CN113782125B

CN113782125B - Clinic scoring method and device based on artificial intelligence, electronic equipment and medium

Info

Publication number: CN113782125B
Application number: CN202111095015.7A
Authority: CN
Inventors: 陈朝海
Original assignee: Shenzhen Ping An Smart Healthcare Technology Co ltd
Current assignee: Shenzhen Ping An Smart Healthcare Technology Co ltd
Priority date: 2021-09-17
Filing date: 2021-09-17
Publication date: 2023-08-08
Anticipated expiration: 2041-09-17
Also published as: CN113782125A

Abstract

The application relates to the technical field of artificial intelligence, and provides an artificial intelligence-based clinic scoring method, an artificial intelligence-based clinic scoring device, electronic equipment and a medium, wherein the method comprises the following steps: extracting evaluation words from the user comment text of the target clinic based on the natural language processing model to obtain the evaluation words; determining a scoring value based on the evaluation weight value tree, and calculating a first scoring value according to the scoring value; obtaining diagnosis record data based on the diagnosis data of the target clinic, and determining a second grading value based on the diagnosis record data, wherein the diagnosis record data comprises patient symptoms and patient medications corresponding to the patient symptoms; determining a plurality of environment dimensions corresponding to the medical environment data based on the medical environment data of the target clinic; determining a third grading value according to the plurality of environment dimensions; calculating a fusion score value based on the first score value, the second score value, and the third score value; based on the fused score values, a score for the target clinic is determined. The utility model provides an efficiency and the rate of accuracy of grading the clinic have been improved.

Description

Clinic scoring method and device based on artificial intelligence, electronic equipment and medium

Technical Field

The application relates to the technical field of artificial intelligence, in particular to an artificial intelligence-based clinic scoring method, an artificial intelligence-based clinic scoring device, electronic equipment and a medium.

Background

At present, the number of big and small clinics in China is huge, but the medical hardware conditions of the clinics and the technical level of medical staff are different, no authoritative evaluation standard and comprehensive authentication system exist in the aspects of medical hardware facilities, operation capability, operation flow, medical quality and the like of the clinics, and people do not have objective channels to know the actual medical level of the clinics when selecting the clinics to visit the doctor.

Currently, grading of clinics is usually completed by means of manual auditing, such as obtaining self-evaluation data obtained by self-evaluating questionnaires filled by related staff of the clinics, authenticating according to the self-evaluation data, or obtaining grading of clinics corresponding to a plurality of patients on line, and determining grading of the clinics based on the grading of the patients. However, manual auditing cannot make flexible comparative analysis for different clinics, and questionnaire accuracy filled out by related staff of the clinics is difficult to guarantee, so that real medical levels of the clinics cannot be accurately scored, and meanwhile, the scores of the clinics are determined based on the scores of patients only, so that the determined scores cannot represent the real medical levels of the clinics.

Disclosure of Invention

In view of the foregoing, it is desirable to provide an artificial intelligence-based clinic scoring method, apparatus, electronic device and medium, which score the medical level of a clinic from multiple dimensions based on artificial intelligence, thereby improving the efficiency and accuracy of clinic scoring.

In a first aspect, the present application provides an artificial intelligence based clinic scoring method comprising:

acquiring user comment text of a target clinic, and extracting evaluation words from the user comment text based on a trained natural language processing model to obtain the evaluation words;

determining a scoring value corresponding to the scoring word based on a preset scoring weight value tree, and calculating a first scoring value corresponding to the target clinic according to the scoring value;

obtaining diagnosis record data of a patient based on the diagnosis data of the target clinic, and determining a second grading value based on the diagnosis record data, wherein the diagnosis record data comprises patient symptoms and patient medications corresponding to the patient symptoms;

determining a plurality of environment dimensions corresponding to the medical environment data based on the medical environment data of the target clinic; determining a third grading value according to the plurality of environment dimensions;

Calculating a fusion score value corresponding to the target clinic based on the first score value, the second score value and the third score value;

a score for the target clinic is determined based on the fused score values.

According to an optional embodiment of the present application, the extracting the evaluation word from the user comment text based on the trained natural language processing model, where obtaining the evaluation word includes:

the method comprises the steps that a subword coding module based on a natural language processing model codes a user comment text to obtain a plurality of subwords, and each subword is characterized by a characterization module of the natural language processing model to obtain a characterization vector corresponding to each subword;

determining the mark of each sub word according to the characterization vector, and carrying out alignment processing on the plurality of sub words according to the mark of each sub word to obtain a characterization sequence;

and calculating the characterization sequence to obtain the matching degree of each sub word, and determining the evaluation word according to the matching degree.

According to an optional embodiment of the present application, the determining the tag of each sub-word according to the token vector, and performing alignment processing on the plurality of sub-words according to the tag of each sub-word to obtain the token sequence includes:

Determining a characterization vector corresponding to a first sub-word as a mark of the first sub-word;

calculating a vector difference value between a characterization vector corresponding to a next sub word and a characterization vector corresponding to a previous sub word to obtain a mark of the next sub word;

calculating the number of marks of each sub word, and judging whether the number of marks is preset with a threshold value or not;

when the number of the marks is larger than a preset threshold value, performing convolution operation on the marks to obtain target marks;

when the number of marks is equal to a preset threshold value, determining the marks as target marks;

and obtaining a characterization sequence according to the target mark.

