Medical service terminal interaction system based on artificial intelligence

Abstract

The invention discloses a medical service terminal interaction system based on artificial intelligence, and relates to the technical field of medical treatment, and the system comprises a user classification module which obtains a description precision coefficient of a user according to the answer of the user to a service verification question; an interaction identification module; a feature extraction module; the pain point supplementing module forms a user supplementing pain point; the scheme forming module forms a medical service scheme; a scheme evaluation module; and the service correction module performs correction to obtain a medical service correction scheme. By setting a user classification module, a feature extraction module, a pain point supplement module, a scheme forming module and a service correction module, the situation that services needed by a user are missed due to inaccurate description of the user can be avoided, secondary correction is conducted on a medical service scheme, and therefore it is further guaranteed that the formed scheme meets the requirements of the user, and the user experience is improved. And thus, the interaction experience feeling is improved.

Description

Technical Field

[0001] The present invention relates to the field of medical technology, and specifically relates to a medical service terminal interaction system based on artificial intelligence. Background Art

[0002] With the rapid development of technology, the medical industry is also constantly transforming and upgrading. To meet the growing demand for medical services, intelligence and convenience have become the main development directions of modern medical services. Among them, the rapid development of artificial intelligence technology has brought a revolutionary change to the medical industry. The combination of artificial intelligence and medical service terminals can promote the interaction between users and medical service terminals. Since there are many medical services in medical service terminals, it is difficult for users to find the required medical services by themselves.

[0003] However, due to the different degrees of understanding of medicine by users themselves and the different accuracies of their descriptions, it is difficult to grasp the accuracy of the feedback information of medical service terminals using artificial intelligence, resulting in users having difficulty obtaining services that match them. Summary of the Invention

[0004] To solve the above technical problems, a medical service terminal interaction system based on artificial intelligence is provided, and this technical solution solves the problems raised in the above background art.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] A medical service terminal interaction system based on artificial intelligence, comprising:

[0007] A user classification module, which forms service verification questions based on historical data, uses the service verification questions, and obtains the description accuracy coefficient of the user according to the user's answers to the service verification questions;

[0008] An interaction recognition module, which performs voice recognition on the user's voice or visual recognition on the user's body language to obtain the user's service description text;

[0009] A feature extraction module, which analyzes and recognizes the service description text to obtain at least one user pain point;

[0010] A pain point supplement module, which supplements the user pain points based on the description accuracy coefficient of the user to form user supplementary pain points;

[0011] A solution formation module, which forms a medical service solution based on the user supplementary pain points;

[0012] A solution evaluation module that obtains the user's satisfaction with the medical service solution. When the satisfaction is less than a preset value, the service is corrected for the user; otherwise, the service is provided to the user according to the medical service solution.

[0013] A service correction module that estimates and forms at least one potential pain point of the user, forms a preview service for the potential pain point, displays the preview service for the potential pain point, obtains the target potential pain point based on the user's response, obtains the dissatisfaction marking points of the user with respect to the medical service solution, and corrects and obtains a medical service correction solution based on the target potential pain point and the dissatisfaction marking points.

[0014] Preferably, forming the service verification problem based on historical data includes the following steps:

[0015] Obtain at least one medical service provided in the medical service terminal, obtain the coincidence degrees of the service modes, service populations, and service purposes of two medical services, and based on the analytic hierarchy process, obtain the weights of the coincidence degrees of the service modes, service populations, and service purposes.

[0016] Use the similarity formula to calculate the similarity between two medical services.

[0017] Classify the medical services based on the similarity of the medical services to obtain at least one medical service category, and the difference in the similarity of the medical services in the medical service category is less than a preset difference.

[0018] Obtain the common points of the medical services in the medical service category, form at least one basic attribute of the common points, and form at least one non-basic attribute that the common points do not have.

[0019] Record the basic attribute and the non-basic attribute as the attributes to be judged.

[0020] The service verification problem is: whether the medical service category has the attribute to be judged. When the attribute to be judged is a basic attribute, the service verification answer is yes; when the attribute to be judged is a non-basic attribute, the service verification answer is no.

