CN112700862B

CN112700862B - Determination method and device of target department, electronic equipment and storage medium

Info

Publication number: CN112700862B
Application number: CN202011567030.2A
Authority: CN
Inventors: 潘晶; 吕中伟; 陆斐; 顾佳怡; 沈满
Original assignee: Shanghai Timi Robot Co ltd
Current assignee: Shanghai Timi Robot Co ltd
Priority date: 2020-12-25
Filing date: 2020-12-25
Publication date: 2024-04-16
Anticipated expiration: 2040-12-25
Also published as: CN112700862A

Abstract

The invention discloses a method and a device for determining a target department, electronic equipment and a storage medium. The method comprises the following steps: the method comprises the steps of obtaining target texts for describing symptoms of users, and inputting the target texts into a department classification model to obtain matching degree values of each department and the target texts; inputting the target text into a target symptom extraction model to obtain symptom vocabulary corresponding to the target text; determining a department to be selected corresponding to the target text according to at least one symptom vocabulary to be processed and the target knowledge graph; according to the matching degree value corresponding to each department and at least one to-be-selected department, the target department is determined, so that the target department corresponding to the target text is determined, and the target department of the user is determined efficiently and accurately.

Description

Determination method and device of target department, electronic equipment and storage medium

Technical Field

The embodiment of the invention relates to the technical field of computers, in particular to a method and a device for determining a target department, electronic equipment and a storage medium.

Background

With the progress of medical technology, hospitals are gradually expanding in scale while improving the medical technology. For better service of patients, hospital departments are also increasingly finely classified. When a patient goes to a hospital for examination or medical treatment, the patient is mainly informed of the department to which he should go by consulting a diagnosis and separation table. However, the consultation efficiency of the triage table is gradually reduced due to the fact that the classification of departments is finer and finer at present. Moreover, the staff of the triage table can hardly remember the specific contents of all departments, can hardly guide the specific departments for the staff according to the symptom description of the patient, and the accuracy of the guided departments is low.

Disclosure of Invention

The invention provides a method and a device for determining a target department, electronic equipment and a storage medium, so as to realize high-efficiency and high-precision determination of the target department of a user.

In a first aspect, an embodiment of the present invention provides a method for determining a target department, including:

acquiring target text for describing a user condition;

inputting the target text into a department classification model, and determining the matching degree value of each department and the target text; the method comprises the steps of,

inputting the target text into a target symptom extraction model to obtain at least one symptom vocabulary to be processed corresponding to the target text, and determining at least one department to be selected corresponding to the target text according to the at least one symptom vocabulary to be processed and a target knowledge graph; the target knowledge graph comprises a disease type, a department and a relation between the disease type and symptom vocabulary;

and determining a target department according to the matching degree value corresponding to each department and the at least one department to be selected.

In a second aspect, an embodiment of the present invention further provides a device for determining a target department, including:

the target text acquisition module is used for acquiring target text for describing the symptoms of the user;

The matching degree value generation module is used for inputting the target text into a department classification model and determining the matching degree value of each department and the target text; the method comprises the steps of,

the department to be selected determining module is used for inputting the target file into a target symptom extraction model to obtain at least one symptom vocabulary to be processed corresponding to the target text, and determining at least one department to be selected corresponding to the target text according to the at least one symptom vocabulary to be processed and a target knowledge graph; the target knowledge graph comprises a disease type, a department and a relation between the disease type and symptom vocabulary;

the target department determining module is used for determining the target department according to the matching degree value corresponding to each department and the at least one department to be selected.

In a third aspect, an embodiment of the present invention further provides an electronic device, including:

one or more processors;

storage means for storing one or more programs,

the one or more programs, when executed by the one or more processors, cause the one or more processors to implement a method of determining a target subject matter as provided by any embodiment of the invention.

In a fourth aspect, embodiments of the present invention also provide a storage medium containing computer-executable instructions, which when executed by a computer processor, are used to perform a method of determining a target subject as provided by any of the embodiments of the present invention.

The embodiments of the above invention have the following advantages or benefits:

the method comprises the steps of obtaining target texts for describing symptoms of users, inputting the target texts into a department classification model, and determining matching degree values of each department and the target texts, so that the matching degree values of each department and the symptoms described by the users are obtained; inputting the target text into a target symptom extraction model to obtain at least one symptom vocabulary to be processed corresponding to the target text, thereby obtaining symptom vocabulary corresponding to the user description symptom; according to the at least one symptom vocabulary to be processed and the target knowledge graph, determining at least one department to be selected corresponding to the target text, and accordingly determining the department to be selected corresponding to the user description disorder; according to the matching degree value corresponding to each department and at least one department to be selected, determining a target department, thereby determining the target department corresponding to the user description disorder, improving the accuracy of the target department, and improving the efficiency of determining the patient department.

Drawings

In order to more clearly illustrate the technical solution of the exemplary embodiments of the present invention, a brief description is given below of the drawings required for describing the embodiments. It is obvious that the drawings presented are only drawings of some of the embodiments of the invention to be described, and not all the drawings, and that other drawings can be made according to these drawings without inventive effort for a person skilled in the art.

FIG. 1 is a flow chart of a method for determining a target department according to an embodiment of the invention;

FIG. 2 is a flow chart of a method for determining a target department according to a second embodiment of the present invention;

fig. 3 is a flow chart of a method for determining a target department according to a third embodiment of the present invention;

fig. 4 is a flow chart of a method for determining a target department according to a fourth embodiment of the present invention;

FIG. 5 is a flowchart of a method for determining a preferred target department according to a fourth embodiment of the present invention;

fig. 6 is a schematic structural diagram of a determining device for a target department according to a fifth embodiment of the present invention;

fig. 7 is a schematic structural diagram of an electronic device according to a sixth embodiment of the present invention.

Detailed Description

The invention is described in further detail below with reference to the drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting thereof. It should be further noted that, for convenience of description, only some, but not all of the structures related to the present invention are shown in the drawings.

Example 1

Fig. 1 is a flowchart of a method for determining a target department according to an embodiment of the present invention, where the method is applicable to a case of determining a target department based on disorder information described by a user, for example, in a case of determining that current disorder information corresponds to a target department in a case of more departments, the method may be performed by a determining device of a target department provided in an electronic device, and the device may be implemented by hardware and/or software.