According to an optional embodiment of the present application, the determining, based on a preset evaluation weight value tree, a score value corresponding to the evaluation word includes:

traversing an evaluation weight value tree according to the evaluation word, and determining a node corresponding to the evaluation word;

determining a node level corresponding to the node and a corresponding node value;

and determining a scoring value corresponding to the evaluation word based on the node level and the node value.

According to an alternative embodiment of the present application, the determining the second score value based on the diagnostic record data comprises:

Determining medication characteristic information corresponding to the medication of the patient based on the medication information corresponding to the medication of the patient;

calculating the medication characteristic information and the patient symptoms based on a cosine similarity algorithm to obtain the matching degree of the medication characteristic information and the patient symptoms;

determining a grade of disorder from the patient's disorder;

a second scoring value is determined based on the degree of matching and the grade of the condition.

According to an optional embodiment of the application, the determining a third score value according to the plurality of environment dimensions comprises:

determining a plurality of environment dimensions corresponding to the medical environment data;

determining dimension information corresponding to each environment dimension from the medical environment data according to the plurality of environment dimensions;

determining a measurement unit of each environmental dimension, and carrying out standardization processing on dimension information corresponding to each environmental dimension based on the measurement units to obtain a standard value;

determining the dimension type of each environment dimension;

generating a fusion standard value of the medical environment data based on the dimension type and the standard value;

generating an environment curve corresponding to the medical environment data according to the fusion standard value;

And calculating the similarity between the environment curve and a preset curve, and determining a third grading value based on the similarity.

According to an optional embodiment of the present application, the calculating the fusion score value corresponding to the target clinic based on the first score value, the second score value and the third score value includes:

constructing a scoring triangle based on the first scoring value, the second scoring value and the third scoring value;

calculating the gravity center of the scoring triangle, and obtaining fusion scoring coordinates corresponding to the target clinic based on the gravity center;

and determining the fusion score value corresponding to the target clinic based on the fusion score coordinates.

In a second aspect, the present application provides an artificial intelligence based clinic scoring device comprising:

the clinic evaluation processing module is used for acquiring user comment texts of the target clinic, and extracting evaluation words from the user comment texts based on the trained natural language processing model to obtain the evaluation words;

the first scoring calculation module is used for determining a scoring value corresponding to the evaluation word based on a preset evaluation weight value tree and calculating a first scoring value corresponding to the target clinic according to the scoring value;

The second score calculation module is used for obtaining diagnosis record data of a patient based on the diagnosis data of the target clinic and determining a second score value based on the diagnosis record data, wherein the diagnosis record data comprises symptoms of the patient and medication of the patient corresponding to the symptoms of the patient;

a third score calculation module, configured to determine, based on the medical environment data of the target clinic, a plurality of environment dimensions corresponding to the medical environment data; determining a third grading value according to the plurality of environment dimensions;

the fusion score calculating module is used for calculating a fusion score value corresponding to the target clinic based on the first score value, the second score value and the third score value;

and the clinic score calculating module is used for determining the score of the target clinic based on the fusion score value.

In a third aspect, the present application provides an electronic device comprising a processor and a memory, the processor for implementing the artificial intelligence based clinic scoring method when executing a computer program stored in the memory.

In a fourth aspect, the present application provides a computer readable storage medium having a computer program stored thereon, which when executed by a processor, implements the artificial intelligence based clinic scoring method.

In summary, according to the method, the device, the electronic equipment and the medium for scoring clinics based on artificial intelligence, the efficiency and the accuracy for determining the evaluation word can be improved by acquiring the user comment text of the target clinic, extracting the evaluation word from the user comment text based on the trained natural language processing model to obtain the evaluation word, and extracting the evaluation word from the user comment text by using the natural language processing model; then, based on a preset evaluation weight value tree, obtaining a score value corresponding to each evaluation word, calculating a first score value corresponding to the target clinic according to the score values, and traversing the evaluation weight tree to improve the accuracy and efficiency of determining the score values, so that the determined first score values are more accurate; obtaining diagnosis record data of a patient based on the diagnosis data of the target clinic, and determining a second grading value based on the diagnosis record data, wherein the diagnosis record data comprises symptoms of the patient and medication of the patient corresponding to the symptoms of the patient, obtaining medical environment data of the target clinic, and determining a third grading value based on the medical environment data; and then, calculating a fusion score value corresponding to the target clinic based on the first score value, the second score value and the third score value, determining the score of the target clinic based on the fusion score value, and obtaining fusion score values through the score values of three dimensions, wherein the fusion score values can comprise richer clinic information, so that the determined score is more accurate and more in line with the actual situation, and the effectiveness of the score is improved.

Drawings

Fig. 1 is a flow chart of an artificial intelligence based clinic scoring method provided in an embodiment of the present application.

Fig. 2 is a block diagram of an artificial intelligence based clinic scoring device according to a second embodiment of the present application.

Fig. 3 is a schematic structural diagram of an electronic device according to a third embodiment of the present application.

Detailed Description

In order that the above-recited objects, features and advantages of the present application will be more clearly understood, a more particular description of the application will be rendered by reference to specific embodiments thereof which are illustrated in the appended drawings. It should be noted that, in the case of no conflict, the embodiments of the present application and the features in the embodiments may be combined with each other.

Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein in the description of the present application is for the purpose of describing the embodiments in one alternative embodiment only and is not intended to be limiting of the present application.

The clinic scoring method based on the artificial intelligence is executed by the electronic equipment, and accordingly, the clinic scoring device based on the artificial intelligence is operated in the electronic equipment. The electronic device may include a cell phone, tablet computer, notebook computer, desktop computer, personal digital assistant, wearable device, and the like.

The embodiment of the application can process medical images based on artificial intelligence technology. Among these, artificial intelligence (Artificial Intelligence, AI) is the theory, method, technique and application system that uses a digital computer or a digital computer-controlled machine to simulate, extend and extend human intelligence, sense the environment, acquire knowledge and use knowledge to obtain optimal results.

Artificial intelligence infrastructure technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, operation/interaction systems, mechatronics, and the like. The artificial intelligence software technology mainly comprises a computer vision technology, a robot technology, a biological recognition technology, a voice processing technology, a natural language processing technology, machine learning/deep learning and other directions.

Example 1

Fig. 1 is a flow chart of an artificial intelligence based clinic scoring method provided in an embodiment of the present application. The method for scoring the clinic based on the artificial intelligence specifically comprises the following steps, the sequence of the steps in the flow chart can be changed according to different requirements, and some can be omitted.

S11, obtaining user comment texts of the target clinic, and extracting evaluation words from the user comment texts based on the trained natural language processing model to obtain the evaluation words.

For example, a target website may be determined, and user evaluation text corresponding to the target clinic may be obtained from the target website. For example, web crawler technology may be used to collect user evaluation text corresponding to a target clinic in a plurality of target websites. The target website may be a medical website. User evaluation texts corresponding to the target clinics are collected from the target websites, and the collection efficiency of the user evaluation texts can be improved.

The evaluation word is a word for representing user evaluation, such as good, bad, satisfactory, dissatisfied, good, dislike, favorites, offensive, and the like. The content corresponding to the evaluation word may be set according to the actual situation, and is not limited in any way.

The natural language processing model may be a BERT (Bidirectional Encoder Representations from Transformers) model or a modified model RoBERTa (Robustlyoptimized BERT Pretraining approach) model of the BERT model. The RoBERTa model achieves more advanced results by improving training tasks and data generation, training longer, using larger batches, using more data, etc.

In an alternative embodiment, before extracting the evaluation word from the user comment text based on the trained natural language processing model, the method further includes the steps of:

Obtaining a user comment sample, wherein the user comment sample comprises comment words and comment labels corresponding to the comment words; training a natural language processing model based on the user comment sample, and judging whether comment word extraction rate of the natural language processing model meets a preset expected value, wherein the natural language processing model is used for extracting comment words in the user comment sample; if the comment word extraction rate of the natural language processing model does not meet a preset expected value, adjusting model parameters of the natural language processing model; and training the natural language processing model by iteratively using the user comment sample until judging that the comment word extraction rate of the natural language processing model meets a preset expected value, stopping training the natural language processing model, and obtaining a natural language processing model after training.

The natural language processing model may include an input layer, a hidden layer, and an output layer, for example. The input layer is used for inputting comment texts to be identified, such as user comment samples; the hidden layer is used for carrying out text recognition on the input comment text and extracting comment words in the comment text; the output layer is used for outputting the comment word.

In an optional implementation manner, the extracting the evaluation word from the user comment text based on the trained natural language processing model, and obtaining the evaluation word includes:

And inputting the user comment text to an input layer of the trained natural language processing model, wherein the input layer inputs the user comment text to a subword coding module for coding, and a plurality of subwords (subwords) in the user comment text are output through the subword coding module. The sub word coding module inputs the plurality of sub words to the characterization module for further coding, and a characterization vector of each sub word is output through the characterization module, wherein the characterization vector is used for uniquely representing the sub word, and the characterization module adopts Byte-Pair Encoding (BPE) technology for coding.

The characterization module firstly maps each sub word into a 768-dimensional word vector according to a word list; mapping the position into a 768-dimensional position vector according to the position of each sub word in the comment text of the user; and finally, adding each element in the word vector and the position vector to be used as a representation vector corresponding to the sub word. The characterization module not only can efficiently capture semantic dependency relations of longer distances, but also can acquire bidirectional context information to generate vector representations with richer semantic information.

Because the characterization module encodes each word by adopting Byte-Pair Encoding (BPE) technology, the characterization vectors generated for each sub word are different, the sequence length of the characterization vectors of some sub words is the same as the sequence length corresponding to the sub word, and the sequence length of the characterization vectors of some sub words is longer than the sequence length corresponding to the sub word, so that the length of a coding sequence obtained based on the characterization vectors of a plurality of sub words is longer than the length of an input text sequence, the token (token) of each sub word is calculated, and the alignment processing is carried out on the characterization vectors of the plurality of sub words according to the token of each sub word, so that the sequence length of the aligned characterization vectors corresponding to each sub word can be the same, the length of the coding sequence obtained based on the characterization vectors of the plurality of sub words is consistent with the length of the comment text of a user, and the matching degree of each sub word is conveniently calculated by using a full-connection layer.