[0021] The similarity formula is as follows:

[0022] A = αa + βb + γc

[0023] Where A is the similarity between two medical services, α, β, and γ are the weights of the coincidence degrees of the service mode, service population, and service purpose respectively, and a, b, and c are the coincidence degrees of the service modes, service populations, and service purposes of the two medical services respectively.

[0024] Preferably, obtaining the user's description accuracy coefficient according to the user's answer to the service verification question includes the following steps:

[0025] Obtain the user's answer to the service verification question, count the number of service verification questions answered by the user, and count the number of service verification questions answered correctly by the user. Among them, the number of service verification questions answered by the user is not less than the preset number;

[0026] Divide the number of service verification questions answered correctly by the user by the number of service verification questions answered by the user to obtain the user's description accuracy coefficient.

[0027] Preferably, analyzing and identifying the service description text to obtain at least one user pain point includes the following steps:

[0028] Segment the service description text to obtain at least one text phrase;

[0029] Take one of the at least one text phrases as the target text phrase, and take the part of the service description text other than the target text phrase as the part to be compared;

[0030] Based on big data, construct a thesaurus, and based on the thesaurus, obtain at least one synonym of the target text phrase;

[0031] Obtain the number of occurrences of the target text phrase and its synonyms in the part to be compared as the characteristic number of times;

[0032] When the target text phrase traverses at least one text phrase, obtain at least one characteristic number of times;

[0033] Select one of the target text phrases with equal characteristic numbers of times as the keyword;

[0034] Based on big data, obtain the probability of the parallel occurrence of the medical service and the keyword as the correlation degree between the keyword and the medical service;

[0035] Take the keywords with a correlation degree greater than the preset probability and the keywords in the medical service as user pain points, and correspond the user pain points to the medical service.

[0036] Preferably, supplementing the user pain points based on the user's description accuracy coefficient to form user supplementary pain points includes the following steps:

[0037] Obtain the medical service corresponding to the user pain point as the pain point medical service;

[0038] Based on historical data, obtain the value range (d, e) of the description accuracy coefficient, obtain the value range (f, g) of the similarity, and form a data mapping function;

[0039] Substitute the description accuracy coefficient into the data mapping function to obtain the similarity threshold value;

[0040] Obtain the medical services whose similarity to the pain point medical service is greater than the similarity threshold value as supplementary medical services, where the similarity of the pain point medical service to itself is 1;

[0041] Take the user pain points corresponding to the supplementary medical services as user supplementary pain points;

[0042] The data mapping function is as follows:

[0043]

[0044] Where B is the similarity threshold value and x is the description accuracy coefficient.

[0045] Preferably, forming a medical service plan based on the user supplementary pain points includes the following steps:

[0046] Based on historical data, obtain the probability that two user supplementary pain points do not appear synchronously in at least one medical service scenario as the mutual exclusion coefficient of the two user supplementary pain points;

[0047] Summarize the user supplementary pain points to form a user supplementary pain point set;

[0048] Summarize the user supplementary pain points whose mutual exclusion coefficient is greater than the preset ratio to obtain a mutual exclusion set of user supplementary pain points;

[0049] Sort the user supplementary pain points from largest to smallest according to the number of elements in the mutual exclusion set of user supplementary pain points to obtain a user supplementary pain point sequence;

[0050] Delete the user supplementary pain points in the user supplementary pain point set in the order of the user supplementary pain point sequence until there are no two user supplementary pain points in the user supplementary pain point set whose mutual exclusion coefficient is greater than the preset ratio;

[0051] Summarize the medical services corresponding to the user supplementary pain points in the reduced user supplementary pain point set to form a medical service plan.