The method specifically comprises the following steps:

before the technical solution of the present embodiment is described, an application scenario may be exemplarily described. The method for determining the target department disclosed by the embodiment of the invention can be integrated in the application program, a user can install the application program on the target terminal equipment, and when the user triggers the application program, the condition information can be edited in the condition editing page of the application program so as to determine the target department according to the edited condition information. Of course, the method for determining the target department may be integrated into an intelligent robot, and the intelligent robot may further include a question-answering system. The questions in the question-answering system may be the name, age, sex and condition information of the user, answer information of the visiting user may be received, and the method provided by the embodiment may be performed to determine the target department according to the answer information. In the embodiment, the advantage of integrating the robot into the robot is that the robot can be applied to various hospitals so as to determine a target department according to the answer of the user, thereby improving the registering convenience of the user for visiting the doctor; the integration into the application has the advantage that the versatility of determining the target department can be improved.

S110, acquiring target text for describing the symptoms of the user.

Wherein, the user disease refers to the disease condition information of the user. The target text may be information describing a user's condition. For example, the user condition is dyspepsia and the target text describing the user condition may be postprandial indigestion. The target text includes specific symptoms, and may also include the age, sex, affected parts, etc. of the user. In this embodiment, when determining a target department corresponding to a user disorder, a target text describing the user disorder may be acquired first.

Optionally, S110 includes: acquiring text content edited by a user in a content editing control of an application program, and generating a target text based on the text content; or, acquiring voice information fed back by the user to the question-answering system, and generating a target text according to the voice information.

The application program may refer to a program running on the terminal device, including a content editing control. The content editing controls include, but are not limited to, symptom-specific editing controls for enabling a user to enter symptom-specific information. Alternatively, the content editing controls may further include an age editing control, a sex editing control, an affected part editing control, and the like. The age editing control is used for enabling a user to input age; the gender editing control is used for enabling a user to enter gender; the diseased site editing control is used to allow a user to enter a diseased site.

The text content in the embodiment may be text information input by the user in the content editing control, or may be text information selected by the user in the content editing control based on a preset selection tag. The target text may be text information integrated from text content.

For example, the application may present a content editing interface to the user upon detecting that the user triggers the referral control, wherein the content editing interface includes a content editing control to enable the user to edit the symptom content in the content editing control of the content editing interface. In one embodiment, text content edited by a user in a content editing control may be obtained upon detection of a user triggering a confirmation submission control.

The question and answer system in this embodiment may be mounted on an intelligent robot. The question-answering system may be a system for asking a user. For example, the question-answering system may query the user upon detecting that the user triggers a corresponding question key, i.e. play a voice prompting the user to describe the condition, for example: and (3) detecting what symptoms you have, and acquiring voice information fed back by the user after the user answers. In this embodiment, the target text may be generated by converting voice information fed back by the user in real time through a voiceprint matching method.

In these optional embodiments, the target text is obtained through the content editing control of the application program or the question-answering system of the intelligent robot, so that the diversity of the target text obtaining method is realized, and the experience of the user is improved.

S120, inputting the target text into a department classification model, and determining the matching degree value of each department and the target text.

The department classification model is a model trained in advance according to training sample data or a referenced existing trained model, and is used for predicting a department classification result to which the target text belongs. Illustratively, the department classification model may be one of a trained supervised learning convolutional neural network model, a support vector regression model, a deep neural network model, a recurrent neural network model, and the like.

The input of the department classification model is a target text, and the output comprises each department and a matching degree value corresponding to each department. In this embodiment, the matching degree value may be a probability value that the target text belongs to the department, or the matching degree value may also be a correlation degree value between the target text and the department, but not limited to this, and in other embodiments, the matching degree value may also be a converted probability value or a converted correlation degree value, for example, the converted probability value and the converted correlation degree value may be represented by a pixel value of an image or a size of a preset shape.

For example, if the number of departments is 100, after the target text is input into the pre-trained department classification model, the department classification model outputs probability values corresponding to 100 departments.

Specifically, the training process of the department classification model is as follows: obtaining training sample data, training a department classification model based on the training sample data, wherein the training sample data comprises a large number of pre-acquired texts and labels corresponding to the affiliated departments, after the training sample data is input into the department classification model to be trained, the department classification model to be trained can output a plurality of probability values, the probability values and the values corresponding to the department labels can be processed based on a loss function in the department classification model to be trained according to the department labels and the probability values corresponding to the department labels, model parameters in the department classification model to be trained are corrected based on the processing results, the operations are repeatedly executed, namely the training sample data are repeatedly executed and input into the department classification model to be trained, and when the loss function in the model is detected to be converged, the department classification model to be trained obtained through training can be used as the department classification model which can be finally used.

The processing of the target text based on the trained department classification model may be: after the target text is input into the department classification model, the department classification model can output probability values between all departments and the target text, namely matching degree values between the departments and the target text, so that the final target departments can be determined based on the matching degree values. The training of the department classification model is completed by inputting the pre-collected texts into the department classification model to be trained, and correcting the probability value of the department actually output by the department classification model to be trained based on the expected output corresponding to each text, namely the label of the department to which the corresponding text belongs, so that the matching degree value corresponding to each department is obtained after inputting the target text into the trained department classification model.

S130, inputting a target text into a target symptom extraction model to obtain at least one symptom vocabulary to be processed corresponding to the target text, and determining at least one department to be selected corresponding to the target text according to the at least one symptom vocabulary to be processed and a target knowledge graph; the target knowledge graph comprises the relation between the symptom type and the department and the relation between the symptom type and the symptom vocabulary.

The target symptom extraction model is pre-trained and is used for extracting a model of symptom vocabulary in a target text. And taking the vocabulary extracted by the target symptom extraction model as the vocabulary of the symptom to be processed. The target knowledge graph is pre-constructed to determine the data network of the department to be selected based on the symptom-like words to be processed. The specific process of constructing the target knowledge graph can be as follows: at least one triplet is created first, for example, the triplet may be: symptom vocabulary-relation-symptom type, also can be symptom type-relation-department to be selected, namely, creating relation between symptom vocabulary and symptom type, and relation between symptom type and department. Wherein the disease type refers to specific type of disease, such as appendicitis, gastrorrhagia or upper respiratory tract infection. The target knowledge graph can process the to-be-processed symptom words output by the symptom word extraction model to determine corresponding departments to be selected.

