Medical literature classification system based on intelligent labeling

Abstract

The invention is suitable for the technical field of literature data processing, and provides a medical literature classification system based on intelligent annotation, which comprises a database and a classification module in communication connection with the database, the classification module comprises a retrieval unit, an extraction unit, a training unit, an integration unit, a labeling unit, a management unit and a classification unit which are in communication connection in sequence; the training unit is used for training the classification model by taking the annotation data as training data to obtain a medical literature classification model; according to the system, data extraction is carried out on retrieved literatures, data mining analysis is carried out on obtained key information, and training based on a professional classification system and labeling rules is carried out on the literatures, so that the problem that classification of existing medical literatures is not refined and professional enough can be solved, regular word screening is carried out on literature word segmentation, and the classification efficiency of the medical literatures is improved. And the problem of relatively large classification data in the prior art can be solved, and an efficient classification effect is obtained.

Description

Technical Field

[0001] The present invention relates to the technical field of document data processing, and more specifically, to a medical document classification system based on intelligent annotation. Background Art

[0002] Keywords are usually annotated during the compilation process of scientific and technological literature to avoid difficulties in understanding when non-professionals read them. The existing keyword-based scientific and technological literature annotation method is to input the source and analysis of keywords word by word, but the keywords in scientific and technological literature appear more frequently, and the same keywords in the same type of medical literature appear more frequently. Therefore, it takes a long time to annotate keywords from the same source, and this annotation method is less efficient.

[0003] The keyword-based annotation of medical literature requires a simple and fixed pattern to achieve the effect of being easy to understand and convenient for reference. Currently, the online literature service platforms provided on the market have few medical-related literature services and are not professional enough. For example, the Wanfang platform: for medical literature, Wanfang has made a coarse-grained division of medical-related literature from the aspects of medicine and health; Pubmed: provides biomedical paper searches and abstracts, and its core theme is medicine. PubMed's literature does not include the full text of journal articles, but can provide links to full-text providers (paid or free); the above two platforms do not provide a complete classification of medical literature, and only classify from one dimension, so when doctors look for a certain type of literature, they cannot meet the search needs well.

[0004] Therefore, it is very necessary to design a method for classifying medical literature by adding keywords for intelligent annotation. Summary of the invention

[0005] In view of the shortcomings of the prior art, the purpose of the present invention is to provide a medical document classification system based on intelligent annotation. By training the documents based on a professional classification system and annotation rules, the problem that the classification of existing medical documents is not detailed and professional enough can be solved. By screening the document segmentation words according to the rules, the problem that the classification data of the prior art is large can be solved, and an efficient classification effect can be obtained. By extracting data from the retrieved documents and performing data mining and analysis on the acquired key information, a keyword data group is obtained, and the relevant medical documents are retrieved according to the combination of keywords in the keyword data group, and the medical document keyword combination is used as the storage identifier of the traditional Chinese medicine document, thereby achieving efficient and accurate acquisition of medical document data and completing the precise classification and storage of medical documents.

[0006] To achieve the above object, the present invention provides the following technical solutions: A medical document classification system based on intelligent annotation comprises a database and a classification module connected to the database in communication; the classification module comprises a retrieval unit, an extraction unit, a training unit, an integration unit, an annotation unit, a management unit and a classification unit connected in sequence through communication; the training unit is used to train a classification model using annotation data as training data to obtain a medical document classification model; the integration unit is used to integrate the extracted keyword annotation information into a fixed query path and display format, generate an orderly arrangement of the keyword information through a connectable capture form, and directly display the original source text of the keyword in the database through the integrated display information; the annotation unit is used to select the keyword that needs to be annotated in the medical document, quote the annotation symbol of the keyword, and then add the integrated annotation information corresponding to the keyword into the annotation position of the medical document.

[0007] The present invention is further configured as follows: the retrieval unit searches the database using medical retrieval keywords to obtain a number of medical documents matching the medical retrieval keywords as a basic medical document database; the documents are integrated into a flat file in the order of downloading and saved in a specific coding format.

