranking is based on the corrected evaluation. Or to correct the sentence example based on the biological information, behavior information, biological log information, and comparison target person information of the health management target person. Details of the correction processing will be described in <Operation flow> described later. The health management subject information is shown in <Data> described later.

  The calculation unit 155 has a function of calculating the usage frequency of the sentence example. Specifically, the calculation unit 155 counts up with the subject sentence example selected by the expert as a trigger. The calculation unit 155 may count up the use frequency for each health care curriculum of the health care target person and reset the curriculum when the curriculum is completed.

  The storage unit 160 stores health management subject information, question information, and answer information. The storage unit 150 is typically realized by various recording media such as a hard disk drive (HDD), a solid state drive (SSD), and a flash memory.

  The transmission unit 170 transmits the sentence examples and evaluations generated and corrected by the server 100 as advice messages, or the sentence examples and evaluations selected and edited by experts as advice messages, to the user terminal 200. Further, the server 100 has a function of transmitting the sentence example and evaluation generated and corrected by the server 100 to the user terminal 300. Communication in the transmission is the same as that of the reception unit 110.

Further, as shown in FIG. 2, the user terminal 300 is configured to include a display unit 310.
The display unit 310 communicates with the health management server 100 via the network 400 to display and output the sentence examples and evaluations received from the server 100.

The display unit 310 displays and outputs information related to the comparison target person when the health management target person is grouped and the comparison target person is determined. Specifically, the display is output such as “Mr. XX is the current rival”.
Further, when the sentence example transmitted to the comparison target person within a predetermined time is specified together with the display output of the comparison target person, the display unit 310 is specified when the sentence example is displayed and output. Displays information about the sentence examples. Specifically, a message such as “Sent to rival Mr. XX” is displayed and output directly under the target sentence.
This completes the description of the functional configurations of the health management server 100 and the user terminal 300.

<Data>
Here, in the present embodiment, the data configuration of the health management target person information and the learning information stored in the storage unit 160 is shown in FIG.

  FIG. 6 is a data conceptual diagram illustrating an example of a data configuration of the health management target person information and the learning information stored in the storage unit 160.

  As shown in FIG. 6, the health management target person information is largely composed of biometric information, curriculum information, sentence example usage information, behavior information, and biometric log information.

  The biometric information is composed of information related to the health management subject from both the subjective and objective aspects of the health management subject, such as genetic information, blood information, screening information (height and weight) and lifestyle questionnaire information of the health management subject. Of the information related to the living body of the health management subject, basic information and invariant information can be included.

  The curriculum information is composed of program information selected by the health care target person, course information in which one or more program information is combined, and set target information.

  The sentence usage information is composed of the sentence usage history and the sentence usage frequency. Here, the sentence usage history information includes the sentence examples used in the past for each health care target person and the last sentence example. This refers to either one or both of the last used date and time.

  The behavior information includes meal information (meal time and meal contents), position information and movement information, and login information (login time and login frequency). The behavior information is information regarding the daily activities of the health management target person acquired by the user terminal 200, and includes text data and image data indicating the contents of the meal received from the user terminal 200, and time associated therewith. The meal information is included in the stamp, or the current position information obtained from the GPS data recorded in the user terminal 200 and the movement information obtained from the position information measured at a certain time and the position information measured at the next time. And information such as the login time of the service provided by the health management server 100 and the login frequency within a certain period (one day, week, month, etc.) can be included. Furthermore, the behavior information can include a login access history obtained from the user terminal 200 or the information site of the login destination, and a purchase history in which a health care target person has purchased health foods such as supplements, health appliances, training machines, and the like. Further, the behavior information can include login information and logout information on the user terminal 200 that informs selection or completion of a task on an application for health management such as a diet.

  The biometric log information includes heartbeat and pulse information, blood pressure information, the number of steps, the amount of activity, and posture information. The biometric log information is information that records the biological status and changes of the health management subject. When the user terminal 200 is a wearable device or a terminal that cooperates with the wearable device, the wearable device Information can be measured and recorded information can be included. For example, the biological log information includes sleep (such as REM sleep or non-REM sleep) information that is determined by sensing the movement of the health care target person based on an acceleration sensor built in the user terminal 200.

