Symptom Word Classification Method, Device, and Terminal

Abstract

An embodiment of the present invention provides a method, apparatus, and terminal for symptom word classification. The method includes: inputting a medical text into a symptom label classification model to obtain a first classification label of the medical text; matching the medical text with multiple symptom attribute word labels respectively to obtain a second classification label of the medical text; classifying the medical text according to the first classification label and the second classification label to obtain a symptom word classification result. By annotating the medical text with the first classification label and the second label, it provides a basis for subsequent relationship reasoning between entity words.

Description

Technical Field

[0001] The present invention relates to the technical field of medical knowledge graphs, and in particular to a method and device for classifying symptom words and a terminal. Background Art

[0002] In the medical field, each medical source has its own annotation. For example, the classification of diseases and surgical operations has the International Classification of Diseases (ICD) code, which is a system that classifies diseases according to certain characteristics of the diseases and represents them by coding methods. Due to the existence of various complex logical relationships between symptoms in clinical practice, one symptom name can derive many logically related symptom names, and the same symptom can also have multiple manifestations, and there are also certain differences in the description of symptoms, forming structured symptom words. However, there is no unified standard for symptom words, which increases the difficulty of actual medical use. Summary of the Invention

[0003] Embodiments of the present invention provide a method and device for classifying symptom words and a terminal to solve one or more technical problems in the prior art.

[0004] In a first aspect, embodiments of the present invention provide a method for classifying symptom words, including:

[0005] Inputting a medical text into a symptom label classification model to obtain a first classification label of the medical text;

[0006] Matching the medical text with multiple symptom attribute word labels respectively to obtain a second classification label of the medical text;

[0007] Classifying the medical text according to the first classification label and the second classification label to obtain a symptom word classification result.

[0008] In an implementation manner, it further includes:

[0009] Constructing a symptom label set, the set includes multiple symptom attribute word labels, and the set further includes one or more of multiple symptom name labels, multiple symptom sign word labels, and multiple symptom status word labels;

[0010] Training a symptom label classification model according to one or more of multiple symptom name labels, multiple symptom sign word labels, and multiple symptom status word labels.

[0011] In an implementation manner, constructing a symptom label set includes:

[0012] Constructing a basic symptom vocabulary, the vocabulary including a plurality of symptom attribute keywords, the vocabulary also including one or more words selected from a plurality of symptom proper nouns, a plurality of symptom sign keywords, and a plurality of symptom state keywords;

[0013] When the accuracy of the symptom basic vocabulary is greater than a first accuracy threshold, combining the plurality of symptom attribute keywords with one or more words from the plurality of symptom proper nouns, the plurality of symptom sign keywords, and the plurality of symptom state key keywords, respectively, to obtain one or more labels from the plurality of symptom name labels, the plurality of symptom sign word labels, and the plurality of symptom state word labels;

[0014] The symptom tag set is composed of a plurality of symptom attribute word tags, a plurality of symptom name tags, a plurality of symptom sign word tags, and a plurality of symptom state word tags.

[0015] In one embodiment, the plurality of symptom attribute keywords include one or more of: symptom location, symptom frequency, symptom attribute description, symptom color, symptom direction, symptom time point, and symptom degree modification.

[0016] In one embodiment, training a symptom label classification model based on one or more of a plurality of symptom name labels, a plurality of symptom sign word labels, and a plurality of symptom state word labels includes:

[0017] Determine whether the accuracy of the symptom label set is greater than a second accuracy threshold; if so, use the symptom label set to train a label classification model.

[0018] In a second aspect, an embodiment of the present invention provides a symptom word classification device, comprising:

[0019] A first classification label acquisition module is used to input the medical text into the symptom label classification model to obtain the first classification label of the medical text;

[0020] A second classification label acquisition module is used to match the medical text with a plurality of symptom attribute word labels respectively to obtain a second classification label of the medical text;

[0021] The symptom word classification module is used to classify the medical text according to the first classification label and the second classification label to obtain a symptom word classification result.