[0052] Preferably, estimating to form at least one potential pain point of the user and forming a preview service for the potential pain point includes the following steps:

[0053] Take the interval composed of 0 and the description accuracy coefficient as the potential interval, and evenly take at least one identification point in the potential interval;

[0054] Substitute the value of the identification point into the position of the description accuracy coefficient in the data mapping function to obtain the similarity identification value;

[0055] Obtain medical services with a similarity to the pain point medical service greater than the similarity recognition value as potential medical services, where the similarity of the pain point medical service to itself is 1;

[0056] Summarize the potential medical services corresponding to the recognition points to obtain a set of potential medical services;

[0057] When the set of potential medical services covers the medical services at the dissatisfied annotation points, regard the recognition points as target recognition points;

[0058] Take the user pain points corresponding to the potential medical services in the set of potential medical services of the target recognition point with the largest value as potential pain points;

[0059] Visually display the medical services corresponding to the potential pain points to form a preview service.

[0060] Preferably, obtaining the target potential pain points based on the user's reaction includes the following steps:

[0061] Based on visual recognition or voice recognition, take the potential pain points corresponding to the preview services recognized by the user as the target potential pain points.

[0062] Preferably, the method for correcting and obtaining a medical service correction plan based on the target potential pain points and the dissatisfied annotation points includes the following steps:

[0063] Obtain the medical services at the dissatisfied annotation points as the medical services to be replaced;

[0064] Take the similarity between the medical services corresponding to the target potential pain points and the medical services to be replaced as the similarity between the target potential pain points and the medical services to be replaced;

[0065] Obtain the minimum value of the similarity between the target potential pain points and at least one medical service to be replaced, and replace the medical service to be replaced corresponding to the minimum value of the similarity with the medical service corresponding to the target potential pain points to obtain a medical service correction plan.

[0066] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0067] By setting up a user classification module, a feature extraction module, a pain point supplement module, a plan formation module and a service correction module, calculate the description accuracy coefficient of the user, and based on the description accuracy coefficient, form the user supplementary pain points, and based on this, form a medical service plan. Thus, when forming the plan, fuzzy recognition can be used for the user's description, so as to avoid missing the services required by the user due to inaccurate user descriptions, and perform secondary correction on the medical service plan, thereby further ensuring that the formed plan meets the user's needs, and thus improving the interactive experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 The pain points shown in the process of the medical service terminal interaction system based on artificial intelligence of the present invention;

[0069] Figure 2 The pain points shown in the process of forming service verification questions based on historical data of the present invention;

[0070] Figure 3 The pain points shown in the process of obtaining the user's description accuracy coefficient according to the user's answer to the service verification question of the present invention;

[0071] Figure 4 The pain points shown in the process of analyzing and identifying the service description text to obtain at least one user pain point of the present invention;

[0072] Figure 5 The pain points shown in the process of supplementing user pain points based on the user's description accuracy coefficient to form user supplementary pain points of the present invention;

[0073] Figure 6 The pain points shown in the process of forming a medical service plan based on user supplementary pain points of the present invention;

[0074] Figure 7 The pain points shown in the process of estimating and forming at least one potential pain point of the user to form a preview service of potential pain points of the present invention;

[0075] Figure 8 The pain points shown in the process of correcting to obtain a medical service correction plan based on the target potential pain points and dissatisfaction marking points of the present invention. Detailed implementation manners

[0076] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments in the following description are only examples, and those skilled in the art can think of other obvious variations.

[0077] Refer to Figure 1 As shown, a medical service terminal interaction system based on artificial intelligence includes:

[0078] A user classification module, which forms service verification questions based on historical data, uses the service verification questions, and obtains the user's description accuracy coefficient according to the user's answer to the service verification questions;

[0079] An interaction recognition module, which performs voice recognition on the user's voice or visual recognition on the user's body language to obtain the user's service description text;

[0080] A feature extraction module, which analyzes and identifies the service description text to obtain at least one user pain point;

[0081] A pain point supplement module, wherein the pain point supplement module supplements the user's pain points based on the user's description accuracy coefficient to form user supplementary pain points;

[0082] A plan forming module, wherein the plan forming module forms a medical service plan based on the user's supplementary pain points;

[0083] A program evaluation module, wherein the program evaluation module obtains the user's satisfaction with the medical service program, and when the satisfaction is less than a preset value, the user is provided with service corrections; otherwise, the user is provided with service according to the medical service program;

[0084] A service correction module estimates at least one potential pain point of a user, forms a preview service of the potential pain point, displays the preview service of the potential pain point, obtains the target potential pain point based on the user's response, obtains the user's dissatisfaction marking points for the medical service plan, and corrects the medical service correction plan based on the target potential pain point and the dissatisfaction marking points.