It should be noted that the target knowledge graph is constructed based on a priori knowledge, that is, the symptom vocabulary-relation-symptom type and symptom type-relation-department to be selected in the target knowledge graph are determined based on a large amount of actual historical inquiry data or according to the content in the medical textbook. Therefore, the accuracy of the department to be selected corresponding to the target text determined based on the target knowledge graph is higher.

The target symptom extraction model in the present embodiment may be any one of a convolutional neural network model with supervised learning, a support vector regression model, a deep neural network model, a cyclic neural network model, and the like. The training process of the target symptom extraction model comprises the following steps: training sample data is obtained, and the target symptom extraction model is trained based on the training sample data. Wherein, training sample data comprises a plurality of pre-collected symptom description texts and symptom vocabularies in the symptom description texts. After the training of the target symptom extraction model is completed, the target text is input into the trained target symptom extraction model, and core keywords corresponding to the target text, namely symptom vocabulary to be processed, can be obtained.

Specifically, after the target knowledge graph is constructed, the symptom vocabulary to be processed is input into the constructed target knowledge graph, the symptom type corresponding to the symptom vocabulary to be processed is determined based on the relation between the symptom vocabulary and the symptom type, and further, the department to be selected corresponding to the symptom type is determined based on the relation between the symptom type and the department, so that the department to be selected corresponding to the symptom vocabulary to be processed is obtained based on the target knowledge graph.

Note that, in this embodiment, the execution order of S120 and S130 is not separately and may be executed at the same time as S120 and S130.

And S140, determining a target department according to the matching degree value corresponding to each department and at least one department to be selected.

In this embodiment, the target department may be a department for which a corresponding target text is finally determined. The number of departments to be selected may be one or more. For example, the number of departments to be selected is one, and if the matching degree value of the departments to be selected is higher than a preset threshold, the departments to be selected may be determined as target departments. For example, the department to be selected is an otorhinolaryngology department, the matching degree value corresponding to the otorhinolaryngology department is 0.92, the preset threshold value is 0.90, and the target department is an otolaryngology department.

The number of departments to be selected can be two or more, and at this time, the department with the highest matching degree value with the target text can be selected from all the departments to be selected as the target departments. For example, the matching degree value corresponding to each department includes: otorhinolaryngological department-0.80, infectious/infectious department-0.15, oncology department-0.05, general surgery department-0, gastroenterology-0, neurology-0, hematology 0; the departments to be selected include otorhinolaryngology department and infection/infection department, and the target department is otorhinolaryngology department.

According to the technical scheme, the target text for describing the symptoms of the user is obtained, the target text is input into a pre-trained department classification model, and the matching degree value of each department and the target text is determined; inputting the target text into a pre-trained target symptom extraction model to obtain at least one symptom vocabulary to be processed corresponding to the target text; determining at least one department to be selected corresponding to the target text according to at least one symptom vocabulary to be processed and a pre-constructed target knowledge graph; according to the matching degree value corresponding to each department and at least one to-be-selected department, determining the target department corresponding to the target text, thereby determining the target department corresponding to the user, and realizing high-efficiency and high-precision determination of the target department of the user.

Example two

Fig. 2 is a schematic flow chart of a method for determining a target department according to a second embodiment of the present invention, where on the basis of the foregoing embodiments, a department classification model includes a word segmentation sub-model, a word vector conversion sub-model, and a classification matching degree value output sub-model, and accordingly, a specific implementation of "inputting a target text into a department classification model to determine a matching degree value between each department and the target text" may refer to a technical scheme of the embodiment. Wherein the explanation of the same or corresponding terms as those of the above embodiments is not repeated herein.

Referring to fig. 2, the method for determining a target department provided in this embodiment includes:

s210, acquiring target text for describing the symptoms of the user.

S220, dividing the target text into at least one word to be processed based on the word segmentation sub-model, and eliminating preset stop words in the word to be processed to obtain a word to be used; converting the vocabulary to be used through a word vector conversion sub-model, and determining word vectors of the vocabularies to be used; and inputting the word vector into the classification matching degree value output submodel to obtain the matching degree value between the target text and each department.

The word segmentation sub-model may refer to a network sub-model, such as a hidden markov model, for determining word segmentation results of the target text. The object file may be divided into a plurality of words to be processed based on the word segmentation sub-model. The preset stop words are words which are preset and have no actual meaning, such as 'woolen', 'approximately', 'possible' or 'having the same' and the like. The vocabulary to be used refers to the vocabulary except for the preset stop words in the vocabulary to be processed. In this embodiment, the word vector conversion sub-model may be a network sub-model for determining word vectors of words to be used, such as a language processing model of continuous word packages or Skip-Gram. Each vocabulary to be used can be converted into a corresponding vector based on a word vector dictionary and/or a word vector conversion sub-model, and the method has the advantages that the vector is conveniently processed by a subsequent classification matching degree value output sub-model to obtain at least one classification result value, and the classification result value can be a probability value corresponding to each department. The classification matching degree value output submodel may be an input-multiple-output submodel, for example, the total number of departments is 100, after the word vector of the target text is input into the submodel, 100 output results may be obtained, and the output results may be 0 or 1, that is, the output results are values between [0,1 ].

Specifically, inputting a target text into a word segmentation sub-model to obtain a word to be processed corresponding to the target text, and removing preset stop words in the word to be processed to obtain a word to be used corresponding to the target text; inputting the vocabulary to be used into a word vector conversion sub-model to obtain word vectors of the vocabulary to be used; and inputting the word vectors of each vocabulary to be used into the classification matching degree value output submodel to obtain the probability value between the target text and each department.

S230, inputting the target text into a target symptom extraction model to obtain at least one symptom vocabulary to be processed corresponding to the target text, and determining at least one department to be selected corresponding to the target text according to the at least one symptom vocabulary to be processed and the target knowledge graph; the target knowledge graph comprises the relation between the symptom type and the department and the relation between the symptom type and the symptom vocabulary.

In this embodiment, the symptom type is determined based on the constructed target knowledge graph, and further, the department to be selected is determined based on the symptom type.

Specifically, the symptom vocabulary to be processed can be searched in the target knowledge graph, and the symptom type corresponding to the symptom vocabulary to be processed can be obtained based on the relation between the symptom vocabulary and the symptom type; and obtaining the department corresponding to the disease type, namely the department to be selected, based on the relation between the disease type and the department, thereby realizing the determination of the department to be selected corresponding to the target text.