[0022] In one embodiment, it further includes:

[0023] A symptom label set construction module is used to construct a symptom label set, wherein the set includes a plurality of symptom attribute word labels, and the set also includes one or more labels selected from a plurality of symptom name labels, a plurality of symptom sign word labels, and a plurality of symptom state word labels;

[0024] A symptom label classification model training module, configured to train a symptom label classification model according to one or more of multiple symptom name labels, multiple symptom sign word labels, and multiple symptom status word labels.

[0025] In one implementation, the symptom label set construction module includes:

[0026] A symptom basic vocabulary construction unit, configured to construct a symptom basic vocabulary, the vocabulary including multiple symptom attribute keywords, and the vocabulary further including one or more of multiple symptom proper nouns, multiple symptom sign keywords, and multiple symptom status keywords;

[0027] A symptom basic vocabulary expansion unit, configured to combine multiple symptom attribute keywords with one or more of multiple symptom proper nouns, multiple symptom sign keywords, and multiple symptom status keywords respectively when the accuracy rate of the symptom basic vocabulary is greater than a first accuracy rate threshold, to obtain one or more of multiple symptom name labels, multiple symptom sign word labels, and multiple symptom status word labels;

[0028] A symptom label set acquisition unit, configured to form the symptom label set with multiple symptom attribute word labels and one or more of multiple symptom name labels, multiple symptom sign word labels, and multiple symptom status word labels.

[0029] In one implementation, the symptom label classification model training module includes:

[0030] A symptom label accuracy rate judgment unit, configured to judge whether the accuracy rate of the symptom label set is greater than a second accuracy rate threshold, and if so, use the symptom label set to train a label classification model.

[0031] In a third aspect, an embodiment of the present invention provides a symptom word classification terminal, and the functions of the symptom word classification terminal can be implemented by hardware or by hardware executing corresponding software. The hardware or software includes one or more modules corresponding to the above functions.

[0032] In a possible design, the structure of the symptom word classification terminal includes a processor and a memory, the memory is configured to store a program for supporting the symptom word classification terminal to execute the above symptom word classification method, and the processor is configured to execute the program stored in the memory. The symptom word classification terminal may further include a communication interface for communicating with other devices or communication networks.

[0033] Fourthly, an embodiment of the present invention provides a computer-readable storage medium for storing computer software instructions used by a symptom word classification terminal, which includes a program for executing the above-mentioned symptom word classification method.

[0034] One of the above technical solutions has the following advantages or beneficial effects: By annotating the medical text with the first classification label and the second label, it provides a basis for the subsequent relationship reasoning between entity words.

[0035] The above summary is only for the purpose of the specification and is not intended to be limiting in any way. In addition to the above-described illustrative aspects, embodiments, and features, further aspects, embodiments, and features of the present invention will be readily apparent by reference to the drawings and the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In the drawings, unless otherwise specified, the same reference numerals throughout the several views denote the same or similar parts or elements. These drawings are not necessarily drawn to scale. It should be understood that these drawings only depict some embodiments disclosed in accordance with the present invention and should not be regarded as limiting the scope of the present invention.

[0037] Figure 1 A flowchart showing a symptom word classification method according to an embodiment of the present invention is shown.

[0038] Figure 2 A flowchart showing another symptom word classification method according to an embodiment of the present invention is shown.

[0039] Figure 3 A flowchart showing another symptom word classification method according to an embodiment of the present invention is shown.

[0040] Figure 4 A structural block diagram showing a symptom word classification device according to an embodiment of the present invention is shown.

[0041] Figure 5 A structural block diagram showing another symptom word classification device according to an embodiment of the present invention is shown.

[0042] Figure 6 A structural block diagram showing another symptom word classification device according to an embodiment of the present invention is shown.

[0043] Figure 7 A structural schematic diagram showing a symptom word classification terminal according to an embodiment of the present invention is shown. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0044] In the following text, only some exemplary embodiments are simply described. As those skilled in the art can recognize, the described embodiments can be modified in various different ways without departing from the spirit or scope of the present invention. Therefore, the drawings and the description are considered to be exemplary in nature rather than restrictive.