[0085] In this solution, users have received different educations, and therefore have different understandings of medical services. Therefore, when using artificial intelligence medical service terminals, the accuracy of the medical services they need may be insufficient. Therefore, if accurate identification is performed based on the user's description and services are formed based on the results of accurate identification, the services will most likely not meet the user's needs, which will affect the interactive experience. Therefore, in this solution, the user's description accuracy coefficient is obtained through the user classification module, and medical service plans are formed for the user with different accuracies based on the description accuracy coefficients of different users. Therefore, it is ensured that the search range includes the medical plan required by the user, so that a service plan that meets their needs can be formed.

[0086] Reference Figure 2 As shown, based on historical data, forming a service verification question includes the following steps:

[0087] Obtain at least one medical service provided in the medical service terminal, obtain the overlap of service modes, overlap of service populations and overlap of service purposes of two medical services, and obtain the weights of the overlap of service modes, overlap of service populations and overlap of service purposes based on the hierarchical analysis method;

[0088] Use the similarity formula to calculate the similarity of two medical services;

[0089] Based on the similarity of the medical services, the medical services are classified to obtain at least one medical service category, wherein the difference in the similarity of the medical services in the medical service category is less than a preset difference;

[0090] Obtain the common points of medical services in a medical service category, form at least one basic attribute of the common points, and form at least one non-basic attribute that the common points do not have;

[0091] Record the basic attribute and the non-basic attribute as the attributes to be judged;

[0092] The service verification question is: whether the medical service category has the attribute to be judged. When the attribute to be judged is a basic attribute, the service verification answer is yes. When the attribute to be judged is a non-basic attribute, the service verification answer is no;

[0093] The similarity formula is as follows:

[0094] A = αa + βb + γc

[0095] Where, A is the similarity of two medical services, α, β, and γ are the weights of the coincidence degree of service modes, the coincidence degree of service populations, and the coincidence degree of service purposes respectively, and a, b, and c are the coincidence degree of service modes, the coincidence degree of service populations, and the coincidence degree of service purposes of two medical services respectively.

[0096] The calculation of similarity is to classify medical services. Therefore, when setting service verification questions, the number of questions set can be reduced, rather than asking all medical services;

[0097] Since the service verification questions contain medical service categories, by setting service verification questions, the degree of understanding of the attributes of medical service categories by users can be determined. In actual testing, the test questions cannot cover all service verification questions. As long as the number is large enough, such as the preset number set according to experience, the obtained description accuracy coefficient is sufficient to estimate the description accuracy of users, and then the recognition range of subsequent services can be adjusted according to this accuracy.

[0098] Refer to Figure 3 As shown, obtaining the user's description accuracy coefficient according to the user's answer to the service verification question includes the following steps:

[0099] Obtain the user's answer to the service verification question, count the number of service verification questions answered by the user, and count the number of service verification questions answered correctly by the user. Among them, the number of service verification questions answered by the user is not less than the preset number;

[0100] Divide the number of service verification questions answered correctly by the user by the number of service verification questions answered by the user to obtain the user's description accuracy coefficient.

