Medical retrieval system based on semantic logic

Abstract

The invention is applicable to the technical field of intelligent medical treatment, and provides a semantic logic-based medical retrieval system, which comprises an acquisition module, a construction module, a semantic analysis module, a processing module, a storage module and a retrieval module, the construction module is used for constructing a semantic word bank by the words acquired from the target data source by the acquisition module; the semantic analysis module is used for classifying words in the constructed semantic word bank through semantic analysis; the storage module is used for storing the words analyzed and classified by the semantic analysis module into a storage unit; the processing module is used for determining a keyword in a to-be-processed statement and determining a data source within a certain range; when the system is used for searching the search word, the search is only carried out in the literature containing the medical statistics method, so that the accuracy of the search result is improved, and the search word and the word with the same semantics or association relationship with the search word are used for searching, namely, the search range is expanded, and the recall ratio is improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent medical technology, and more specifically, to a medical retrieval system based on semantic logic. Background Art

[0002] As medical research continues to deepen, researchers are increasingly demanding the authenticity, scientificity, and accuracy of research results. As an important auxiliary tool for medical research, medical statistics is closely related to the design, implementation, and results of medical research projects.

[0003] The researchers chose to obtain relevant information about medical statistics by querying the database. However, there are the following problems in querying the relevant information about medical statistics through the database: the amount of information in the existing database is very large, and when searching for relevant information about medical statistics, there will be problems such as missed detection and low retrieval accuracy, which causes researchers to search the database repeatedly and waste a lot of time; most of the existing common medical retrieval systems obtain the final output through linear networks, do not effectively use other feature information of the sentence, have weak generalization ability, and have poor keyword extraction effect Summary of the invention

[0004] In view of the deficiencies in the prior art, the purpose of the present invention is to provide a medical retrieval system based on semantic logic. When searching for a search term, it is equivalent to searching only in the literature containing medical statistical methods, thereby improving the accuracy of the search results. At the same time, since each classified word in the storage unit is a word with the same semantics or association relationship, when searching based on the search term stored in the storage unit, the search term is actually expanded. The search is not simply performed using the search term, but the search is performed using the search term and the words with the same semantics or association relationship as the search term, namely, the search scope is expanded and the recall rate is improved. The improvement in the recall rate and the precision rate makes it unnecessary for users to search repeatedly, thereby greatly saving the user's search time.

[0005] To achieve the above object, the present invention provides the following technical solutions: A medical retrieval system based on semantic logic includes an acquisition module, a construction module, a semantic analysis module, a processing module, a storage module and a retrieval module; the construction module is used to construct a semantic vocabulary with the words obtained by the acquisition module from the target data source; the semantic analysis module is used to classify the words in the constructed semantic vocabulary through semantic analysis; the storage module is used to store the words analyzed and classified by the semantic analysis module into a storage unit; the processing module is used to determine the keywords in the sentence to be processed and determine the data source within a certain range; the retrieval module obtains and outputs the target document corresponding to the search term based on the data source determined by the screening term.

[0006] The present invention is further configured as follows: the acquisition module includes a first acquisition unit and a second acquisition unit; the first acquisition unit is connected to the Internet through communication; the output end of the first acquisition unit is electrically connected to the second acquisition unit; the construction module includes a construction unit and an input unit; the input end of the input unit is electrically connected to the second acquisition unit; the output end of the input unit is connected to the construction unit through telecommunication.

[0007] The present invention is further configured as follows: the input end of the semantic analysis module is electrically connected to the construction unit; the output end of the semantic analysis module is electrically connected to the processing module; the output end of the processing module is electrically connected to the storage module; and the output end of the storage module is electrically connected to the retrieval module.

[0008] The present invention is further configured as follows: the first acquisition unit is used to obtain literature containing medical statistical methods from an Internet target data source library as a target data source; the second acquisition unit is used to obtain target words and extended words of the target words from the target data source obtained by the first acquisition module.

[0009] The present invention is further configured as follows: the input unit inputs the target word and the extended sentence of the target word acquired by the second acquisition unit into the construction unit; the construction unit constructs a semantic vocabulary using the words acquired from the target data source through the input target word and the extended sentence of the target word.

[0010] The present invention is further configured as follows: the processing module includes a neural network model construction unit, a target language model generation unit, a target sentence block generation unit, a key information generation unit, a keyword determination unit and a data source determination unit.

[0011] The present invention is further configured as follows: the construction unit of the neural network model extracts data from a semantic vocabulary for construction; the target language model generation unit trains and adjusts the neural network model based on the keywords in the semantic vocabulary; the markup sentence block generation unit constructs target sentence blocks based on the target language model; the key information generation unit generates key information based on the multiple groups of target sentence blocks; the keyword determination unit receives the sentence to be processed and obtains the relevant keyword dictionary in the semantic vocabulary to determine the keywords in the sentence to be processed.

[0012] The present invention is further configured as follows: the data source determination unit compares the keywords in the sentence to be processed with the words in the storage unit that have been analyzed and classified by the semantic analysis module, determines the filtering words within an error range, and determines the data source within a certain range through the filtering words.