[0045] Embodiment 1

[0046] In a specific implementation, as Figure 1 shown, a method for classifying symptom words is provided, including:

[0047] Step S10: Input the medical text into the symptom label classification model to obtain the first classification label that the medical text has.

[0048] Step S20: Match the medical text with multiple symptom attribute word labels respectively to obtain the second classification label that the medical text has.

[0049] Step S30: Classify the medical text according to the first classification label and the second classification label to obtain the symptom word classification result.

[0050] In an example, the medical text can come from various websites on the Internet, electronic medical records, medical books, etc. The medical text can include multiple structured symptom words or other structured medical words. The symptom label classification model can extract multiple structured medical words in the medical text and output one or more classification rules for splitting the structured symptom words. The classification rules can include splitting the structured symptom words according to non - split medical proper nouns. Or, splitting the structured symptom words according to symptom sign words. Or, splitting the structured symptom words according to symptom state words. One or more classification rules constitute the first classification label. The structured symptom words in the medical text can be labeled with the first classification label to obtain the medical text with the first classification label. The multiple symptom attribute word labels can include symptom location labels, symptom frequency labels, symptom attribute description labels, symptom color labels, symptom orientation labels, symptom time - point labels, symptom degree modification labels, etc. The structured symptom words in the medical text are labeled with the aforementioned one or more symptom attribute word labels to obtain the medical text with the second classification label. The structured symptom words in the medical text that have both the first classification label and the second classification label can be split into a set where each structured symptom word has one or more labels, which is the symptom word classification result.

[0051] It should be noted that steps S10 and S20 are not limited in sequence. After annotating symptom attribute word tags such as part, color, and direction for the medical text, it can be input into the symptom label classification model for annotating the first classification label. For example, "abscess in the right lower lobe of the liver" has part tags "liver" and "lower lobe", orientation tag "right lower", and the remaining word "abscess" is classified using the symptom label classification model, and the "abscess" is annotated as a symptom status word tag. "Red and swollen skin on both lower extremities" has part tags "both lower extremities" and "skin", color tag "red", and the remaining word "swollen" is classified using the symptom label classification model, and the "swollen" is annotated as a symptom status word tag.

[0052] The normalization and disambiguation of diseases can be performed using the classification results of symptom words. For example, when normalizing "headache" and "abdominal pain", it can be detected that the label of "headache" is the head, and the label of "abdominal pain" is the abdomen. Since the parts conflict, normalization cannot be performed. When normalizing "lumbar disc herniation", "lumbar disc annulus fibrosus rupture", and "lumbar disc prolapse", it can be detected that the label of "lumbar disc herniation" is lumbar disc and herniation. The label of "lumbar disc annulus fibrosus rupture" is lumbar disc, annulus fibrosus, and rupture. The label of "lumbar disc prolapse" is lumbar disc and prolapse. Since they all have the label lumbar disc, the three can be normalized to "lumbar disc herniation". The classification results of symptom words can also assist doctors in prescribing medicine reminders. For example, a doctor prescribes a medicine for treating the head, but during the patient symptom detection, an abdominal-related label is detected, then it will remind the doctor whether the medicine is prescribed wrongly.

[0053] In one implementation, as Figure 2 shown, it further includes:

[0054] Step S01: Construct a symptom label set, the set includes multiple symptom attribute word tags, and the set also includes one or more of multiple symptom name tags, multiple symptom sign word tags, and multiple symptom status word tags;

[0055] Step S02: Train a symptom label classification model according to one or more of multiple symptom name tags, multiple symptom sign word tags, and multiple symptom status word tags.