Optionally, determining at least one department to be selected corresponding to the target text according to the at least one symptom vocabulary to be processed and the target knowledge graph, including: for each symptom vocabulary to be processed, determining a similarity value between the current symptom vocabulary to be processed and each symptom vocabulary in the target knowledge graph, and based on the similarity value, obtaining the target symptom vocabulary; and aiming at each target symptom vocabulary, searching the symptom type associated with the current target symptom vocabulary and at least one department to be selected associated with the symptom type in the target knowledge graph.

In this embodiment, the similarity value is used to represent the degree of similarity between the symptom vocabulary currently to be processed and each symptom vocabulary in the target knowledge graph. The similarity value may be determined based on a vector between the two words, such as determining cosine similarity of the two words, or based on an edit distance between the two words. The target symptom vocabulary can be the symptom vocabulary with the highest similarity value with the current symptom vocabulary to be processed in the target knowledge graph. The reasons and benefits of determining the target symptom vocabulary are: because the extracted current symptom vocabulary to be processed possibly has a certain error with the target knowledge graph, in order to avoid the error, the target symptom vocabulary corresponding to the current symptom vocabulary to be processed can be determined, and the conversion has the advantage that the accuracy of determining the departments to be selected in the target knowledge graph can be improved.

Specifically, the target knowledge graph comprises relations between each symptom type and each symptom word, and the symptom word with the highest similarity value is used as the target symptom word by calculating the similarity value between the symptom word to be processed currently and each symptom word in the target knowledge graph, so that the effect of mapping the symptom word to be processed currently to the target knowledge graph is achieved. After the target symptom vocabulary is determined, determining the symptom type associated with the target symptom vocabulary according to the relation between the symptom type and the symptom vocabulary in the target knowledge graph, and further determining the department associated with the symptom type associated with the target symptom vocabulary based on the relation between the symptom type and the department. The number of the target symptom vocabulary determined according to the current symptom processing vocabulary can be one or more, and correspondingly, the number of departments to be selected can also be a plurality. Further, since there may be a plurality of departments associated with a certain symptom, for example, the departments associated with thyroid may be endocrinology and thyroid, even one target symptom vocabulary may be associated with a plurality of departments to be selected.

In these optional embodiments, the similarity value between the current symptom vocabulary to be processed and each symptom vocabulary in the target knowledge graph is determined for each symptom vocabulary to be processed, and the target symptom vocabulary is based on the similarity value, so that mapping of the current symptom vocabulary to be processed into the target symptom vocabulary in the target knowledge graph is realized, and the situation that the current symptom vocabulary to be processed cannot be queried in the target knowledge graph is avoided, and therefore the situation that a target department cannot be determined is avoided.

S240, determining a target department according to the matching degree value corresponding to each department and at least one department to be selected.

According to the technical scheme, a target text is divided into at least one word to be processed based on a word segmentation sub-model, and preset stop words in the word to be processed are removed to obtain words to be used; converting the vocabulary to be used through a word vector conversion sub-model, and determining word vectors of the vocabularies to be used; the word vector is input into the classification matching degree value output submodel to obtain the matching degree value between the target text and each department, so that the matching degree value between the target text and each department is accurately obtained, and the accuracy of the target department is improved.

Example III

Fig. 3 is a schematic flow chart of a method for determining a target department according to a third embodiment of the present invention, where the number of departments to be selected may be multiple based on the foregoing embodiments, and if the number of departments to be selected includes multiple, a specific implementation manner of determining the target department may be referred to the technical solution of the embodiment. Wherein the explanation of the same or corresponding terms as those of the above embodiments is not repeated herein.

Referring to fig. 3, the method for determining a target department provided in this embodiment includes:

S310, acquiring target text for describing the symptoms of the user.

S320, inputting the target text into the department classification model, and determining the matching degree value of each department and the target text.

S330, inputting the target text into a target symptom extraction model to obtain at least one symptom vocabulary to be processed corresponding to the target text, and determining at least one department to be selected corresponding to the target text according to the at least one symptom vocabulary to be processed and the target knowledge graph; the target knowledge graph comprises the relation between the symptom type and the department and the relation between the symptom type and the symptom vocabulary.

S340, determining a target department to be determined corresponding to the matching degree value higher than a preset matching degree threshold value; and when the fact that the target department to be determined is included in the at least one department to be selected is detected, the target department to be determined is taken as the target department.

The preset matching degree threshold is preset, for example, may be 0.9. According to the matching degree value and the preset matching degree threshold value of each department output by the department classification model, part of departments can be screened out from all departments, namely, the target departments to be determined. At this time, the screened target departments to be determined are departments with the matching degree value higher than the preset matching degree threshold. The number of target departments to be determined may be one or more.

It should be noted that, the preset matching degree threshold value may be set according to actual requirements, and the specific numerical value of the preset matching degree threshold value is not limited in the present application.

Specifically, a department with a matching degree value higher than a preset matching degree threshold value is determined as a target department to be determined. Meanwhile, at least one department to be selected can be determined according to a pre-constructed target knowledge graph. If the department to be selected contains the target department to be determined, the target department to be determined can be used as the target department. The preset matching degree threshold is 0.9, the departments with the probability value higher than 0.9 are determined to be thyroid departments according to the probability values corresponding to the departments, and at least one department to be selected, which is determined based on the target knowledge graph, comprises endocrinology and thyroid departments. At this time, the target subject to be determined is a thyroid subject, and the subject to be selected includes a thyroid subject, so that the thyroid subject can be determined as the target subject.

Optionally, on the basis of S310-S330, S340 may also be: when the number of at least one department to be selected is detected to be one, and the matching degree value corresponding to the department to be selected is smaller than the preset matching degree threshold value, the department to be selected is taken as the target department.

Specifically, the number of departments to be selected determined through the pre-constructed target knowledge graph is only one, and at this time, whether the matching degree value of each department to be selected is smaller than the preset matching degree threshold value or not, the departments to be selected can be used as target departments. The reason is that: the target knowledge graph is constructed based on a large amount of actual historical inquiry data or according to the content in the medical textbook, and the result obtained based on the target knowledge graph is accurate, namely the accuracy of the department to be selected corresponding to the target text determined by the target knowledge graph is higher.

Optionally, if the matching degree value corresponding to the department to be selected is greater than the preset matching degree threshold, the department to be selected can be further determined to be the target department, and the determined result is more accurate.

That is, when the number of departments to be selected is one, the matching degree value of the departments to be selected may not be considered.