[0008] The present invention is further configured as follows: the extraction unit is used to extract key information from the downloaded unstructured text data, save it in a specific data format, and establish a coding table for key information of traditional Chinese medical literature, wherein the key information corresponds to a binary code respectively; data mining and analysis are performed on the acquired key information to obtain a keyword data group, and the keywords in the keyword data group are combined to retrieve relevant medical literature, and the medical literature keyword combination is used as the medical literature storage identifier.

[0009] The present invention is further configured as follows: the training unit extracts data to be annotated from a basic literature database; annotates the data to be annotated using a pre-established classification system and annotating rule library; and trains a classification model using the annotated data as training data to obtain a medical literature classification model.

[0010] The present invention is further configured as follows: the annotation unit matches one or more rule classes involved in the data to be annotated from a pre-established classification system and an annotation rule library; constructs a corresponding rule class set tree based on the one or more rule classes involved in the data to be annotated; matches the constructed rule class set tree with the annotation rule library, and annotates the document category associated with the matched annotation rule as the document category of the sample document.

[0011] The present invention is further configured as follows: the management unit manages the integrated keyword information and the annotated sample documents in a unified manner, directly extracts the archived keyword annotation information when applied, and the application of the same keyword in different positions in the medical documents can have a variety of different extraction sources and interpretations.

[0012] The present invention is further configured as follows: the classification unit extracts features from the annotated data, constructs a maximum entropy classification model as a medical document classification model based on the features, and constructs a maximum entropy classification model to classify the documents to be classified.

[0013] The present invention is further configured as follows: a denoising unit is communicatively connected between the integration unit and the annotation unit; the denoising unit is used to filter and denoise the annotated data before annotating the data, so as to remove the same document from multiple sources.

[0014] The advantages of the present invention are: 1. The present invention extracts data from the retrieved documents, performs data mining and analysis on the acquired key information to obtain a keyword data group, and retrieves relevant medical documents based on the combination of keywords in the keyword data group, and uses the medical document keyword combination as the storage identifier of the traditional Chinese medical document, thereby achieving efficient and accurate acquisition of medical document data and completing the precise classification and storage of medical documents.

[0015] 2. The present invention extracts a variety of different sources and interpretations of keywords, and the most accurate analysis and source are added to the annotation, thereby realizing diversified application and management of keywords, achieving different interpretation effects of the same keyword in different scenarios, and improving the annotation accuracy.

[0016] 3. The present invention can solve the problem that the classification of existing medical literature is not detailed and professional enough by training the literature based on professional classification system and annotation rules, screen the regular words for the literature segmentation, and solve the problem that the classification data of the existing technology is large, thereby obtaining efficient classification effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 The present invention is a flowchart of a medical literature classification system based on intelligent annotation. DETAILED DESCRIPTION

[0018] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application may be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0019] It should be noted that, unless otherwise specified, all technical and scientific terms used in this application have the same meanings as commonly understood by ordinary technicians in the technical field to which this application belongs.

[0020] In the present invention, unless otherwise specified, the directions used, such as "up" and "down", usually refer to the directions shown in the drawings, or to the vertical, perpendicular or gravity directions; similarly, for ease of understanding and description, "left" and "right" usually refer to the left and right shown in the drawings; "inside" and "outside" refer to the inside and outside relative to the outline of each component itself, but the above-mentioned directions are not used to limit the present invention.

[0021] For example, see Figure 1 , the present invention provides the following technical solutions: A medical document classification system based on intelligent annotation, specifically, comprises a database and a classification module connected to the database by communication; the classification module comprises a retrieval unit, an extraction unit, a training unit, an integration unit, a annotation unit, a management unit and a classification unit connected in sequence by communication; the training unit is used to train a classification model using the annotation data as training data to obtain a medical document classification model; the integration unit is used to integrate the extracted keyword annotation information into a fixed query path and display format, generate an orderly arrangement of the keyword information through a connectable capture form, and directly display the original source text of the keyword in the database through the integrated display information; the annotation unit is used to select the keyword to be annotated in the medical document, and quote the annotation symbol of the keyword, and then add the integrated annotation information corresponding to the keyword into the annotation position of the medical document.