[0056] In one example, multiple symptom attribute word tags can be derived from multiple symptom attribute keywords. The symptom attribute word tags include one or more of: symptom location tags, symptom frequency tags, symptom attribute description tags, symptom color tags, symptom orientation tags, symptom time point tags, and symptom severity modifier tags. Specifically, the symptom location tag represents the organ location where the patient has an abnormality, such as tags like "abdomen", "stomach", "head", "back", etc. The symptom frequency tag represents the frequency status of the patient's abnormal sensations or signs. For example, tags like "occasional", "intermittent", "sudden", etc. The symptom attribute description tag represents additional descriptions of symptoms and signs. For example, "symmetric", "clear", "metastasis", "diffuse", etc. The symptom color tag represents the color status of the substance of the patient's abnormal sensations or signs, such as tags like "yellow" for hair and "purple" for lip color. The symptom orientation tag represents the orientation of the location of the patient's abnormal sensations or signs, usually used in combination with the location, such as "right" in right upper abdominal pain, and "upper", "lower", "right", "left", "lower left", "upper right", "upper left", etc. The symptom time point tag represents the starting time point of the patient's abnormal sensations or signs, for example, "night" in night cough, "morning", etc. The symptom severity modifier tag represents the severity of the status of the patient's abnormal sensations or signs, such as "severe" or "mild" in severe cough. The symptom name tag represents the proper name of the symptom, which can independently represent the specific meaning of a certain symptom and cannot be divided into finer granularity, for example, tags like "liver palms", "wheezing", etc. The symptom sign word tag represents the tag that can characterize the physical characteristics of the patient's abnormal sensations, such as tags like "breathing", "body temperature", "pulse", "blood pressure", "nasal mucus", "urine", etc. The symptom status word tag represents the word tag of the status of the patient's physical characteristics, such as "necrosis", "swelling", "peculiar smell", "vomiting", "fever", "murmur", "rebound tenderness", "rigidity", "obstruction", "pain", "dullness", "rale", etc.

[0057] Since the symptom name tag, symptom sign word tag, and symptom status word tag cannot be directly marked in medical texts, a large number of samples are needed to train the symptom label classification model. Input the new medical text into the symptom label classification model, and the output medical text is marked with a first classification label composed of one or more of the symptom name tag, symptom sign word tag, and symptom status word tag. Through matching, the medical text can be directly marked with the symptom attribute word tag, so a medical text with a second classification label can be obtained without training the model.

[0058] It should be noted that, in order to improve the speed and efficiency of labeling multiple symptom attribute word tags for medical texts, it is also possible to implement the labeling of multiple symptom attribute word tags by training another symptom label classification model. Taking the location tag as an example: after "eye pain" and "weakened pupillary light reflex" pass through another symptom label classification model, the "eye" location tag is obtained. After "low back and back pain" passes through another symptom label classification model, the "lower back" location tag, "back" location tag, and "lower back and back" location tag are obtained.

[0059] In one implementation, as Figure 3 shown, step S01 includes:

[0060] Step S011: Construct a symptom base vocabulary, the vocabulary includes multiple symptom attribute keywords, and the vocabulary also includes one or more of multiple symptom proper nouns, multiple symptom sign keywords, and multiple symptom status keywords;

[0061] Step S012: When the accuracy rate of the symptom base vocabulary is greater than the first accuracy rate threshold, combine multiple symptom attribute keywords with one or more of multiple symptom proper nouns, multiple symptom sign keywords, and multiple symptom status keywords respectively to obtain one or more of multiple symptom name tags, multiple symptom sign word tags, and multiple symptom status word tags;

[0062] Step S013: Multiple symptom attribute word tags and one or more of multiple symptom name tags, multiple symptom sign word tags, and multiple symptom status word tags form a symptom tag set.

[0063] In one example, the words in the symptom base vocabulary can come from the Internet, books, etc., and can also come from manually predefined words. Multiple symptom attribute keywords, multiple symptom proper nouns, multiple symptom sign keywords, and multiple symptom status keywords can be constructed into a small sample tag set. The small sample tag set can be expanded. Using multiple symptom proper nouns, multiple symptom sign keywords, and multiple symptom status keywords in the symptom base vocabulary as roots, combine them with multiple symptom attribute keywords respectively to expand the symptom base vocabulary. For example, "abdominal pain" is the root of "upper abdominal pain", then "upper" may be the word to be expanded. After the scope of the symptom base vocabulary is expanded, a new and larger tag set is formed.