[0013] The advantages of the present invention are: 1. When searching for a search term, the present invention is equivalent to searching only in documents containing medical statistical methods, thereby improving the accuracy of the search results; at the same time, since each classified word in the storage unit is a word with the same semantics or association relationship, when searching based on the words stored in the storage unit using the search term, the search term is actually expanded, and the search is not simply performed using the search term, but the search is performed using the search term and the words with the same semantics or association relationship as the search term, that is, the search scope is expanded and the recall rate is improved; the improvement of the recall rate and the precision rate makes it unnecessary for users to search repeatedly, thereby greatly saving the user's search time.

[0014] 2. The present invention uses a pre-trained language model as a feature extraction tool for sentences to be processed, and adds rich underlying information such as part-of-speech features and dictionary features to expand the differentiation of words, so that the difference between keywords and non-keywords in the sentences to be processed is clearer, thereby more accurately obtaining the keywords in the sentences to be processed. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 This is a flow chart of a medical retrieval system based on semantic logic of the present invention.

[0016] Figure 2 It is a composition diagram of the processing module of the present invention. DETAILED DESCRIPTION

[0017] 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.

[0018] 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.

[0019] 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.

[0020] For example, see Figure 1-2 , the present invention provides the following technical solutions: A medical retrieval system based on semantic logic, specifically, a collection module, a construction module, a semantic analysis module, a processing module, a storage module and a retrieval module; the construction module is used to construct a semantic vocabulary with the words obtained by the collection module from the target data source; the semantic analysis module is used to classify the words in the constructed semantic vocabulary through semantic analysis; the storage module is used to store the words analyzed and classified by the semantic analysis module into a storage unit; the processing module is used to determine the keywords in the sentence to be processed and determine the data source within a certain range; the retrieval module obtains and outputs the target document corresponding to the retrieval word based on the data source determined by the screening word.

[0021] The acquisition module includes a first acquisition unit and a second acquisition unit; the first acquisition unit is connected to the Internet through communication; the output end of the first acquisition unit is electrically connected to the second acquisition unit; the construction module includes a construction unit and an input unit; the input end of the input unit is electrically connected to the second acquisition unit; the output end of the input unit is connected to the construction unit through telecommunication.

[0022] The input end of the semantic analysis module is electrically connected to the construction unit; the output end of the semantic analysis module is electrically connected to the processing module; the output end of the processing module is electrically connected to the storage module; and the output end of the storage module is electrically connected to the retrieval module.

[0023] The first acquisition unit is used to obtain literature containing medical statistical methods as a target data source from an Internet target data source library; the second acquisition unit is used to obtain target words and extended words of the target words from the target data source obtained by the first acquisition module.

[0024] The processing module includes a neural network model construction unit, a target language model generation unit, a target sentence block generation unit, a key information generation unit, a keyword determination unit and a data source determination unit.

[0025] Working principle of the first embodiment: When searching for a search term, it is equivalent to searching only in the literature containing medical statistical methods, which improves the accuracy of the search results; at the same time, since each classified word in the storage unit is a word with the same semantics or association relationship, when the search term is used to search based on the word stored in the storage unit, the search term is actually expanded. It is not simply searched with the search term, but the search term and the words with the same semantics or association relationship as the search term are used to search, that is, the search scope is expanded and the recall rate is improved; the improvement of the recall rate and the precision rate makes it unnecessary for users to search repeatedly, which greatly saves the user's search time.

[0026] For example 2, please refer to Figure 1-2, this embodiment 2 makes the following improvements on the basis of embodiment 1. Specifically, the construction unit of the neural network model extracts data from the semantic vocabulary for construction; the target language model generation unit trains and adjusts the neural network model based on the keywords in the semantic vocabulary; the markup sentence block generation unit constructs the target sentence block based on the target language model; the key information generation unit generates key information based on multiple groups of target sentence blocks; the keyword determination unit receives the sentence to be processed and obtains the relevant keyword dictionary in the semantic vocabulary to determine the keyword in the sentence to be processed.

[0027] The data source determination unit compares the keywords in the sentence to be processed with the words in the storage unit that have been analyzed and classified by the semantic analysis module, determines the screening words within the error range, and determines the data source within a certain range through the screening words.

[0028] Working principle of the second embodiment: By using the pre-trained language model as a feature extraction tool for the sentences to be processed and adding rich underlying information such as part-of-speech features and dictionary features to expand the differentiation of words, the difference between keywords and non-keywords in the sentences to be processed is made clearer, so that the keywords in the sentences to be processed can be obtained more accurately.

[0029] The keywords generate semantic vectors of the corresponding keywords respectively; according to the grammar and semantic features of semantic logic, the medical knowledge is divided into multiple semantic blocks, and the corresponding semantic vectors are generated respectively.