[0101] Refer to Figure 4 As shown, analyzing and identifying the service description text to obtain at least one user pain point includes the following steps:

[0102] Segment the service description text to obtain at least one text phrase;

[0103] Select one of the at least one text phrases as the target text phrase, and use the part of the service description text other than the target text phrase as the part to be compared;

[0104] Based on big data, construct a thesaurus, and based on the thesaurus, obtain at least one synonym of the target text phrase;

[0105] Obtain the number of occurrences of the target text phrase and its synonyms in the part to be compared as the characteristic number;

[0106] When traversing at least one text phrase for the target text phrase, obtain at least one characteristic number;

[0107] Select one of the target text phrases with equal characteristic numbers as the keyword;

[0108] Based on big data, obtain the probability of the parallel occurrence of the medical service and the keyword as the correlation degree between the keyword and the medical service;

[0109] Take the keywords with a correlation degree greater than the preset probability and the keywords in the medical service as user pain points, and correspond the user pain points to the medical service.

[0110] When two target text phrases are synonyms, according to the way of obtaining the characteristic number, it can be known that the characteristic numbers of the two target text phrases are equal. Because the characteristic number actually counts the number of a class of synonyms, thus, when the characteristic numbers of the two target text phrases are equal, it can be considered that the two target text phrases are synonyms. In fact, there is a very small probability that they are not synonyms, but as long as the probability is small enough, it will not affect the operation of the system, so it can be ignored. Then select one of the target text phrases with equal characteristic numbers as the keyword, which can remove duplicates from the service description text, and then retain the key part of the service description text. Thus, the user pain points are obtained.

[0111] Refer to Figure 5 As shown, based on the description accuracy coefficient of the user, supplement the user pain points to form user supplementary pain points, including the following steps:

[0112] Obtain the medical service corresponding to the user pain point as the pain point medical service;

[0113] Based on historical data, obtain the value range (d, e) of the description accuracy coefficient, obtain the value range (f, g) of the similarity, and form a data mapping function;

[0114] Substitute the description accuracy coefficient into the data mapping function to obtain the similarity threshold;

[0115] Obtain medical services with a similarity to the pain point medical service greater than the similarity threshold as supplementary medical services, where the similarity of the pain point medical service to itself is 1;

[0116] Take the user pain points corresponding to the supplementary medical services as user supplementary pain points;

[0117] The data mapping function is as follows:

[0118]

[0119] Where B is the similarity threshold and x is the description precision coefficient.

[0120] The key point of user pain point supplementation is to determine the identification range of medical services, and this identification range is determined by the description precision coefficient. When the description precision coefficient is larger, the identification range of medical services is smaller, but there is no mapping relationship between the description precision coefficient and the similarity. Therefore, it is impossible to obtain the similarity threshold based on the description precision coefficient to determine the identification range of medical services. In this solution, since the value range (d, e) of the description precision coefficient and the value range (f, g) of the similarity are corresponding, therefore, according to their corresponding relationship, both the description precision coefficient and the similarity threshold exist as critical functions. Therefore, the relative positions of the description precision coefficient and the similarity threshold in the value range (d, e) of the description precision coefficient and the value range (f, g) of the similarity are the same, that is Therefore, according to this proportional relationship, the data mapping function can be obtained, and further, obtain medical services with a similarity to the pain point medical service greater than the similarity threshold as supplementary medical services.

[0121] Refer to Figure 6 As shown, forming a medical service plan based on user supplementary pain points includes the following steps:

[0122] Based on historical data, obtain the probability that two user supplementary pain points do not appear synchronously in at least one medical service scenario as the mutual exclusion coefficient of the two user supplementary pain points;

[0123] Summarize the user supplementary pain points to form a user supplementary pain point set;

[0124] Summarize the user supplementary pain points with a mutual exclusion coefficient greater than the preset ratio to obtain the mutual exclusion set of user supplementary pain points;

[0125] Sort the user supplementary pain points from largest to smallest according to the number of elements in the mutual exclusion set of user supplementary pain points to obtain the user supplementary pain point sequence;

[0126] Delete the user-supplemented pain points in the user-supplemented pain point set in the order of the user-supplemented pain point sequence until there are no two user-supplemented pain points in the user-supplemented pain point set whose mutual exclusion coefficient is greater than the preset ratio;

[0127] Summarize the medical services corresponding to the user-supplemented pain points in the user-supplemented pain point set after deletion to form a medical service plan.