[0064] In one implementation, multiple symptom attribute keywords include one or more of: symptom location, symptom frequency, symptom attribute description, symptom color, symptom orientation, symptom time point, and symptom degree modification.

[0065] In one implementation, as Figure 3 shown, step S02 includes:

[0066] Step S021: Determine whether the accuracy rate of the symptom label set is greater than the second accuracy threshold. If so, train the label classification model using the symptom label set.

[0067] In one example, the symptom base vocabulary and the vocabulary expanded according to the above method can be sampled manually to determine the accuracy rate, and finally it is determined that the symptom base vocabulary and the vocabulary expanded according to the above method are effective. Use the symptom labels composed of the expanded vocabulary as samples to train the label classification model.

[0068] Embodiment 2

[0069] In another specific implementation, as Figure 4 shown, a symptom word classification device is provided, including:

[0070] The first classification label acquisition module 10 is used to input the medical text into the symptom label classification model to obtain the first classification label possessed by the medical text;

[0071] The second classification label acquisition module 20 is used to match the medical text with multiple symptom attribute word labels respectively to obtain the second classification label possessed by the medical text;

[0072] The symptom word classification module 30 is used to classify the medical text according to the first classification label and the second classification label to obtain the symptom word classification result.

[0073] In one implementation, as Figure 5 shown, it further includes:

[0074] The symptom label set construction module 11 is used to construct a symptom label set, the set includes multiple symptom attribute word labels, and the set also includes one or more of multiple symptom name labels, multiple symptom sign word labels, and multiple symptom status word labels;

[0075] The symptom label classification model training module 12 is used to train the symptom label classification model according to one or more of multiple symptom name labels, multiple symptom sign word labels, and multiple symptom status word labels.

[0076] In one implementation, as Figure 6 shown, the symptom label set construction module 11 includes:

[0077] The symptom base vocabulary construction unit 110 is used to construct a symptom base vocabulary, the vocabulary includes multiple symptom attribute keywords, and the vocabulary also includes one or more of multiple symptom proper nouns, multiple symptom sign keywords, and multiple symptom status keywords;

[0078] The symptom basic vocabulary expansion unit 111 is used to combine multiple symptom attribute keywords with one or more of multiple symptom proper nouns, multiple symptom sign keywords, and multiple symptom state keyword combinations respectively when the accuracy rate of the symptom basic vocabulary is greater than the first accuracy threshold, so as to obtain one or more of multiple symptom name labels, multiple symptom sign word labels, and multiple symptom state word labels;

[0079] The symptom label set acquisition unit 112 is used to form the symptom label set with multiple symptom attribute word labels and one or more of multiple symptom name labels, multiple symptom sign word labels, and multiple symptom state word labels.

[0080] In one implementation, as Figure 6 shown, the symptom label classification model training module 12 includes:

[0081] The symptom label accuracy rate judgment unit 120 is used to judge whether the accuracy rate of the symptom label set is greater than the second accuracy threshold. If so, the label classification model is trained using the symptom label set.

[0082] For the functions of the modules in each device of the embodiments of the present invention, reference may be made to the corresponding descriptions in the above methods, which will not be elaborated here.

[0083] Embodiment III

[0084] Figure 7 Show the structural block diagram of the symptom word classification terminal according to the embodiment of the present invention. As Figure 7 shown, the terminal includes: a memory 910 and a processor 920, and a computer program that can run on the processor 920 is stored in the memory 910. When the processor 920 executes the computer program, the symptom word classification method in the above embodiment is implemented. The number of the memory 910 and the processor 920 can be one or more.

[0085] The terminal further includes:

[0086] A communication interface 930 is used to communicate with external devices for data interaction and transmission.

[0087] The memory 910 may include a high-speed RAM memory and may also include a non-volatile memory, such as at least one disk memory.