According to the technical scheme, when the matching degree value is higher than the preset matching degree threshold value, a target department to be determined is corresponding to the matching degree value; when at least one department to be selected is detected to comprise the target department to be determined, the target department to be determined is taken as the target department, so that the accurate determination of the target department is realized, and the accuracy of the target department is improved.

Example IV

Fig. 4 is a flow chart of a method for determining a target department according to a fourth embodiment of the present invention, where, based on the foregoing embodiments, the matching degree values corresponding to the departments are all smaller than a preset matching degree threshold, and the number of the at least one department to be selected includes a plurality of or is empty, and in this case, the technical scheme of the embodiment may be adopted to determine the target department. Wherein the explanation of the same or corresponding terms as those of the above embodiments is not repeated herein.

S410, acquiring target text for describing the symptoms of the user.

S420, inputting the target text into the department classification model, and determining the matching degree value of each department and the target text.

S430, inputting the target text into a target symptom extraction model to obtain at least one symptom vocabulary to be processed corresponding to the target text, and determining at least one department to be selected corresponding to the target text according to the at least one symptom vocabulary to be processed and the target knowledge graph; the target knowledge graph comprises the relation between the symptom type and the department and the relation between the symptom type and the symptom vocabulary.

S440, if the matching degree values corresponding to the departments are smaller than the preset matching degree threshold, and the number of the at least one department to be selected comprises a plurality of or is empty, displaying at least one corpus to be selected associated with the target text.

The corpus to be selected may refer to text information associated with the target text. The corpus to be selected is preset corpus corresponding to various disorder information.

Specifically, the corpus to be selected may be supplementary information of the target text, synonymous substitute information of the target text, or similar text information of the target text. For example, the target text is "abdominal pain", and the corpus to be selected may be "intermittent pain", "abdominal pain", or "stomach pain", or the like.

In this embodiment, when the number of departments to be selected is empty, or the number of departments to be selected is multiple, and the matching degree values corresponding to the multiple departments to be selected are all smaller than the preset matching degree threshold, a page or a dialog box for selecting the corpus to be selected may be popped up on the display interface of the target client, so that the user triggers to select at least one corpus to be selected associated with the target text.

S450, updating the target text according to the corpus to be selected and the target text triggered by the user.

The number of the corpus to be selected triggered by the user can be one or more. The updated target text may be regenerated to more closely match the textual information of the user's condition, i.e., the regenerated text based on the original target text and the user-triggered corpus, and may be used to redefine the target department.

In this embodiment, the updated target text may be obtained by supplementing the vocabulary in the target text based on the expectation to be selected, or may be obtained by combining the target text with the corpus to be selected. For example, the target text and each corpus to be selected in the above example, the corresponding updated target text includes: intermittent abdominal pain, or stomach pain.

S460, inputting the updated target text into the department classification model and the target symptom model to obtain a matching degree value corresponding to the updated target text and at least one department to be selected, and repeatedly displaying at least one corpus to be selected associated with the updated target text and re-updating the target text according to the corpus to be selected and the updated target text triggered by the user if the matching degree value corresponding to each department is smaller than a preset matching degree threshold and the number of the at least one department to be selected comprises a plurality of or is empty.

Specifically, the matching degree value of each department and the updated target text is obtained by inputting the updated target text into the department classification model; and (3) inputting the updated target text into the target symptom model to obtain at least one symptom vocabulary to be processed corresponding to the updated target text, and determining at least one department to be selected corresponding to the updated target text based on the at least one vocabulary to be processed and a pre-constructed target knowledge graph, so that the target departments are determined according to the matching degree value corresponding to each department and the at least one department to be selected, namely repeatedly executing S110 to S140 or S410 to S430.

It should be noted that, if the matching degree value corresponding to each department is smaller than the preset matching degree threshold, and the number of at least one department to be selected corresponding to the updated target text includes a plurality of or is empty, at least one corpus to be selected associated with the updated target text may be continuously displayed, so as to re-update the target text according to the corpus to be selected triggered by the user and the updated target text.

It should be noted that, if the target department is determined based on the updated target text, the repeated execution of the above steps may be stopped.

And when the target department is still not determined based on the updated target text, the target text can be updated again for a plurality of times and the steps are repeatedly executed, wherein if the number of times of repeated execution reaches a preset threshold value, it is possible that the target text input by the user is a text without practical meaning, and the corresponding target department can not be determined at the moment.

Optionally, the method for determining the target department further comprises: when detecting that the actual circulation times of the updated target text reach the preset circulation times threshold value according to the corpus to be selected and the target text triggered by the user repeatedly, and the number of at least one department to be selected is empty and departments with the matching degree value higher than the preset matching degree threshold value exist, determining the target departments according to the departments corresponding to the departments higher than the preset matching degree threshold value.

The preset cycle number threshold may be used to define the number of times that the user selects the corpus to be selected. The preset cycle number threshold may be set according to actual requirements, and the setting of a specific value of the preset cycle number threshold is not limited. The actual number of loops may refer to the number of times the update target text is repeatedly determined.

Specifically, when the determined number of departments to be selected is empty and departments with the matching degree value higher than a preset matching degree threshold exist, determining the departments with the matching degree value higher than the preset matching degree threshold as target departments. It should be noted that the number of departments with the matching degree value higher than the preset matching degree threshold may be one or more, and if there are multiple departments with the matching degree value higher than the preset matching degree threshold, determining the department with the highest matching degree value as the target department. If the target department still cannot be determined according to the current updated target text, the failure determination information can be fed back, and at the moment, the user can be reminded that the information is invalid or the user can be reminded of the feedback information of consultation to the consultation desk.

In the actual application process, when the number of actual cycles reaches the preset cycle number threshold, the number of departments to be selected may still include a plurality of situations, and the processing manner at this time may be: when detecting that the actual circulation times of the updated target text reach a preset circulation times threshold value according to the corpus to be selected and the target text triggered by the user repeatedly and the number of at least one department to be selected comprises a plurality of departments, taking the at least one department to be selected as the target department.

That is, when it is detected that the actual cycle number has reached the preset cycle number threshold, the number of departments to be selected still includes a plurality of departments, and the determined plurality of departments to be selected may be fed back to the corresponding terminal device, so that the user determines the target departments according to the fed back departments to be selected.

Optionally, the method for determining the target department further comprises: and feeding back the target department to the target client or the target terminal equipment.

The target terminal device may be a communication terminal to which the application program belongs, such as an electronic device including a mobile phone, an intelligent watch, or a tablet, or may be an intelligent robot. The target client may be an application program that performs the above-described determination method of the target department.