[0128] When forming a medical service plan, since the user-supplemented pain points obtained are not verified, it is possible that two of the user-supplemented pain points may not actually occur simultaneously. Therefore, it is necessary to delete the user-supplemented pain points in this case to ensure that there are no unreasonable situations among the user-supplemented pain points in the user-supplemented pain point set. When deleting, form a mutual exclusion set for each user-supplemented pain point. The elements of the mutual exclusion set are the elements in the user-supplemented pain point set that conflict with this user-supplemented pain point. Since in practice, the user-supplemented pain points belonging to the same medical service plan do not conflict with the other user-supplemented pain points, when deleting, it is inevitable to give priority to deleting the user-supplemented pain point with more elements in the mutual exclusion set because it has a greater probability of not belonging to the current plan.

[0129] Refer to Figure 7 As shown, estimate at least one potential pain point of the user. The steps for forming the preview service of the potential pain point are as follows:

[0130] Take the interval formed by 0 and the description accuracy coefficient as the potential interval, and evenly take at least one identification point in the potential interval;

[0131] Substitute the value of the identification point into the position of the description accuracy coefficient in the data mapping function to obtain the similarity identification value. Here, it is to replace the position of the description accuracy coefficient in the data mapping function with the value of the identification point, that is, let x be equal to the value of the identification point;

[0132] Obtain the medical services whose similarity to the pain point medical service is greater than the similarity identification value as potential medical services, where the similarity between the pain point medical service and itself is 1;

[0133] Summarize the potential medical services corresponding to the identification points to obtain a potential medical service set;

[0134] When the potential medical service set covers the medical service at the dissatisfaction marking point, take the identification point as the target identification point;

[0135] Take the user pain points corresponding to the potential medical services in the potential medical service set of the target identification point with the largest value as potential pain points;

[0136] Visualize the medical services corresponding to the potential pain points to form a preview service.

[0137] When obtaining potential pain points, it is actually an expansion of the supplementary pain points of users. Since the acquisition of users' supplementary pain points is based on the description precision coefficient, when obtaining potential pain points, the description precision coefficient is gradually decreased, that is, for each identification point, and then the potential medical service set corresponding to each identification point is gradually obtained. To avoid adding completely irrelevant potential pain points, the user pain points corresponding to the potential medical services in the potential medical service set of the target identification point with the largest value are used as potential pain points.

[0138] Based on the user's response, obtaining the target potential pain points includes the following steps:

[0139] Based on visual recognition or speech recognition, the potential pain points corresponding to the preview services recognized by the user are used as the target potential pain points.

[0140] Refer to Figure 8 As shown, based on the target potential pain points and the dissatisfaction annotation points, the steps to correct and obtain the medical service correction plan include the following:

[0141] Obtain the medical service at the dissatisfaction annotation point as the medical service to be replaced;

[0142] The similarity between the medical service corresponding to the target potential pain point and the medical service to be replaced is used as the similarity between the target potential pain point and the medical service to be replaced;

[0143] Obtain the minimum value of the similarity between the target potential pain point and at least one medical service to be replaced, and replace the medical service to be replaced corresponding to the minimum value of the similarity with the medical service corresponding to the target potential pain point to obtain the medical service correction plan.

[0144] Furthermore, this solution also proposes a storage medium on which a computer-readable program is stored. When the computer-readable program is called, it executes the above-mentioned medical service terminal interaction system based on artificial intelligence.

[0145] It can be understood that the storage medium can be a magnetic medium, such as a floppy disk, a hard disk, a magnetic tape; an optical medium, such as a DVD; or a semiconductor medium, such as a solid-state drive (SSD).

[0146] In summary, the advantages of the present invention are as follows: By setting up a user classification module, a feature extraction module, a pain point supplement module, a solution formation module, and a service correction module, the description accuracy coefficient of the user is calculated, and based on the description accuracy coefficient, the formation of the user's supplementary pain points is carried out, and a medical service solution is formed based on this. Thus, when forming the solution, fuzzy recognition can be used for the user's description, thereby avoiding missing the services required by the user due to inaccurate user descriptions, and making a secondary correction to the medical service solution, so as to further ensure that the formed solution meets the needs of the user, thereby enhancing the interactive experience.