[0088] If the memory 910, the processor 920, and the communication interface 930 are implemented independently, the memory 910, the processor 920, and the communication interface 930 can be interconnected through a bus and communicate with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The bus can be divided into an address bus, a data bus, a control bus, etc. For the sake of convenience of representation, Figure 7 it is only represented by a thick line in the figure, but it does not mean that there is only one bus or one type of bus.

[0089] Optionally, in a specific implementation, if the memory 910, the processor 920, and the communication interface 930 are integrated on a single chip, the memory 910, the processor 920, and the communication interface 930 can communicate with each other through an internal interface.

[0090] An embodiment of the present invention provides a computer-readable storage medium storing a computer program, which when executed by a processor implements any one of the methods in the above embodiments.

[0091] In the description of this specification, the descriptions referring to terms such as "one embodiment", "some embodiments", "example", "specific example", or "some examples" mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples. In addition, without contradiction, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0092] In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the quantity of the indicated technical features. Thus, the features defined with "first" and "second" can explicitly or implicitly include at least one of the features. In the description of the present invention, "a plurality of" means two or more unless otherwise specifically defined.

[0093] Any process or method description represented in a flowchart or otherwise described herein can be understood to represent a module, segment, or portion of code including one or more executable instructions for implementing a specific logical function or process. The scope of the preferred embodiments of the present invention includes additional implementations in which functions may be executed in a substantially simultaneous manner or in an order opposite to that shown or discussed, according to the functions involved, which should be understood by those skilled in the art to which the embodiments of the present invention pertain.

[0094] The logic and / or steps represented in a flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other systems that can fetch and execute instructions from the instruction execution system, apparatus, or device), or in conjunction with such instruction execution systems, apparatus, or devices. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by or in conjunction with an instruction execution system, apparatus, or device. More specific examples (non-exhaustive list) of computer-readable media include the following: an electrical connection portion with one or more wirings (electronic device), a portable computer diskette (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable read-only memory (CDROM). Additionally, the computer-readable medium can even be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpretation, or other appropriate processing as necessary, and then stored in a computer memory.

[0095] It should be understood that various parts of the present invention can be implemented by hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, any one or a combination of the following techniques well known in the art can be used: discrete logic circuits having logic gate circuits for implementing logical functions on data signals, application-specific integrated circuits having appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0096] Those of ordinary skill in the art can understand that all or part of the steps carried out in the methods of the above embodiments can be completed by instructing relevant hardware through a program. The program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiments.

[0097] In addition, in each of the embodiments of the present invention, the functional units can be integrated into a processing module, or each unit can exist physically alone, or two or more units can be integrated into one module. The above integrated module can be implemented in the form of hardware or in the form of a software functional module. When the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. The storage medium can be a read-only memory, a magnetic disk, an optical disk, or the like.

[0098] The above is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of various changes or substitutions, and these should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. A symptom word classification method, characterized in that: include: Inputting the medical text into the symptom label classification model to obtain the first classification label of the medical text; The first classification tag is used to represent the splitting of the structured symptom words in the medical text using at least one classification rule; Matching the medical text with a plurality of symptom attribute word labels respectively to obtain a second classification label of the medical text; Using the first classification label and the second classification label to label the structured symptom words in the medical text respectively, to obtain a medical text with the first classification label and the second classification label; The medical text is classified according to the first classification label and the second classification label to obtain a symptom word classification result. The symptom word classification result includes: splitting the structured symptom words in the medical text that have both the first classification label and the second classification label, and decomposing them into a set formed by each structured symptom word having one or more labels as the symptom word classification result.

2. The method according to claim 1, characterized in that Also includes: Constructing a symptom label set, the set including a plurality of symptom attribute word labels, the set also including one or more labels selected from a plurality of symptom name labels, a plurality of symptom sign word labels, and a plurality of symptom state word labels; A symptom label classification model is trained based on one or more labels among multiple symptom name labels, multiple symptom sign word labels, and multiple symptom state word labels.