Specifically, the target department may be fed back to the communication terminal bound to the user in various manners, such as a short message notification, a WeChat public number notification, a mail notification, or a web page link, which is not limited in this application. In addition, the information of the target department can be fed back in a voice broadcasting mode. The target department is fed back to the target client or the target terminal equipment to remind the user that the department is determined, so that the experience of the user is improved.

Illustratively, as shown in FIG. 5, a flow chart of a preferred method of determining a target subject is shown. The method comprises the following specific steps:

step 1: and acquiring the target text.

Step 2: and determining departments to be selected corresponding to the target text, and matching degree values of the departments and the target text.

Step 3: judging whether the number of departments to be selected is empty or not; if not, executing the step 4; if yes, go to step 8.

Step 4: judging whether the number of departments to be selected is 1; if yes, executing the step 5; if not, executing step 6.

Step 5: taking the department to be selected as a target department.

Specifically, when the number of departments to be selected is 1, the departments to be selected are determined as target departments.

Step 6: judging whether the matching degree values corresponding to at least two departments to be selected are smaller than a preset matching degree threshold value or not; if yes, executing the step 8; if not, go to step 7.

Step 7: and taking the department to be selected with the highest matching degree value in the at least two departments to be selected as a target department.

Specifically, if the matching degree value corresponding to the to-be-selected department among the at least two to-be-selected departments is not smaller than the preset matching degree threshold, taking the to-be-selected department with the highest matching degree value among the at least two to-be-selected departments as the target department.

Step 8: updating the target text according to the corpus to be selected and the target text triggered by the user; meanwhile, the actual number of cycles is increased by 1.

Wherein the initial value of the actual number of cycles is 0. And updating the target text according to the to-be-selected prediction triggered by the user and the target text when the number of departments to be selected corresponding to the target text is 0 or the number of departments to be selected is more than 2 and the matching degree value of each department to be selected is smaller than the preset matching degree threshold value.

Step 9: judging whether the actual cycle times are smaller than a preset cycle times threshold value or not; if yes, returning to the execution step 1; if not, step 10 is performed.

Step 10: determining departments to be selected corresponding to the updated target text, and judging whether the number of the departments to be selected is empty or not; if not, executing the step 11; if yes, go to step 12.

Specifically, when the actual cycle number reaches a preset cycle number threshold, the number of departments to be selected is judged.

Step 11: the target department is determined based on the at least one department to be selected.

When the actual circulation times reach a preset circulation times threshold value and the number of departments to be selected is a plurality of, selecting the department to be selected with the highest matching degree value from the departments to be selected as a target department; and when the number of departments to be selected is 1, determining the departments to be selected as target departments.

Step 12: judging whether the matching degree values corresponding to all departments are lower than a preset matching degree threshold value or not; if yes, executing step 13; if not, step 14 is performed.

Step 13: and feeding back the failure information.

Step 14: and determining the departments with the matching degree values higher than the preset matching degree threshold as target departments.

Specifically, when the actual cycle number reaches a preset cycle number threshold and the number of departments to be selected is zero, if departments with the matching degree value higher than the preset matching degree threshold exist, determining the departments with the matching degree value higher than the preset matching degree threshold as target departments; if no department with the matching degree value higher than the preset matching degree threshold value exists, feeding back failure determination information to prompt a user that the corresponding target department cannot be determined according to the target text.

According to the technical scheme, when the matching degree value corresponding to each department is smaller than the preset matching degree threshold value and the number of at least one department to be selected comprises a plurality of or is empty, at least one corpus to be selected which is associated with the target text is displayed; according to corpus to be selected and target texts triggered by a user, a matching degree value and a department to be selected corresponding to the updated target texts are obtained again, and according to the matching degree value and at least one department to be selected, the target department is determined, so that the target texts are updated when the target departments cannot be determined, the target departments are determined again based on the updated target texts, and the situation that the target departments corresponding to the user cannot be determined is avoided.

Example five

Fig. 6 is a schematic structural diagram of a determining device for a target department according to a fifth embodiment of the present invention, where the present embodiment is applicable to a case of determining a corresponding target department based on disorder information described by a user, and particularly applicable to a case of performing department determination on a disorder described by a user in a scenario of multiple departments. The device specifically comprises: the system comprises a target text acquisition module 610, a matching degree value generation module 620, a department to be selected determination module 630 and a target department determination module 640.

A target text acquisition module 610 for acquiring target text for describing a user's condition;

the matching degree value generating module 620 is configured to input the target text into the department classification model, and determine a matching degree value between each department and the target text; the method comprises the steps of,

the department to be selected determining module 630 is configured to input the target file into the target symptom extraction model, obtain at least one symptom vocabulary to be processed corresponding to the target text, and determine at least one department to be selected corresponding to the target text according to the at least one symptom vocabulary to be processed and the target knowledge graph; the target knowledge graph comprises the relation between the symptom type and the department and the relation between the symptom type and the symptom vocabulary;

The target department determining module 640 is configured to determine a target department according to the matching degree value corresponding to each department and at least one department to be selected.

According to the technical scheme, a target text for describing the symptoms of the user is acquired through a target text acquisition module, the target text is input into a department classification model through a matching degree value generation module, and the matching degree value of each department and the target text is determined; inputting the target text into a target symptom extraction model through a department determination module to be selected to obtain at least one symptom vocabulary to be processed corresponding to the target text; determining at least one department to be selected corresponding to the target text according to the at least one symptom vocabulary to be processed and the target knowledge graph; the target department corresponding to the target text is determined by the target department determining module according to the matching degree value corresponding to each department and at least one department to be selected, so that the target department corresponding to the user is determined, and the target department of the user is determined efficiently and accurately.

Optionally, the target text obtaining module 610 is specifically configured to obtain text content edited by a user in a content editing control of an application program, and generate a target text based on the text content; or, acquiring voice information fed back by the user to the question-answering system, and generating a target text according to the voice information.

Optionally, the department classification model includes: the word segmentation sub-model, the word vector conversion sub-model and the classification matching degree value output sub-model are respectively used for dividing the target text into at least one word to be processed based on the word segmentation sub-model, and eliminating preset stop words in the word to be processed to obtain a vocabulary to be used; converting the vocabulary to be used through a word vector conversion sub-model, and determining word vectors of the vocabularies to be used; and inputting the word vector into the classification matching degree value output submodel to obtain the matching degree value between the target text and each department.