  As described above, the information on the subject of health management includes the genetic information of the subject health management subject himself / herself, blood information, biometric information such as examination information, curriculum information, sentence usage information, and health management target selected by the health management subject. It is not limited to attribute information such as behavior information and receipt information acquired from the person himself. The health management target person information is a concept including sleep state identification information obtained from the user terminal 200 in addition to the biological and health attribute information provided from the health management target person himself. Furthermore, the health management target person information includes the user terminal 200 of the health management target person such as the purchase history of health foods such as supplements, health appliances, training machines, etc., the login time, and the click rate (CTR) to the advertisement / information website. It is a concept that includes a wide range of information obtained from information sites at login destinations.

  According to the health management server and health management server control method and health management program according to the present invention, these pieces of information are stored in the storage unit 160 and can be learned by feeding back various information. In addition to providing timely content (messages, etc.) to health care subjects, it is possible to provide a health guidance service that can automatically improve accuracy.

The user terminal 200 question information refers to a set of words or the like broken down into formal elements that can be recognized by a computer as input information for deriving answer information from language information as shown in FIG.

  The answer information includes correspondence with question information, answer contents (answer word and index), and past evaluations. The association with the question information only needs to be able to associate the question information with the answer information, and may be a number assigned to the question information in a sequence, for example. The answer contents are composed of answer words and indicators, and the answer words include the contents of recipes, fitness, tasks, restaurants, etc. that are output information for the question, for example, the answer to the consultation or question, or the advice for the inquiry Can do. The index indicates characteristics associated with the answer word, and can include, for example, numerical values indicating characteristics such as task load, efficiency, ease, effect, fashion, and economy included in the answer word. The past evaluation is a set of evaluations that each answer information has received so far. Specifically, the evaluation unit 140 can include confidence statistics and the like calculated so far. Further, the evaluation unit 140 calculates a certainty factor based on the past evaluation.

<Operation>
The operation of the health management server 100 according to this embodiment will be described.
The operation of the health management server 100 will be described with reference to FIG.
FIG. 9 is a flowchart showing the operation of the health management server 100.
The receiving unit 110 receives image information or message information sent from the user terminal 200 of the health care target person (step S11). The receiving unit 110 transmits the received image information or message information to the analyzing unit 120.

  The analysis unit 120 analyzes language information from the received image information or message information, and obtains question information (step S12). For example, as shown in FIG. 8A, for example, when message information inquiring about a recommended task from a health care target person such as “What is recommended for resolving lack of exercise” is received, Natural language is processed, and as language information, as shown in FIG. 8B, it is decomposed into morphemes such as “insufficient /// resolved / to / for / what / what / recommended /?” Then, determine each part of speech. For example, as shown in FIG. 8 (c), information such as age 20 years, height 170cm, weight 60kg is used as a search key as the linguistic information and the subject health care subject information as a search key. The corresponding question information is searched from the question information stored in 160, and the question information is acquired as shown in FIG. The acquired question information is transmitted to the generation unit 130.

  As shown in FIG. 8E, for example, the generation unit 130 extracts answer information associated with the question information, generates a hypothesis from the extracted answer information, and generates a hypothesis. Verification is performed with the supporting grounds (S13). Further, as illustrated in FIG. 8F, the generation unit 130 generates a sentence example as an example from the verified answer information (step S15), and generates the generated sentence example and the certainty degree evaluated by the evaluation unit 140. In addition, it is transmitted to the correction unit 150. In FIG. 8F, a radio button for selecting a recommended recommended task is provided in the sentence example, and action selection information is acquired by specifying the radio button. However, the radio button is not limited thereto. For example, the evaluation unit 140 may acquire and select the action selection information from message information such as “I would like to stretch among recommended tasks” as reply message information of the health care target person.

  The evaluation unit 140 evaluates the certainty factor of the response information extracted by the generation unit 130 based on the grounds verified by the generation unit 130 and the past evaluation of the response information (step S14). The evaluated certainty factor is transmitted to the generating unit 130.

16). Specifically, as an example, as shown in FIG. 7 (a), when the extracted response information is evaluated for certainty in the order of jogging, stretching, swimming, and dietary restrictions, the recommendation ranking is also evaluated. Once set in order. Here, for example, with respect to the “load” of the index of each recommended task, as shown in FIG. 7C, “high load” in the chart indicating the tendency value of the behavior of the corresponding health management subject If the action tends to be “does not execute against the advice”, there is a possibility that the recommended content does not necessarily match the subject of the health care.