In an optional embodiment, the determining the tag of each sub-word according to the token vector, and performing alignment processing on the plurality of sub-words according to the tag of each sub-word to obtain the token sequence includes:

and obtaining a characterization sequence according to the target mark.

Wherein the first sub-word refers to a word located at a first position in the user comment text. The preset threshold may be set to 1, without any limitation.

In some embodiments, since the length of the characterization vector extracted through the natural language processing model is greater than the length of the user comment text, and the marks of each sub word are calculated through the characterization corresponding to each sub word, the characterization sizes of the sub words in the characterization vector can be rapidly determined, and finally, the marks with the number greater than a preset threshold are convolved, so that the simplification of the characterization of the sub word corresponding to the mark is realized, and therefore, the length of the characterization sequence obtained according to the target mark corresponding to each sub word is effectively ensured to be consistent with the length of the user comment text.

For example, the token sequence may be calculated based on the connection layer of the natural language processing model, resulting in a degree of matching for each subword. For example, the full concatenation layer may convert the representation of the subwords in the token sequence that contain the context information into a 2-dimensional representation and combine the subwords into phrases, resulting in a score for whether each phrase is an evaluation word. The score is used for indicating the matching degree of the phrase serving as the evaluation word. The larger the matching degree is, the larger the possibility that the phrase is an evaluation word is, the smaller the matching degree is, and the smaller the possibility that the phrase is the evaluation word is.

For example, the plurality of sub words may be ranked in a reverse order based on the matching degree, and a preset number of sub words in the plurality of sub words ranked in the reverse order may be obtained as the evaluation word.

S12, determining a scoring value corresponding to the scoring word based on a preset scoring weight value tree, and calculating a first scoring value corresponding to the target clinic according to the scoring value.

Illustratively, a plurality of evaluation words are acquired, and an evaluation level corresponding to each evaluation word is determined, wherein the evaluation level is used for representing the satisfaction degree of the user. For example, the higher the rating, the higher the user's corresponding satisfaction, and the lower the rating, the lower the user's corresponding satisfaction. The evaluation level corresponding to the evaluation word can be determined according to the emotion color corresponding to the evaluation word. For example, the evaluation level corresponding to the evaluation words "best", "most excellent" is higher than the evaluation level corresponding to the evaluation words "worst", "worst".

And constructing an evaluation weight value tree according to the evaluation grades corresponding to the evaluation words. The evaluation weight value tree comprises a root node and a plurality of leaf nodes, and each node comprises an evaluation word corresponding to the node and a node value corresponding to the node. The node corresponding to the evaluation word with high evaluation level is closer to the root node of the evaluation weight value tree, and the node value corresponding to the node is larger as the node level corresponding to the node is higher; the node corresponding to the evaluation word with low evaluation level is far away from the root node of the evaluation weight value tree, and the lower the node level corresponding to the node is, the smaller the node value corresponding to the node is.

Different weight values can be set according to node levels corresponding to different nodes. For example, the scoring value corresponding to the evaluation word may be determined based on the weight value of the node-to-node hierarchy multiplied by the node value corresponding to the node.

For example, if there are multiple evaluation words, multiple evaluation values may be determined based on a preset evaluation weight value tree, and the first evaluation value corresponding to the target clinic may be calculated based on an average value of the multiple evaluation values.

In an optional embodiment, the determining, based on the preset evaluation weight value tree, a score value corresponding to the evaluation word includes:

Different weight values can be set according to node levels corresponding to different leaf nodes. For example, the score value corresponding to the evaluation word may be determined based on the weight value of the node hierarchy corresponding to the leaf node multiplied by the node value corresponding to the leaf node.

And S13, obtaining diagnosis record data of a patient based on the diagnosis data of the target clinic, and determining a second grading value based on the diagnosis record data, wherein the diagnosis record data comprises patient symptoms and patient medication corresponding to the patient symptoms.

The diagnostic data includes diagnostic record data for the patient, such as a patient condition and a patient medication corresponding to the patient condition, and a second scoring value is determined based on the diagnostic data.

In an alternative embodiment, the determining a second score value based on the diagnostic record data includes:

determining a grade of disorder from the patient's disorder;

The medication characteristic information is used for representing applicable symptoms corresponding to the medication of the patient. For example, a medicine database may be preset, in which a mapping relationship between medicines and medication characteristic information is recorded. And inquiring the medicine database based on the medicine information to determine the medicine characteristic information corresponding to the medicine.

Based on a cosine similarity algorithm, calculating the medication characteristic information and the patient symptoms to obtain the matching degree of the medication characteristic information and the patient symptoms, so as to determine whether the medication of the patient accords with the patient symptoms or not, wherein the higher the matching degree is, the more the medication of the patient accords with the patient symptoms; the lower the matching degree is, the less the patient takes medicine and the disease symptoms of the patient are in agreement, and medical accidents are easy to happen. Based on the patient's symptoms, a corresponding symptom grade is calculated for the patient, the symptom grade is used for representing the severity of the disease, and the higher the symptom grade is, the more difficult the disease is to treat.

Illustratively, the calculating the medication characteristic information and the patient condition based on the cosine similarity algorithm, and obtaining the matching degree between the medication characteristic information and the patient condition includes: and obtaining the representation vectors of the medication characteristic information and the two texts of the patient symptoms, calculating cosine similarity of the two representation vectors, and determining the matching degree of the medication characteristic information and the patient symptoms based on the similarity. The larger the cosine similarity, the more similar the medication characteristic information is to the patient's condition, and the larger the matching degree is.