3. The method according to claim 2, characterized in that Build a symptom label set, including: Constructing a basic symptom vocabulary, the vocabulary including a plurality of symptom attribute keywords, the vocabulary also including one or more words selected from a plurality of symptom proper nouns, a plurality of symptom sign keywords, and a plurality of symptom state keywords; When the accuracy of the symptom basic vocabulary is greater than a first accuracy threshold, combining the plurality of symptom attribute keywords with one or more words from the plurality of symptom proper nouns, the plurality of symptom sign keywords, and the plurality of symptom state key keywords, respectively, to obtain one or more labels from the plurality of symptom name labels, the plurality of symptom sign word labels, and the plurality of symptom state word labels; The symptom tag set is composed of a plurality of symptom attribute word tags, a plurality of symptom name tags, a plurality of symptom sign word tags, and a plurality of symptom state word tags.

4. The method according to claim 3, characterized in that The multiple symptom attribute keywords include: one or more of: symptom location, symptom frequency, symptom attribute description, symptom color, symptom direction, symptom time point, and symptom degree modification.

5. The method according to claim 3, characterized in that Training a symptom label classification model based on one or more of a plurality of symptom name labels, a plurality of symptom sign word labels, and a plurality of symptom state word labels includes: Determine whether the accuracy of the symptom label set is greater than a second accuracy threshold; if so, use the symptom label set to train a label classification model.

6. A symptom word classification device, characterized in that: include: A first classification label acquisition module is used to input the medical text into the symptom label classification model to obtain the first classification label of the medical text; The first classification tag is used to represent the splitting of the structured symptom words in the medical text using at least one classification rule; A second classification label acquisition module is used to match the medical text with a plurality of symptom attribute word labels respectively to obtain a second classification label of the medical text; a symptom word classification module, configured to mark the structured symptom words in the medical text using the first classification label and the second classification label, respectively, to obtain a medical text having the first classification label and the second classification label; The medical text is classified according to the first classification label and the second classification label to obtain a symptom word classification result. The symptom word classification result includes: splitting the structured symptom words in the medical text that have both the first classification label and the second classification label, and decomposing them into a set formed by each structured symptom word having one or more labels as the symptom word classification result.

7. The device according to claim 6, characterized in that Also includes: A symptom label set construction module is used to construct a symptom label set, wherein the set includes a plurality of symptom attribute word labels, and the set also includes one or more labels selected from a plurality of symptom name labels, a plurality of symptom sign word labels, and a plurality of symptom state word labels; The symptom label classification model training module is used to train the symptom label classification model based on one or more labels among multiple symptom name labels, multiple symptom sign word labels, and multiple symptom state word labels.

8. The device according to claim 7, characterized in that The symptom label set building module includes: A symptom basic vocabulary building unit is used to build a symptom basic vocabulary, wherein the vocabulary includes a plurality of symptom attribute keywords, and the vocabulary also includes one or more words selected from a plurality of symptom proper nouns, a plurality of symptom sign keywords, and a plurality of symptom state keywords; a symptom basic vocabulary expansion unit, configured to combine, when the accuracy of the symptom basic vocabulary is greater than a first accuracy threshold, a plurality of symptom attribute keywords with one or more words selected from a plurality of symptom proper nouns, a plurality of symptom sign keywords, and a plurality of symptom state key keywords, respectively, to obtain one or more labels selected from a plurality of symptom name labels, a plurality of symptom sign word labels, and a plurality of symptom state word labels; The symptom label set acquisition unit is used for a plurality of symptom attribute word labels, and one or more labels selected from a plurality of symptom name labels, a plurality of symptom sign word labels, and a plurality of symptom state word labels to form the symptom label set.

9. The device according to claim 8, characterized in that The symptom label classification model training module includes: The symptom label accuracy judgment unit is used to judge whether the accuracy of the symptom label set is greater than a second accuracy threshold. If so, the label classification model is trained using the symptom label set.

10. A symptom word classification terminal, characterized in that: include: one or more processors; a storage device for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors are caused to implement the method according to any one of claims 1 to 5.

11. A computer-readable storage medium storing a computer program, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

Images