Optionally, the department determination module 630 to be selected is specifically configured to determine, for each symptom vocabulary to be processed, a similarity value between the symptom vocabulary to be processed and each symptom vocabulary in the target knowledge graph, and determine the target symptom vocabulary based on the similarity value; and aiming at each target symptom vocabulary, searching the symptom type associated with the current target symptom vocabulary and at least one department to be selected associated with the symptom type in the target knowledge graph.

Optionally, the target department determining module 640 includes a first determining unit, configured to determine a target department to be determined corresponding to a matching degree value higher than a preset matching degree threshold; and when the fact that the target department to be determined is included in the at least one department to be selected is detected, the target department to be determined is taken as the target department.

Optionally, the target department determining module 640 includes a second determining unit, configured to, when detecting that the number of at least one department to be selected is one, and when the matching degree value corresponding to the department to be selected is smaller than the preset matching degree threshold, take the department to be selected as the target department.

Optionally, the determining device of the target departments further includes a target text updating module, configured to display at least one corpus to be selected associated with the target text when the matching degree values corresponding to the departments are all smaller than the preset matching degree threshold, and the number of the at least one department to be selected includes a plurality of or is empty; updating the target text according to the corpus to be selected and the target text triggered by the user; and inputting the updated target text into a department classification model and a target symptom model to obtain a matching degree value corresponding to the updated target text and at least one department to be selected, and repeatedly displaying at least one corpus to be selected associated with the updated target text and updating the target text according to the corpus to be selected and the updated target text triggered by a user if the matching degree value corresponding to each department is smaller than a preset matching degree threshold and the number of the at least one department to be selected comprises a plurality of or is empty.

Optionally, the determining device of the target departments further includes a circulation threshold department determining module, configured to, when detecting that the actual circulation times of the target text reach the preset circulation times threshold value according to the corpus to be selected and the target text triggered by the user repeatedly, and the number of at least one department to be selected is empty and departments with a matching degree value higher than the preset matching degree threshold value exist, determine the target departments according to the departments corresponding to the departments higher than the preset matching degree threshold value; or,

when detecting that the actual circulation times of the updated target text reach a preset circulation times threshold value according to the corpus to be selected and the target text triggered by the user repeatedly and the number of at least one department to be selected comprises a plurality of departments, taking the at least one department to be selected as the target department.

The target department determining device provided by the embodiment of the invention can execute the target department determining method provided by any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the executing method.

It should be noted that, the units and modules included in the above system are only divided according to the functional logic, but not limited to the above division, so long as the corresponding functions can be implemented; in addition, the specific names of the functional units are also only for distinguishing from each other, and are not used to limit the protection scope of the embodiments of the present invention.

Example six

Fig. 7 is a schematic structural diagram of an electronic device according to a sixth embodiment of the present invention. Fig. 7 illustrates a block diagram of an exemplary electronic device 12 suitable for use in implementing embodiments of the present invention. The electronic device 12 shown in fig. 7 is merely an example and should not be construed as limiting the functionality and scope of use of embodiments of the present invention. Device 12 is typically an electronic device that assumes the functionality of a determined target department.

As shown in fig. 7, the electronic device 12 is in the form of a general purpose computing device. Components of the electronic device 12 may include, but are not limited to: one or more processors or processing units 16, a memory 28, and a bus 18 connecting the different components, including the memory 28 and the processing unit 16.

Bus 18 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor, and a local bus using any of a variety of bus architectures. By way of example, and not limitation, such architectures include industry standard architecture (Industry Standard Architecture, ISA) bus, micro channel architecture (Micro Channel Architecture, MCA) bus, enhanced ISA bus, video electronics standards association (Video Electronics Standards Association, VESA) local bus, and peripheral component interconnect (Peripheral Component Interconnect, PCI) bus.

Electronic device 12 typically includes a variety of computer-readable media. Such media can be any available media that is accessible by electronic device 12 and includes both volatile and nonvolatile media, removable and non-removable media.

Memory 28 may include computer device readable media in the form of volatile memory, such as random access memory (Random Access Memory, RAM) 30 and/or cache memory 32. The electronic device 12 may further include other removable/non-removable, volatile/nonvolatile computer storage media. By way of example only, storage device 34 may be used to read from or write to a non-removable, non-volatile magnetic media (not shown in FIG. 7, commonly referred to as a "hard disk drive"). Although not shown in fig. 7, a disk drive for reading from and writing to a removable nonvolatile magnetic disk (e.g., a "floppy disk"), and an optical disk drive for reading from and writing to a removable nonvolatile optical disk (e.g., a Compact Disc-Read Only Memory (CD-ROM), digital versatile Disc (Digital Video Disc-Read Only Memory, DVD-ROM), or other optical media) may be provided. In such cases, each drive may be coupled to bus 18 through one or more data medium interfaces. Memory 28 may include at least one program product 40, with program product 40 having a set of program modules 42 configured to perform the functions of embodiments of the present invention. Program product 40 may be stored, for example, in memory 28, such program modules 42 include, but are not limited to, one or more application programs, other program modules, and program data, each or some combination of which may include an implementation of a network environment. Program modules 42 generally perform the functions and/or methods of the embodiments described herein.

The electronic device 12 may also communicate with one or more external devices 14 (e.g., keyboard, mouse, camera, etc., and display), with one or more devices that enable a user to interact with the electronic device 12, and/or with any device (e.g., network card, modem, etc.) that enables the electronic device 12 to communicate with one or more other computing devices. Such communication may occur through an input/output (I/O) interface 22. Also, the electronic device 12 may communicate with one or more networks (e.g., local area network (Local Area Network, LAN), wide area network Wide Area Network, WAN) and/or a public network, such as the internet) via the network adapter 20. As shown, the network adapter 20 communicates with other modules of the electronic device 12 over the bus 18. It should be appreciated that although not shown, other hardware and/or software modules may be used in connection with electronic device 12, including, but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, disk array (Redundant Arrays of Independent Disks, RAID) devices, tape drives, data backup storage devices, and the like.

The processor 16 executes various functional applications and data processing by running a program stored in the memory 28, for example, to implement the target department determination method provided by the above-described embodiment of the present invention, including:

Acquiring target text for describing a user condition;

inputting the target text into a target symptom extraction model to obtain at least one symptom vocabulary to be processed corresponding to the target text, and determining at least one department to be selected corresponding to the target text according to the at least one symptom vocabulary to be processed and the target knowledge graph; the target knowledge graph comprises the relation between the symptom type and the department and the relation between the symptom type and the symptom vocabulary;

and determining a target department according to the matching degree value corresponding to each department and at least one department to be selected.