[0030] Specifically, the semantic vector contains the deep semantic information of the keywords, which is beneficial to the subsequent key information extraction, semantic matching, text classification and other tasks. At the same time, by processing each keyword, the generated semantic vector has better generalization and robustness, which can reduce the problems of word ambiguity and description differences in legal documents. The construction of the target sentence block is based on the target language model. The weight coefficients of the semantic vectors corresponding to each semantic block and the semantic vectors of each keyword are calculated respectively. The sum of the weight coefficients of the semantic vector corresponding to any semantic block and the semantic vectors of all keywords is calculated to obtain the comprehensive weight coefficient corresponding to each semantic block. When it is determined that there is a comprehensive weight coefficient greater than the set weight coefficient, the semantic block corresponding to the comprehensive weight coefficient is used as the target sentence block.

[0031] Specifically, by simultaneously focusing on the semantic vectors corresponding to each semantic block and the semantic vectors of each keyword, it is possible to automatically focus on the most critical sentence fragments (i.e., the sentence fragments corresponding to the target sentence block) from complex medical papers, thereby effectively alleviating the limitations of keyword matching, capturing the implicit contextual associations in medical papers, and improving the accuracy and coverage of key information positioning.

[0032] Key information construction is to identify key sentences in at least one target sentence block among multiple target sentence blocks to form an initial key sentence set, cluster the key sentences in the initial key sentence set using syntactic and semantic similarity algorithms to obtain multiple target key sentences, and reorganize and sort the multiple target key sentences according to the structure of the medical paper to generate key information.

[0033] The keywords in the sentence to be processed are determined by first receiving the sentence to be processed, and obtaining a keyword dictionary corresponding to the sentence to be processed, obtaining dictionary features of the keyword dictionary based on a feature extraction model, and obtaining sentence features of the sentence to be processed based on a language model, and performing part-of-speech tagging on the sentence to be processed to obtain part-of-speech features corresponding to the part-of-speech of the sentence to be processed.

[0034] Specifically, some target sentence blocks that may contain key information can be located from the medical knowledge base, but the fragments in the target sentence blocks may have some redundant, irrelevant or incoherent problems. Therefore, the key sentences in the target sentence blocks can be screened and reorganized to form key information that is clear and semantically coherent.

[0035] Based on the above-mentioned dictionary features, part-of-speech features and sentence features, the keywords in the sentence to be processed are determined. Based on the above-mentioned dictionary features, part-of-speech features and sentence features, the keywords in the sentence to be processed are determined.

[0036] 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.

[0037] 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.

[0038] 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.

[0039] 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.

[0040] 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 retrieval system based on semantic logic, characterized by: It includes acquisition module, construction module, semantic analysis module, processing module, storage module and retrieval module; The construction module is used to construct a semantic word library with the words acquired by the acquisition module from the target data source; The semantic analysis module is to classify the words in the constructed semantic vocabulary through semantic analysis; The storage module is used to store the words analyzed and classified by the semantic analysis module into a storage unit; The processing module is used to determine the keywords in the sentence to be processed and determine the data source within a certain range; The search module obtains and outputs target documents corresponding to the search terms based on the data source determined by the screening terms.

2. A medical retrieval system based on semantic logic according to claim 1, characterized in that: The acquisition module includes a first acquisition unit and a second acquisition unit; the first acquisition unit is connected to the Internet via communication; the output end of the first acquisition unit is electrically connected to the second acquisition unit; The construction module includes a construction unit and an input unit; The input end of the recording unit is electrically connected to the second collection unit; the output end of the recording unit is electrically connected to the construction unit.

3. A medical retrieval system based on semantic logic according to claim 2, characterized in that: The input end of the semantic analysis module is electrically connected to the construction unit; the output end of the semantic analysis module is electrically connected to the processing module; the output end of the processing module is electrically connected to the storage module; and the output end of the storage module is electrically connected to the retrieval module.

4. A medical retrieval system based on semantic logic according to claim 3, characterized in that: The first acquisition unit is used to obtain literature containing medical statistical methods from an Internet target data source library as a target data source; The second acquisition unit is used to acquire the target word and the extended word of the target word from the target data source acquired by the first acquisition module.

5. A medical retrieval system based on semantic logic according to claim 4, characterized in that: The input unit inputs the target word and the extended sentence of the target word acquired by the second acquisition unit into the construction unit; The construction unit constructs a semantic word library by using the input target words and the extended sentences of the target words to obtain the words from the target data source.

6. A medical retrieval system based on semantic logic according to claim 5, characterized in that: The processing module includes a neural network model construction unit, a target language model generation unit, a target sentence block generation unit, a key information generation unit, a keyword determination unit and a data source determination unit.

7. A medical retrieval system based on semantic logic according to claim 6, characterized in that: The construction unit of the neural network model is constructed by extracting data from a semantic vocabulary; The target language model generation unit trains and adjusts the neural network model based on the keywords in the semantic vocabulary; The marked sentence chunk generating unit constructs a target sentence chunk based on the target language model; The key information generating unit generates key information based on the multiple groups of target sentence blocks; The keyword determination unit receives the sentence to be processed and obtains a related keyword dictionary in the semantic vocabulary to determine the keywords in the sentence to be processed.

8. A medical retrieval system based on semantic logic according to claim 7, characterized in that: The data source determination unit compares the keywords in the sentence to be processed with the words in the storage unit that are analyzed and classified by the semantic analysis module, determines the screening words within the error range, and determines the data source within a certain range through the screening words.

Images