[0147] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. What is described in the above embodiments and the specification is only the principle of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of protection required by the present invention is defined by the appended claims and their equivalents.

Claims

1. A medical service terminal interaction system based on artificial intelligence, characterized in that, Including: A user classification module, which forms service verification questions based on historical data, uses the service verification questions, and obtains the user's description accuracy coefficient according to the user's answers to the service verification questions; An interaction recognition module, which performs voice recognition on the user's voice or visual recognition on the user's body language to obtain the user's service description text; A feature extraction module, which analyzes and recognizes the service description text to obtain at least one user pain point; A pain point supplement module, which supplements the user pain points based on the user's description accuracy coefficient to form user supplementary pain points; A solution formation module, which forms a medical service solution based on the user supplementary pain points; A solution evaluation module, which obtains the user's satisfaction with the medical service solution. When the satisfaction is less than the preset value, the service for the user is corrected. Otherwise, the service is provided to the user according to the medical service solution; A service correction module, which estimates and forms at least one potential pain point of the user, forms a preview service for the potential pain point, displays the preview service for the potential pain point, obtains the target potential pain point based on the user's reaction, obtains the dissatisfaction marking points of the user with respect to the medical service solution, and corrects to obtain a medical service correction solution based on the target potential pain point and the dissatisfaction marking points.

2. The medical service terminal interaction system based on artificial intelligence according to claim 1, wherein The forming of the service verification questions based on historical data includes the following steps: Obtain at least one medical service provided in the medical service terminal, obtain the coincidence degrees of the service modes, service populations, and service purposes of two medical services, and obtain the weights of the coincidence degrees of the service modes, service populations, and service purposes based on the analytic hierarchy process; Use the similarity formula to calculate the similarity of two medical services; Classify the medical services based on the similarity of the medical services to obtain at least one medical service category, where the difference in the similarity of the medical services in the medical service category is less than the preset difference; Obtain the common points of the medical services in the medical service category, form at least one basic attribute of the common points, and form at least one non-basic attribute that the common points do not have; Record the basic attribute and the non-basic attribute as the attributes to be judged; The service verification question is: whether the medical service category has the attribute to be judged. When the attribute to be judged is the basic attribute, the service verification answer is yes. When the attribute to be judged is the non-basic attribute, the service verification answer is no; The similarity formula is as follows: A = αa + βb + γc Wherein, A is the similarity of two medical services, α, β, and γ are respectively the weights of the coincidence degrees of the service mode, service population, and service purpose, and a, b, and c are respectively the coincidence degrees of the service modes, service populations, and service purposes of two medical services.

3. The medical service terminal interaction system based on artificial intelligence according to claim 2, characterized in that, The obtaining of the user's description accuracy coefficient according to the user's answers to the service verification questions includes the following steps: Obtain the user's answers to the service verification questions, count the number of service verification questions answered by the user, and count the number of service verification questions answered correctly by the user, where the number of service verification questions answered by the user is not less than the preset number; The number of service verification questions answered correctly by the user is divided by the number of service verification questions answered by the user to obtain the user's description accuracy coefficient.

4. The medical service terminal interaction system based on artificial intelligence according to claim 3, wherein, The analysis and recognition of the service description text to obtain at least one user pain point includes the following steps: Segment the service description text to obtain at least one text phrase; Select one of the at least one text phrases as the target text phrase, and use the part of the service description text other than the target text phrase as the part to be compared; Based on big data, construct a thesaurus, and based on the thesaurus, obtain at least one synonym of the target text phrase; Obtain the number of occurrences of the target text phrase and the synonym in the part to be compared as the feature number; When the target text phrase traverses at least one text phrase, obtain at least one feature number; Select one of the target text phrases with equal feature numbers as the keyword; Based on big data, obtain the probability of the parallel occurrence of the medical service and the keyword as the correlation degree between the keyword and the medical service; Use the keywords with a correlation degree greater than the preset probability and the keywords in the medical service as user pain points, and correspond the user pain points to the medical service.