  Therefore, in the case of this tendency, the weight of the “load” indicator is negatively weighted, the sum of the indicators is calculated as a correction value, and the certainty factor is corrected by, for example, adding the correction value to the certainty factor. be able to. The correction method is not limited to the calculation method. For example, the correlation coefficient between the two values of the biological feature value or the behavior tendency value and the value of one of the indices is expressed as two values. It can also be obtained from the covariance and standard deviation and corrected using the correlation coefficient. By replacing the recommendation ranking as shown in FIG. 7B based on the evaluation after the correction, it is possible to advise a recommended task that more closely matches the health management subject. In addition, in the said correction | amendment, you may correct | amend from the log | history and frequency which used the task for the sentence example as shown in FIG. The corrected sentence example and evaluation are transmitted to the evaluation unit 140.

  The evaluation unit 140 determines whether or not the health management target person is grouped. If the group is grouped (YES in step S17), the group that transmits the same, similar, or opposite illustration within a predetermined time The other health management target person is determined as a comparison target person as a rival or target person, and the determination is transmitted to the correction unit 150.

  The correcting unit 150 corrects the sentence example and the evaluation with the information regarding the determined comparison target person (step S18). Specifically, for example, the name of the person to be compared is specified in the example, and “Mr. △△ (the person to be compared) is jogging recommended to Mr. XX (the person who is subject to health care). It can be corrected by adding a sentence such as “!”. The correction unit 150 transmits the determined and corrected sentence example and evaluation to the transmission unit 170.

  The transmission unit 170 determines from the curriculum information whether or not the curriculum has a follower of the expert. If the curriculum has the follower (YES in step S19), the transmission unit 170 generates and corrects the display unit 310 of the user terminal 300 of the expert. A sentence example and an evaluation are displayed and output (step S20).

  Regarding the sentence example displayed and output on the display unit 310, the expert selects the sentence example as the one to be transmitted to the health management subject, and edits the selected sentence example and the evaluation as necessary (step S21). The display unit 310 transmits the selected and edited sentence example and evaluation to the transmission unit 170.

The transmission unit 170 transmits the transmitted sentence example and evaluation to the user terminal 200 (step S22).
The above is the description of the operation of the health management server 100.

As shown in FIG. 10, the health management server 100 performs a cross-analysis based on the lifestyle questionnaire results of the health care subjects and the genetic test results of the health care subjects, and the lifestyle, eating habits, or exercise habits. Can improve the standardization and efficiency of report creation for guidance aimed at improving
Specifically, the health management server 100 extracts concerns about the health of the health management target person and the characteristics of the lifestyle from the lifestyle questionnaire result of the health management target person, and extracts the extracted content and the health management target person's characteristics. The result of the genetic test is stored in the storage unit 160, the stored data is filtered and scored, cross analysis is performed, and a report is automatically generated from the analysis result, thereby standardizing the creation and Increase efficiency. The extraction and cross analysis are performed by the analysis unit 120.

Moreover, as shown in FIG. 11, the health management server 100 performs cross analysis based on the lifestyle questionnaire results of the health management subject and the blood test results of the health management subject, and improves the habits to be improved. In addition, it is possible to standardize and improve the efficiency of creating reports for the purpose of continuation of habits to be continued.
Specifically, the health management server 100 extracts the lifestyle related to the blood of the health management subject from the lifestyle questionnaire result of the health management subject, and extracts the extracted content and the blood test of the health management subject. The results are stored in the storage unit 160, the stored data is filtered and scored, cross analysis is performed, and a report is automatically created from the analysis results, thereby standardizing and improving the efficiency of the creation. Plan. The extraction and cross analysis are performed by the analysis unit 120.

  In addition, as shown in FIG. 12, the health management server 100 is based on the past behavior history (hereinafter referred to as “log”), lifestyle questionnaire results, genetic test results, and blood test results of the health management subject. By cross-analyzing and delivering content aimed at providing tasks, training and fitness videos, ideal meals, and health knowledge that are effective for diet and lifestyle activities performed by health care subjects, It is possible to reduce the burden on expert guidance.

  Specifically, the health management server extracts behavioral changes and concerns of the health management target person from the log, the lifestyle questionnaire result and the life style questionnaire result of the health management target person, and the extracted contents and the health management target The results of the genetic test and blood test of the person are stored in the storage unit 160, the stored data is filtered and scored, cross analysis is performed, and content distribution is performed based on the analysis result, thereby providing expert guidance. To reduce the load. The extraction and cross analysis are performed by the analysis unit 120.