Of course, it will be understood by those skilled in the art that the processor may also implement the technical scheme of the method for determining a target department provided by any embodiment of the present invention.

Example seven

The seventh embodiment of the present invention also provides a storage medium containing computer-executable instructions for performing the steps of the method for determining a target subject room as provided by any embodiment of the present invention when executed by a computer processor.

Acquiring target text for describing a user condition;

The computer storage media of embodiments of the invention may take the form of any combination of one or more computer-readable media. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. The computer readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the computer-readable storage medium would include the following: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.

The computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, either in baseband or as part of a carrier wave. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination of the foregoing. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.

Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

Computer program code for carrying out operations for embodiments of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, smalltalk, C ++ and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computer (for example, through the Internet using an Internet service provider).

Note that the above is only a preferred embodiment of the present invention and the technical principle applied. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, while the invention has been described in connection with the above embodiments, the invention is not limited to the embodiments, but may be embodied in many other equivalent forms without departing from the spirit or scope of the invention, which is set forth in the following claims.

Claims

1. A method for determining a target department, comprising:

acquiring target text for describing a user condition;

Determining a target department according to the matching degree value corresponding to each department and the at least one department to be selected;

determining the target department according to the matching degree value corresponding to each department and the at least one department to be selected comprises the following steps:

determining a target department to be determined corresponding to the matching degree value higher than a preset matching degree threshold value;

when the fact that the target department to be determined is included in the at least one department to be selected is detected, the target department to be determined is used as the target department;

when the number of the at least one department to be selected is detected to be one, and the matching degree value corresponding to the department to be selected is smaller than a preset matching degree threshold value, the department to be selected is taken as a target department;

if the matching degree value corresponding to each department is smaller than the preset matching degree threshold value and the number of the at least one department to be selected comprises a plurality of or is empty, displaying at least one corpus to be selected which is associated with the target text;

updating the target text according to the corpus to be selected and the target text triggered by the user;

And inputting the updated target text into the department classification model and the target symptom extraction model to obtain a matching degree value corresponding to the updated target text and at least one department to be selected, and repeatedly displaying at least one corpus to be selected associated with the updated target text and re-updating the target text according to the corpus to be selected and the updated target text triggered by the user if the matching degree value corresponding to each department is smaller than a preset matching degree threshold and the number of the at least one department to be selected comprises a plurality of or is empty.

2. The method of claim 1, wherein the obtaining target text describing the user condition comprises:

acquiring text content edited by a user in a content editing control of an application program, and generating the target text based on the text content; or alternatively, the first and second heat exchangers may be,

and acquiring voice information fed back by the user to the question-answering system, and generating the target text according to the voice information.

3. The method of claim 1, wherein the department classification model comprises: the word segmentation sub-model, the word vector conversion sub-model and the classification matching degree value output sub-model are used for inputting the target text into a department classification model, and determining the matching degree value of each department and the target text, and the method comprises the following steps:

Dividing the target text into at least one word to be processed based on the word segmentation sub-model, and eliminating preset stop words in the word to be processed to obtain a word to be used;

converting the vocabulary to be used through the vocabulary vector conversion sub-model, and determining the vocabulary vectors of the vocabulary to be used;

and inputting the word vector into the classification matching degree value output submodel to obtain the matching degree value between the target text and each department.

4. The method according to claim 1, wherein the determining at least one department to be selected corresponding to the target text according to the at least one symptom vocabulary to be processed and the target knowledge-graph includes:

determining similarity values between the current symptom vocabulary to be processed and symptom vocabulary in the target knowledge graph, and determining the target symptom vocabulary based on the similarity values;

and aiming at each target symptom vocabulary, searching the symptom type associated with the current target symptom vocabulary and at least one department to be selected associated with the symptom type in the target knowledge graph.

5. The method as recited in claim 1, further comprising:

When detecting that the actual circulation times of the updated target text reach a preset circulation times threshold value according to the corpus to be selected and the target text triggered by the user repeatedly, and the number of at least one department to be selected is empty and departments with matching degree values higher than the preset matching degree threshold value exist, determining the target departments according to the departments corresponding to the matching degree values higher than the preset matching degree threshold value; or,

6. A target department determining apparatus, comprising:

the department to be selected determining module is used for inputting the target text into a target symptom extraction model to obtain at least one symptom vocabulary to be processed corresponding to the target text, and determining at least one department to be selected corresponding to the target text according to the at least one symptom vocabulary to be processed and a target knowledge graph; the target knowledge graph comprises a disease type, a department and a relation between the disease type and symptom vocabulary;

The target department determining module is used for determining a target department according to the matching degree value corresponding to each department and the at least one department to be selected;

the target department determining module comprises a first determining unit, a second determining unit and a third determining unit, wherein the first determining unit is used for determining a target department to be determined corresponding to the matching degree value higher than a preset matching degree threshold value;

when at least one department to be selected is detected to comprise a target department to be determined, taking the target department to be determined as a target department;

the target department determining module comprises a second determining unit, and is used for taking the departments to be selected as target departments when the number of at least one department to be selected is detected to be one and the matching degree value corresponding to the departments to be selected is smaller than a preset matching degree threshold value;

the target text updating module is used for displaying at least one corpus to be selected associated with the target text when the matching degree value corresponding to each department is smaller than a preset matching degree threshold value and the number of at least one department to be selected comprises a plurality of or is empty; updating the target text according to the corpus to be selected and the target text triggered by the user; and inputting the updated target text into a department classification model and a target symptom model to obtain a matching degree value corresponding to the updated target text and at least one department to be selected, and repeatedly displaying at least one corpus to be selected associated with the updated target text and updating the target text according to the corpus to be selected and the updated target text triggered by a user if the matching degree value corresponding to each department is smaller than a preset matching degree threshold and the number of the at least one department to be selected comprises a plurality of or is empty.

7. An electronic device, the electronic device comprising:

one or more processors;

storage means for storing one or more programs,

when executed by the one or more processors, causes the one or more processors to implement the method of determining a goal department as claimed in any one of claims 1-5.

8. A storage medium containing computer executable instructions which, when executed by a computer processor, are for performing the method of determining a target department as claimed in any one of claims 1 to 5.