[0022] Working principle of the first embodiment: by training the documents based on the professional classification system and annotation rules, it is possible to solve the problem that the classification of existing medical documents is not detailed and professional enough, and by screening the document segmentation words based on the rule words, it is also possible to solve the problem that the classification data of the existing technology is large, and obtain efficient classification effect; by and through data extraction of the retrieved documents, data mining and analysis of the acquired key information, a keyword data group is obtained, and the relevant medical documents are retrieved according to the combination of keywords in the keyword data group, and the medical document keyword combination is used as a storage identifier of traditional Chinese medicine documents, thereby realizing efficient and accurate acquisition of medical document data and completing the precise classification and storage of medical documents.

[0023] For example 2, please refer to Figure 1 The second embodiment makes the following improvements on the basis of the first embodiment. Specifically, the retrieval unit searches the database using medical retrieval keywords, and obtains a number of medical documents matching the medical retrieval keywords as a basic medical document database; the medical documents are integrated into a flat file in the order of downloading, and saved in a specific encoding format.

[0024] Key information includes two types: core data and non-core data. The extracted data is first stored in the corresponding database as the basic data for the next step of data processing, and then imported into SQL for the next step of mining and analysis; core data refers to keywords in the downloaded text data that exceed the set threshold frequency, and non-core data refers to keywords that appear at least once in the downloaded text data and are lower than the set threshold.

[0025] SQL (Structured Query Language) is a database language with multiple functions such as data manipulation and data definition. This language is interactive and can provide great convenience for users. Database management systems should make full use of SQL language to improve the work quality and efficiency of computer application systems. SQL language can not only be used independently in terminals, but also as a sub-language to provide effective assistance for other program designs. In this program application, SQL can optimize program functions together with other program languages, thereby providing users with more and more comprehensive information; SQL, as an operation command set, is widely welcomed by industry insiders for its rich functions and has become a guarantee for improving database operation efficiency. The application of SQL Server database can effectively improve the speed of data request and return, effectively deal with the processing of complex tasks, and is the key to improving work efficiency.

[0026] The extraction unit is used to extract key information from the downloaded unstructured text data, save it in a specific data format, and establish a coding table for key information of traditional Chinese medicine literature, where each key information corresponds to a binary code; data mining and analysis are performed on the acquired key information to obtain a keyword data group, and the keywords in the keyword data group are combined to retrieve relevant medical literature, and the medical literature keyword combination is used as a medical literature storage identifier.

[0027] The training unit is to extract the data to be annotated from the literature basic database; annotate the data to be annotated using the pre-established classification system and annotation rule library; train the classification model using the annotated data as training data to obtain a medical literature classification model.

[0028] The annotation unit matches one or more rule classes involved in the data to be annotated from a pre-established classification system and annotation rule library; constructs a corresponding rule class set tree based on the one or more rule classes involved in the data to be annotated; matches the constructed rule class set tree with the annotation rule library, and annotates the document category associated with the matched annotation rule as the document category of the sample document.

[0029] The management unit refers to the unified management of the integrated keyword information and the annotated sample documents, and the direct extraction of archived keyword annotation information during application. The same keyword can have multiple different extraction sources and interpretations when used in different locations in medical documents.

[0030] The classification unit extracts features from the annotated data, constructs a maximum entropy classification model based on the features as a medical literature classification model, and constructs a maximum entropy classification model to classify the literature to be classified.

[0031] The configuration is used to extract data to be classified from the documents to be classified, and to perform classification prediction on the data to be classified by a medical document classification model trained by a training device of a medical document classification model according to each embodiment provided in the present application, to obtain a corresponding prediction category and a corresponding prediction probability; and to compare the prediction probability with a set threshold to determine a classification result.

[0032] The medical literature classification model trained by the medical literature classification model training method performs classification prediction on the data to be classified, and obtains two prediction data, one is the prediction category, and the other is the prediction probability. For a document to be classified, the above two data are correlated, that is, the prediction probability is the probability that the document to be classified belongs to the prediction category.