<Others>
Although the health guidance service is taken up as a service according to the present invention, the present invention can also be used in other businesses. In particular, businesses that use people's psychology and spiritual actions (for example, businesses that give advice to increase human motivation), welfare (for example, when specialists provide remote advice on the welfare of local communities, etc.) ), Can be used in business such as education.

100 health management server 110 reception unit 120 analysis unit 130 generation unit 140 evaluation unit 150 correction unit 160 storage unit 170 transmission unit 200 user terminal (person subject to health management)
300 User terminal (expert)
310 Display 400 Network

Claims (10)

A health management server connected to a terminal via a network,
A storage unit for storing health management target person information and answer information of a health management target person having the terminal;
A receiving unit for receiving input information transmitted from the terminal;
A generator that extracts the answer information corresponding to the received input information and generates a sentence example based on the extracted answer information;
An evaluation unit that evaluates the extracted answer information with a certainty factor indicating the certainty of the answer information;
A correction unit that corrects at least one of the sentence example or the evaluation based on at least one of the health management subject information or the content of the input information;
A health management server comprising: a transmitting unit that transmits the corrected sentence example to the terminal. The health management server is connected to the terminal via a network,
2. The health management server according to claim 1, wherein the corrected sentence example and the evaluation are displayed and output on a display unit of the terminal. The transmitter is
The health management server according to claim 2, wherein advice information including sentence examples and evaluations selected or edited and selected by the expert on the display unit is transmitted to the terminal. The pre-evaluation unit determines whether the evaluation confidence level is high or low, and corrects the sentence based on the determination result, or generates an alert message and notifies the expert. One to choose,
The correction unit corrects the example sentence based on the determination result and answer information for avoidance when the evaluation unit selects to correct the example sentence,
2. The health according to claim 1, wherein when the evaluation unit selects to notify the expert of an alert message, the transmission unit notifies the expert terminal of the alert message. Management server. The receiving unit receives expert sentence example selection information, action selection information of a health care target person, action execution information, and biological log information from the terminal,
The evaluation unit performs a combination based on the received sentence example selection information, action selection information, action execution information and biological log information, or a time stamp of these information, and the example sentence selection information, action selection information, The health management server according to any one of claims 1 to 4, wherein the response information associated with the action execution information and the biolog information is evaluated. The image information is taken by the terminal,
6. The health according to claim 1, wherein the analysis unit calculates a feature of the captured image information and analyzes language information based on the feature and the question information. Management server. The health management server includes a calculation unit that calculates the use frequency of response information for each health management target person,
The storage unit stores past use history and use frequency of the answer information of the health care target person as use history information,
The health management server according to claim 1, wherein the correction unit corrects a sentence example based on the use history information. The receiving unit receives image information or message information transmitted from a plurality of terminals grouped in advance,
The evaluation unit identifies a sentence example that is the same as, similar to, or opposite to a sentence example transmitted to a terminal, and identifies a terminal in a group that is transmitted within a predetermined time, and compares health management subjects who have the identified terminal Judge as a target person,
The health management server according to claim 1, wherein the correction unit corrects the sentence example and the evaluation based on the comparison target person information. A method for controlling a health management server connected to a terminal via a network,
Storing health care target person information and answer information of a health care target person having the terminal;
Receiving input information transmitted from the terminal;
Extracting the answer information corresponding to the received input information, and generating a sentence example based on the extracted answer information;
Evaluating the extracted answer information with a certainty factor indicating the certainty of the answer information;
Correcting at least any of the sentence examples or evaluations based on at least one of the health management subject information or the content of the input information;
And a step of transmitting the corrected sentence example to the terminal. A health care program connected to a terminal via a network,
On the computer,
A storage function for storing health management target person information and answer information of a health management target person having the terminal;
A receiving function for receiving input information transmitted from the terminal;
A function of extracting the answer information corresponding to the received input information, and generating a sentence example based on the extracted answer information;
An evaluation function for evaluating the extracted answer information with a certainty factor indicating the certainty of the answer information;
A correction function for correcting at least one of the sentence examples or the evaluation based on at least one of the health management subject information or the content of the input information;
A health management program for realizing a transmission function for transmitting the corrected sentence example to the terminal.

Images