The second scoring value may be calculated based on a weight value corresponding to the degree of matching and the grade of the condition. For example, the degree of matching may be multiplied by a weight value corresponding to the grade of the condition to obtain a second scoring value. Wherein, the weight value corresponding to the disease level can be set according to the actual situation. For example, the higher the disease level, the smaller the corresponding weight may be set.

For example, if there is diagnostic data corresponding to a plurality of patients, a plurality of score values may be determined according to the matching degree and the disorder level, and an average value corresponding to the plurality of score values may be determined as a second score value.

S14, determining a plurality of environment dimensions corresponding to the medical environment data based on the medical environment data of the target clinic; and determining a third scoring value based on the plurality of environmental dimensions.

The medical environment data is used to represent the quality of the clinical environment and may include, among other things, the clinical temperature, the clinical oxygen content, the clinical carbon dioxide content, and the clinical humidity.

In an alternative embodiment, the determining a third scoring value based on the plurality of environmental dimensions includes:

determining the dimension type of each environment dimension;

The environmental dimensions may include a temperature dimension, an oxygen content dimension, a carbon dioxide content dimension, a humidity dimension, and the like. And determining dimension information corresponding to each environment dimension from the medical environment data, such as temperature measurement data corresponding to a temperature dimension, oxygen measurement data corresponding to an oxygen content dimension, carbon dioxide measurement data corresponding to a carbon dioxide content dimension and humidity measurement data corresponding to a humidity dimension.

And carrying out standardization processing on the dimension information corresponding to each environment dimension based on the measurement unit corresponding to each environment dimension, and obtaining the standard value by the data in the same format of the dimension information corresponding to each environment dimension.

Different dimension types can be provided with different weight values, and fusion standard values of the medical environment data are generated based on the weight values corresponding to the dimension types and the standard values. For example, a plurality of dimension types exist, a weight value corresponding to each dimension type is multiplied by a standard value corresponding to the dimension type to obtain a plurality of calculated values, and an average value of the plurality of calculated values is determined as a fusion standard value of the medical environment data.

For example, the environmental curve corresponding to the medical environment data may be generated according to the fusion standard value and the time corresponding to the fusion standard value. The time may be a date, a time of day, etc.

And S15, calculating a fusion score value corresponding to the target clinic based on the first score value, the second score value and the third score value.

For example, an average of the first score value, the second score value, and the third score value may be calculated, and the average may be determined as a fusion score value corresponding to the target clinic.

In an optional embodiment, the calculating the fusion score value corresponding to the target clinic based on the first score value, the second score value, and the third score value includes:

For example, a first weight value corresponding to the first score value, a second weight value corresponding to the second score value, and a third weight value corresponding to the third score value may be preset. And calculating coordinates corresponding to each grading value by using the weight corresponding to each grading value, and constructing a grading triangle based on the coordinates. For example, based on the first weight, the coordinate corresponding to the first score value is determined, for example, the value corresponding to the first score value is determined as X, the value obtained by multiplying the value corresponding to the first score value by the first weight value is determined as Y, and the coordinate corresponding to the first score value is obtained.

Illustratively, barycentric coordinates corresponding to the barycenter of the scoring triangle are determined, and the barycentric coordinates are determined as fusion scoring coordinates corresponding to the target clinic. For example, a value of the fusion score coordinate on the X-axis may be determined as a fusion score value corresponding to the target clinic; or calculating the distance value between the fusion score coordinate and the origin coordinate based on the X value and the Y value of the fusion score coordinate, and determining the distance value as the fusion score value corresponding to the target clinic.

S16, determining the score of the target clinic based on the fusion score value.

For example, the fusion score value may be numerically converted, including percentile conversion, scoring conversion, hierarchical conversion, and the like, and the converted value may be determined as the score of the target clinic.

According to the clinic scoring method based on artificial intelligence, the user comment text of the target clinic is obtained, and the evaluation word is extracted from the user comment text based on the natural language processing model after training is completed, so that the evaluation word is obtained, and the natural language processing model is used for extracting the evaluation word from the user comment text, so that the efficiency and the accuracy for determining the evaluation word can be improved; then, based on a preset evaluation weight value tree, obtaining a score value corresponding to each evaluation word, calculating a first score value corresponding to the target clinic according to the score values, and traversing the evaluation weight tree to improve the accuracy and efficiency of determining the score values, so that the determined first score values are more accurate; obtaining diagnosis record data of a patient based on the diagnosis data of the target clinic, and determining a second grading value based on the diagnosis record data, wherein the diagnosis record data comprises symptoms of the patient and medication of the patient corresponding to the symptoms of the patient, obtaining medical environment data of the target clinic, and determining a third grading value based on the medical environment data; and then, calculating a fusion score value corresponding to the target clinic based on the first score value, the second score value and the third score value, determining the score of the target clinic based on the fusion score value, and obtaining fusion score values through the score values of three dimensions, wherein the fusion score values can comprise richer clinic information, so that the determined score is more accurate and more in line with the actual situation, and the effectiveness of the score is improved.