[0033] A denoising unit is connected in communication between the integration unit and the annotation unit; the denoising unit is used to filter and denoise the annotated data before annotating the annotated data, so as to remove the same document from multiple sources.

[0034] Working principle of the second embodiment of the present invention: through the integration module and the management module, the keyword annotation information will be automatically archived and uniformly managed after being input for the first time. If the same keyword appears again in the scientific and technological literature, there is no need to extract the keyword information again. It can be directly retrieved from the uniformly managed annotation library and share the same annotation with the keyword analysis of the previous text to avoid repeated operations. This method improves the efficiency and effect of keyword annotation with scientific and technological literature; by; by and through data extraction of the retrieved literature, data mining and analysis of the acquired key information, obtaining a keyword data group, and searching for relevant medical literature based on the combination of keywords in the keyword data group, and using the medical literature keyword combination as a storage identifier for traditional Chinese medicine literature, thereby achieving efficient and accurate acquisition of medical literature data and completing the precise classification and storage of medical literature.

[0035] Obviously, the above-described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0036] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0037] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein.

[0038] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

[0039] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. A medical literature classification system based on intelligent annotation, characterized by: comprising a database and a classification module in communication with the database; The classification module includes a retrieval unit, an extraction unit, a training unit, an integration unit, a labeling unit, a management unit and a classification unit which are sequentially connected by communication; The training unit is used to train the classification model using the labeled data as training data to obtain a medical literature classification model; The integration unit is used to integrate the extracted keyword annotation information into a fixed query path and display format, generate an orderly arrangement of the keyword information through a connectable capture form, and directly display the original text of the keyword in the database through the integrated display information; The marking unit is used to select the keywords that need to be marked in the medical literature, and quote the marking symbols of the keywords, and then add the integrated marking information corresponding to the keywords into the marking position of the medical literature.

2. According to claim 1, a medical literature classification system based on intelligent annotation is characterized by: The retrieval unit searches the database using medical search keywords to obtain a number of medical documents matching the medical search keywords as a basic medical document database; the documents are integrated into a flat file in the order of downloading and saved in a specific coding format.

3. The medical literature classification system based on intelligent annotation according to claim 2, characterized in that: The extraction unit is used to extract key information from the downloaded unstructured text data, save it in a specific data format, and establish a coding table for key information of traditional Chinese medicine literature, wherein each key information corresponds to a binary code; Data mining and analysis are performed on the acquired key information to obtain a keyword data group, and the keywords in the keyword data group are combined to retrieve relevant medical literature, and the medical literature keyword combination is used as the medical literature storage identifier.

4. The medical literature classification system based on intelligent annotation according to claim 3, characterized in that: The training unit extracts data to be annotated from a basic literature database; annotates the data to be annotated using a pre-established classification system and annotating rule library; and trains a classification model using the annotated data as training data to obtain a medical literature classification model.

5. The medical literature classification system based on intelligent annotation according to claim 4, characterized in that: The labeling unit matches one or more rule classes involved in the data to be labeled from a pre-established classification system and labeling rule library; Based on one or more rule classes involved in the data to be annotated, a corresponding rule class set tree is constructed; the constructed rule class set tree is matched with the annotation rule library, and the document category associated with the matched annotation rule is annotated as the document category of the sample document.

6. The medical literature classification system based on intelligent annotation according to claim 5, characterized in that: The management unit manages the integrated keyword information and the annotated sample documents in a unified manner, directly extracts the archived keyword annotation information when applying, and the application of the same keyword in different positions in the medical documents can have multiple different extraction sources and interpretations.

7. The medical literature classification system based on intelligent annotation according to claim 6, characterized in that: The classification unit extracts features from the annotated data, constructs a maximum entropy classification model as a medical document classification model based on the features, and constructs the maximum entropy classification model to classify the documents to be classified.

8. The medical literature classification system based on intelligent annotation according to claim 7, characterized in that: A denoising unit is communicatively connected between the integration unit and the annotation unit; the denoising unit is used to filter and denoise the annotated data before annotating the annotated data, so as to remove the same document from multiple sources.

Images