5. An interactive system for a medical service terminal based on artificial intelligence according to claim 4, characterized in that, The supplementation of the user pain points based on the user's description accuracy coefficient to form user supplementary pain points includes the following steps: Obtain the medical service corresponding to the user pain point as the pain point medical service; Based on historical data, obtain the value range (d, e) of the description accuracy coefficient and the value range (f, g) of the similarity to form a data mapping function; Substitute the description accuracy coefficient into the data mapping function to obtain the similarity critical value; Obtain the medical services with a similarity greater than the similarity critical value to the pain point medical service as supplementary medical services, where the similarity between the pain point medical service and itself is 1; Use the user pain points corresponding to the supplementary medical services as user supplementary pain points; The data mapping function is as follows: Where B is the similarity critical value and x is the description accuracy coefficient.

6. The interactive system of a medical service terminal based on artificial intelligence according to claim 5, wherein The formation of a medical service plan based on the user supplementary pain points includes the following steps: Based on historical data, obtain the probability of the asynchronous occurrence of two user supplementary pain points in at least one medical service scenario as the mutual exclusion coefficient of the two user supplementary pain points; Summarize the user supplementary pain points to form a user supplementary pain point set; Summarize the user supplementary pain points with a mutual exclusion coefficient greater than the preset ratio to obtain a mutual exclusion set of user supplementary pain points; Sort the user supplementary pain points from largest to smallest according to the number of elements in the mutual exclusion set of user supplementary pain points to obtain a user supplementary pain point sequence; Delete the user supplementary pain points in the user supplementary pain point set in the order of the user supplementary pain point sequence until there are no two user supplementary pain points with a mutual exclusion coefficient greater than the preset ratio in the user supplementary pain point set; Summarize the medical services corresponding to the user supplementary pain points in the reduced user supplementary pain point set to form a medical service plan.

7. An interactive system for a medical service terminal based on artificial intelligence according to claim 6, characterized in that, The estimation of forming at least one potential pain point of the user and forming a preview service of the potential pain point includes the following steps: Use the interval formed by 0 and the description accuracy coefficient as the potential interval, and uniformly take at least one identification point in the potential interval; Substitute the value of the recognition point into the position of the precise coefficient described in the data mapping function to obtain the similarity recognition value; Obtain the medical services with a similarity to the pain point medical service greater than the similarity recognition value as potential medical services, where the similarity between the pain point medical service and itself is 1; Summarize the potential medical services corresponding to the recognition points to obtain a set of potential medical services; When the set of potential medical services covers the medical services at the dissatisfied annotation points, regard the recognition point as the target recognition point; Take the user pain points corresponding to the potential medical services in the set of potential medical services of the target recognition point with the largest value as potential pain points; Visually display the medical services corresponding to the potential pain points to form a preview service.

8. An interactive system for a medical service terminal based on artificial intelligence according to claim 7, characterized in that, The obtaining of the target potential pain points based on the user's response includes the following steps: Based on visual recognition or voice recognition, regard the potential pain points corresponding to the preview service recognized by the user as the target potential pain points.

9. The interactive system of a medical service terminal based on artificial intelligence according to claim 8, characterized in that The correcting to obtain the medical service correction plan based on the target potential pain points and the dissatisfied annotation points includes the following steps: Obtain the medical services at the dissatisfied annotation points as the medical services to be replaced; Take the similarity between the medical service corresponding to the target potential pain point and the medical service to be replaced as the similarity between the target potential pain point and the medical service to be replaced; Obtain the minimum value of the similarity between the target potential pain point and at least one medical service to be replaced, and replace the medical service to be replaced corresponding to the minimum value of the similarity with the medical service corresponding to the target potential pain point to obtain the medical service correction plan.

Images