Example two

In some embodiments, the artificial intelligence based clinic scoring device 20 may include a plurality of functional modules consisting of computer program segments. The computer program of the individual program segments in the artificial intelligence based clinic scoring device 20 may be stored in a memory of an electronic device and executed by at least one processor to perform (see fig. 1 for details) the functions of the artificial intelligence based clinic scoring method.

In this embodiment, the artificial intelligence based clinic scoring device 20 may be divided into a plurality of functional modules according to the functions it performs. The functional module may include: the system comprises a clinic evaluation processing module 201, a first score calculating module 202, a second score calculating module 203, a third score calculating module 204, a fusion score calculating module 205 and a clinic score calculating module 206. A module as referred to in this application refers to a series of computer program segments, stored in a memory, capable of being executed by at least one processor and of performing a fixed function. In the present embodiment, the functions of the respective modules will be described in detail in the following embodiments.

The clinic evaluation processing module 201 is configured to obtain a user comment text of a target clinic, and extract evaluation words from the user comment text based on a natural language processing model after training is completed, so as to obtain evaluation words.

In an alternative embodiment, the clinic evaluation processing module 201 performs evaluation word extraction on the user comment text based on the trained natural language processing model, and the obtaining the evaluation word includes:

In an alternative embodiment, the clinic evaluation processing module 201 determines a label of each sub-word according to the characterization vector, and performs alignment processing on the plurality of sub-words according to the label of each sub-word to obtain a characterization sequence, which includes:

and obtaining a characterization sequence according to the target mark.

Wherein the first sub-word refers to a word located at a first position in the user comment text.

The preset threshold may be set to 1, without any limitation.

The first score calculating module 202 is configured to determine a score value corresponding to the evaluation word based on a preset evaluation weight value tree, and calculate a first score value corresponding to the target clinic according to the score value.

In an alternative embodiment, the first score calculating module 202 determines, based on a preset evaluation weight value tree, a score value corresponding to the evaluation word, including:

And a second score calculation module 203, configured to obtain diagnosis record data of a patient based on the diagnosis data of the target clinic, and determine a second score value based on the diagnosis record data, where the diagnosis record data includes a patient condition and a patient medication corresponding to the patient condition.

In an alternative embodiment, the second score calculation module 203 determines the second score value based on the diagnostic record data comprises:

Determining a grade of disorder from the patient's disorder;

Based on a cosine similarity algorithm, calculating the medication characteristic information and the patient symptoms to obtain the matching degree of the medication characteristic information and the patient symptoms, so as to determine whether the medication of the patient accords with the patient symptoms or not, wherein the higher the matching degree is, the more the medication of the patient accords with the patient symptoms; the lower the matching degree is, the less the patient takes medicine and the disease symptoms of the patient are in agreement, and medical accidents are easy to happen.

Based on the patient's symptoms, a corresponding symptom grade is calculated for the patient, the symptom grade is used for representing the severity of the disease, and the higher the symptom grade is, the more difficult the disease is to treat.

A third score calculation module 204, configured to determine, based on the medical environment data of the target clinic, a plurality of environment dimensions corresponding to the medical environment data; and determining a third scoring value based on the plurality of environmental dimensions.

In an alternative embodiment, third score calculation module 204 determines a third score value from the plurality of environmental dimensions comprises:

determining the dimension type of each environment dimension;

And a fusion score calculating module 205, configured to calculate a fusion score value corresponding to the target clinic based on the first score value, the second score value and the third score value.

In an alternative embodiment, the calculating a fusion score value corresponding to the target clinic based on the first score value, the second score value, and the third score value by the fusion score calculating module 205 includes:

constructing a scoring triangle based on the first scoring value, the second scoring value and the third scoring value; calculating the gravity center of the scoring triangle, and obtaining fusion scoring coordinates corresponding to the target clinic based on the gravity center; and determining the fusion score value corresponding to the target clinic based on the fusion score coordinates.

And a clinic score calculating module 206, configured to determine a score of the target clinic based on the fused score value.

According to the clinic scoring device based on artificial intelligence, the user comment text of the target clinic is obtained, and the evaluation word is extracted from the user comment text based on the natural language processing model after training is completed, so that the evaluation word is obtained, and the natural language processing model is used for extracting the evaluation word from the user comment text, so that the efficiency and the accuracy for determining the evaluation word can be improved; then, based on a preset evaluation weight value tree, obtaining a score value corresponding to each evaluation word, calculating a first score value corresponding to the target clinic according to the score values, and traversing the evaluation weight tree to improve the accuracy and efficiency of determining the score values, so that the determined first score values are more accurate; obtaining diagnosis record data of a patient based on the diagnosis data of the target clinic, and determining a second grading value based on the diagnosis record data, wherein the diagnosis record data comprises symptoms of the patient and medication of the patient corresponding to the symptoms of the patient, obtaining medical environment data of the target clinic, and determining a third grading value based on the medical environment data; and then, calculating a fusion score value corresponding to the target clinic based on the first score value, the second score value and the third score value, determining the score of the target clinic based on the fusion score value, and obtaining fusion score values through the score values of three dimensions, wherein the fusion score values can comprise richer clinic information, so that the determined score is more accurate and more in line with the actual situation, and the effectiveness of the score is improved.

Example III

The present embodiment provides a computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the above-described embodiment of an artificial intelligence based clinic scoring method, such as S11-S16 shown in fig. 1:

s11, obtaining user comment text of a target clinic, and extracting evaluation words from the user comment text based on a natural language processing model with training completed, so as to obtain the evaluation words;

s12, determining a scoring value corresponding to the scoring word based on a preset scoring weight value tree, and calculating a first scoring value corresponding to the target clinic according to the scoring value;

s13, obtaining diagnosis record data of a patient based on the diagnosis data of the target clinic, and determining a second grading value based on the diagnosis record data, wherein the diagnosis record data comprises patient symptoms and patient medications corresponding to the patient symptoms;

s14, determining a plurality of environment dimensions corresponding to the medical environment data based on the medical environment data of the target clinic; determining a third grading value according to the plurality of environment dimensions;

s15, calculating a fusion score value corresponding to the target clinic based on the first score value, the second score value and the third score value;

Alternatively, the computer program, when executed by a processor, performs the functions of the modules/units in the above-described apparatus embodiments, e.g., modules 201-206 in fig. 2:

the clinic evaluation processing module 201 is configured to obtain a user comment text of a target clinic, and extract evaluation words from the user comment text based on a natural language processing model after training is completed, so as to obtain evaluation words;

the first score calculating module 202 is configured to determine a score value corresponding to the evaluation word based on a preset evaluation weight value tree, and calculate a first score value corresponding to the target clinic according to the score value;

the second score calculating module 203 is configured to obtain diagnosis record data of a patient based on the diagnosis data of the target clinic, and determine a second score value based on the diagnosis record data, where the diagnosis record data includes a patient condition and a patient medication corresponding to the patient condition;

the third score calculating module 204 is configured to determine, based on the medical environment data of the target clinic, a plurality of environment dimensions corresponding to the medical environment data; determining a third grading value according to the plurality of environment dimensions;

The fusion score calculating module 205 is configured to calculate a fusion score value corresponding to the target clinic based on the first score value, the second score value and the third score value;

the clinic score calculating module 206 is configured to determine a score of the target clinic based on the fused score value.

Example IV

Fig. 3 is a schematic structural diagram of an electronic device according to a third embodiment of the present application. In the preferred embodiment of the present application, the electronic device 3 comprises a memory 31, at least one processor 32, at least one transceiver 33 and a communication bus 34.

It will be appreciated by those skilled in the art that the configuration of the electronic device shown in fig. 3 is not limiting of the embodiments of the present application, and that either a bus-type configuration or a star-type configuration is possible, and that the electronic device 3 may also include more or less other hardware or software than illustrated, or a different arrangement of components.

In some embodiments, the electronic device 3 is a device capable of automatically performing numerical calculation and/or information processing according to a preset or stored instruction, and its hardware includes, but is not limited to, a microprocessor, an application specific integrated circuit, a programmable gate array, a digital processor, an embedded device, and the like. The electronic device 3 may further include a client device, where the client device includes, but is not limited to, any electronic product that can interact with a client by way of a keyboard, a mouse, a remote control, a touch pad, or a voice control device, such as a personal computer, a tablet computer, a smart phone, a digital camera, etc.

It should be noted that the electronic device 3 is only used as an example, and other electronic products that may be present in the present application or may be present in the future are also included in the scope of the present application and are incorporated herein by reference.

In some embodiments, the memory 31 has stored therein a computer program that, when executed by the at least one processor 32, performs all or part of the steps in the artificial intelligence based clinic scoring method as described. The Memory 31 includes Read-Only Memory (ROM), programmable Read-Only Memory (PROM), erasable programmable Read-Only Memory (EPROM), one-time programmable Read-Only Memory (One-time Programmable Read-Only Memory, OTPROM), electrically erasable rewritable Read-Only Memory (EEPROM), compact disc Read-Only Memory (Compact Disc Read-Only Memory, CD-ROM) or other optical disc Memory, magnetic tape Memory, or any other medium that can be used for computer-readable carrying or storing data.

Further, the computer-readable storage medium may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function, and the like; the storage data area may store data created from the use of blockchain nodes, and the like.

The blockchain referred to in the application is a novel application mode of computer technologies such as distributed data storage, point-to-point transmission, consensus mechanism, encryption algorithm and the like. The Blockchain (Blockchain), which is essentially a decentralised database, is a string of data blocks that are generated by cryptographic means in association, each data block containing a batch of information of network transactions for verifying the validity of the information (anti-counterfeiting) and generating the next block. The blockchain may include a blockchain underlying platform, a platform product services layer, an application services layer, and the like.

In some embodiments, the at least one processor 32 is a Control Unit (Control Unit) of the electronic device 3, connects the various components of the entire electronic device 3 using various interfaces and lines, and performs various functions of the electronic device 3 and processes data by running or executing programs or modules stored in the memory 31, and invoking data stored in the memory 31. For example, the at least one processor 32, when executing the computer programs stored in the memory, implements all or part of the steps of the artificial intelligence based clinic scoring method described in embodiments of the present application; or to implement all or part of the functionality of an artificial intelligence based clinic scoring device. The at least one processor 32 may be comprised of integrated circuits, such as a single packaged integrated circuit, or may be comprised of multiple integrated circuits packaged with the same or different functionality, including one or more central processing units (Central Processing unit, CPU), microprocessors, digital processing chips, graphics processors, combinations of various control chips, and the like.

In some embodiments, the at least one communication bus 34 is arranged to enable connected communication between the memory 31 and the at least one processor 32 or the like.

Although not shown, the electronic device 3 may further comprise a power source (such as a battery) for powering the various components, which may preferably be logically connected to the at least one processor 32 via a power management device, such that functions of managing charging, discharging, and power consumption are performed by the power management device. The power supply may also include one or more of any of a direct current or alternating current power supply, recharging device, power failure detection circuit, power converter or inverter, power status indicator, etc. The electronic device 3 may further include various sensors, bluetooth modules, wi-Fi modules, etc., which will not be described herein.

The integrated units implemented in the form of software functional modules described above may be stored in a computer readable storage medium. The software functional modules described above are stored in a storage medium and include instructions for causing a computer device (which may be a personal computer, an electronic device, or a network device, etc.) or processor (processor) to perform portions of the methods described in various embodiments of the present application.

In the several embodiments provided in this application, it should be understood that the disclosed apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the modules is merely a logical function division, and there may be other manners of division when actually implemented.

The modules described as separate components may or may not be physically separate, and components shown as modules may or may not be physical units, may be located in one place, or may be distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional module in each embodiment of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units can be realized in a form of hardware or a form of hardware and a form of software functional modules.

It will be evident to those skilled in the art that the present application is not limited to the details of the foregoing illustrative embodiments, and that the present application may be embodied in other specific forms without departing from the spirit or essential characteristics thereof. The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the application being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned. Furthermore, it will be obvious that the term "comprising" does not exclude other elements or that the singular does not exclude a plurality. Several of the elements or devices recited in the specification may be embodied by one and the same item of software or hardware. The terms first, second, etc. are used to denote a name, but not any particular order.

Finally, it should be noted that the above embodiments are merely for illustrating the technical solution of the present application and not for limiting, and although the present application has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solution of the present application may be modified or substituted without departing from the spirit and scope of the technical solution of the present application.

Claims

1. An artificial intelligence based clinic scoring method, the method comprising:

obtaining user comment text of a target clinic, extracting evaluation words from the user comment text based on a trained natural language processing model, and obtaining the evaluation words, wherein the method comprises the following steps: coding the user comment text based on a subword coding module of the natural language processing model to obtain a plurality of subwords, and representing each subword through a representation module of the natural language processing model to obtain a representation vector corresponding to each subword; determining the mark of each sub word according to the characterization vector, and carrying out alignment processing on the plurality of sub words according to the marks to obtain a characterization sequence; calculating the characterization sequence to obtain the matching degree of each sub word, and determining the evaluation word according to the matching degree;

determining a score for the target clinic based on the fused score values;

the determining the mark of each sub word according to the characterization vector, and performing alignment processing on the plurality of sub words according to the mark, so as to obtain a characterization sequence comprises the following steps: determining a characterization vector corresponding to a first sub-word as a mark of the first sub-word; calculating a vector difference value between a characterization vector corresponding to a next sub word and a characterization vector corresponding to a previous sub word to obtain a mark of the next sub word; calculating the number of marks of each sub word, and judging whether the number of marks is preset with a threshold value or not; when the number of the marks is larger than a preset threshold value, performing convolution operation on the marks to obtain target marks; when the number of marks is equal to a preset threshold value, determining the marks as target marks; and obtaining the characterization sequence according to the target mark.

2. The artificial intelligence based clinic scoring method according to claim 1, wherein the determining the scoring value corresponding to the scoring word based on a preset scoring weight value tree comprises:

3. The artificial intelligence based clinic scoring method according to claim 1, wherein the determining a second scoring value based on the diagnostic record data comprises:

determining a grade of disorder from the patient's disorder;

4. The artificial intelligence based clinic scoring method according to claim 1, wherein determining a third scoring value based on the plurality of environmental dimensions comprises:

determining the dimension type of each environment dimension;

5. The artificial intelligence based clinic scoring method according to any one of claims 1 to 4, wherein calculating the fusion score value corresponding to the target clinic based on the first score value, the second score value and the third score value comprises:

6. An artificial intelligence based clinic scoring device, the device comprising:

the clinic evaluation processing module is used for acquiring user comment texts of a target clinic and extracting evaluation words from the user comment texts based on a natural language processing model with training completed, and comprises the following steps: coding the user comment text based on a subword coding module of the natural language processing model to obtain a plurality of subwords, and representing each subword through a representation module of the natural language processing model to obtain a representation vector corresponding to each subword; determining the mark of each sub word according to the characterization vector, and carrying out alignment processing on the plurality of sub words according to the marks to obtain a characterization sequence; calculating the characterization sequence to obtain the matching degree of each sub word, and determining the evaluation word according to the matching degree;

a clinic score calculating module for determining a score of the target clinic based on the fused score value;

7. An electronic device comprising a processor and a memory, wherein the processor is configured to implement the artificial intelligence based clinic scoring method according to any one of claims 1 to 5 when executing a computer program stored in the memory.

8. A computer readable storage medium having stored thereon a computer program, which when executed by a processor implements the artificial intelligence based clinic scoring method according to any one of claims 1 to 5.