Gene ontology medical term vector representation method

Abstract

The invention relates to the technical field of natural language processing, in particular to a gene ontology medical term vector representation method. The problem that the vector representation of the medical terms does not consider the individual semantic weights of the medical terms in the related vector representation of the medical terms is solved. According to the technical scheme, the gene ontology medical term vector representation method comprises the following steps of S1, data collection, training and preprocessing; s2, developing a Gene Ontology: GO medical term vector on the basis of the medical term vector of the Word2Vec model; s3, calculating individual semantic weights of the GO medical terms; and S4, GO medical term vector representation based on semantic weight. The method has the beneficial effects that the medical term vector representation efficiency is improved, and the efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of natural language processing, and in particular to a gene ontology medical term vector representation method. Background Art

[0002] Medical term vectors are crucial for AI-based applications such as disease diagnosis, information retrieval, and disease classification, and by improving the efficiency of medical term vector representations, the efficiency of these methods can be improved. Reference URL: (https: / / radimrehurek.com / gensim / models / word2vec.html), (https: / / geneontology.org / ), existing Word2Vec methods and vector representations of medical terms based on Gene Ontology (GO) do not consider the individual semantic weights of medical terms in their associated vector representations, which reduces efficiency and effectiveness.

[0003] The present invention solves the following problems:

[0004] (1) How to use the classification structure of the GO graph to calculate the individual semantic weights of medical terms;

[0005] (2) How to integrate semantic weights and Gene Ontology medical term vectors based on the Word2Vec method to improve its vector representation efficiency and increase efficiency. Summary of the invention

[0006] The object of the present invention is to provide a method for calculating the individual semantic weights of GO medical terms by using the PMC Open Access Subset (https: / / pmc.ncbi.nl m.nih.gov / tools / textmining / ) for training a Word2Vec model of a data set GeneOntology (GO) graph structure (https: / / geneontology.org / ) to calculate the individual semantic weights of GO medical terms, combining the calculated individual semantic weights of the medical terms with their vectors based on the Word2Vec model, improving the representation of the medical term vectors, improving efficiency, and increasing effectiveness.

[0007] In order to achieve the above-mentioned invention object, the technical solution adopted by the present invention is specifically as follows:

[0008] A method for representing gene ontology medical term vectors, comprising the following steps:

[0009] S1: data collection, training, and preprocessing;

[0010] S2: Development of medical term vectors based on the Word2Vec model Gene Ontology: GO medical term vectors;

[0011] S3: Calculation of individual semantic weights of GO medical terms;

[0012] S4: Vector representation of GO medical terms based on semantic weights.

[0013] Step S1 is specifically as follows:

[0014] Collection of PMC open access datasets: Open Access Subset ( https: / / pmc.ncbi.nlm.nih.gov / tools / textmining / ) and use this data to use Python's Gensim ( https: / / pypi.org / project / gensim / ) library to train the Word2Vec model data ( https: / / radimrehurek.com / gensim / models / word2vec.html ), and then using the Python programming language version 3.10 and the Python Natural Language Toolkit (NLTK) ( https: / / www.nltk.org / ) library to preprocess the data to learn medical terminology vectors.

[0015] Step S2 is specifically as follows:

[0016] The Word2Vec model is trained on the PMC open access dataset: OpenAccess Subset dataset to generate GO medical term vectors as shown in formula (1):

[0017]

[0018] Where: MedTerm represents the medical term in Gene Ontology (GO); the function Word2Vecmodel calculates the medical term vector after training on the PMC open access dataset: Open Access Subset dataset

[0019] Step S3 is specifically as follows:

[0020] Gene Ontology: The graphical structure of the GO graph is used to calculate the individual semantic weights of medical terms, as shown in formula (2):

[0021]

[0022] Among them: Function SW (MedTerm) GOrepresents the individual semantic weight of the medical term in the GO graph; the function |Des(Term)| represents the number of descendants of the medical term in the GO graph; N represents the total number of medical terms in the GO graph; the function Depth(Term) represents the depth of the medical term in the GO graph; α and β are linear contribution parameters, and their values ​​are both α=β=0.50.

[0023] Step S4 is specifically as follows:

[0024] By integrating the individual semantic weights of GO medical terms, they are transformed into vectors based on the Word2Vec model as shown in formula (3):

[0025]

[0026] in: The function represents the newly developed GO medical term vector that combines individual semantic weights and vectors based on the Word2Vec model; * represents the product of a scalar value and a vector.

[0027] Compared with the prior art, the present invention has the following beneficial effects:

[0028] (1) The classification structure of the GO graph can be used to calculate the individual semantic weights of medical terms;

[0029] (2) Formula (3) was used to integrate semantic weights and Gene Ontology medical term vectors based on the Word2Vec method to improve the vector representation. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention.

[0031] Figure 1 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0032] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. Of course, the specific embodiments described here are only used to explain the present invention and are not used to limit the present invention.

[0033] Example 1

[0034] like Figure 1 As shown: A method for representing gene ontology medical term vectors, comprising the following steps:

[0035] S1: data collection, training, and preprocessing;

[0036] S2: Development of medical term vectors based on the Word2Vec model Gene Ontology: GO medical term vectors;

[0037] S3: Calculation of individual semantic weights of GO medical terms;

[0038] S4: Vector representation of GO medical terms based on semantic weights.

[0039] Step S1 is specifically as follows:

[0040] Collection of PMC open access datasets: Open Access Subset ( https: / / pmc.ncbi.nlm.nih.gov / tools / textmining / ) and use this data to use Python's Gensim ( https: / / pypi.org / project / gensim / ) library to train the Word2Vec model data ( https: / / radimrehurek.com / gensim / models / word2vec.html ), and then using the Python programming language version 3.10 and the Python Natural Language Toolkit (NLTK) ( https: / / www.nltk.org / ) library to preprocess the data to learn medical terminology vectors.

[0041] Step S2 is specifically as follows:

[0042] The Word2Vec model is trained on the PMC open access dataset: OpenAccess Subset dataset to generate GO medical term vectors as shown in formula (1):

[0043]

[0044] Where: MedTerm represents the medical term in Gene Ontology (GO); the function Word2Vecmodel calculates the medical term vector after training on the PMC open access dataset: Open Access Subset dataset

[0045] Step S3 is specifically as follows:

[0046] Gene Ontology: The graphical structure of the GO graph is used to calculate the individual semantic weights of medical terms, as shown in formula (2):

[0047]

[0048] Among them: Function SW (MedTerm) GOrepresents the individual semantic weight of the medical term in the GO graph; the function |Des(Term)| represents the number of descendants of the medical term in the GO graph; N represents the total number of medical terms in the GO graph; the function Depth(Term) represents the depth of the medical term in the GO graph; α and β are linear contribution parameters, and their values ​​are both α=β=0.50.

[0049] Step S4 is specifically as follows:

[0050] By integrating the individual semantic weights of GO medical terms, they are transformed into vectors based on the Word2Vec model as shown in formula (3):

[0051]

[0052] in: The function represents the newly developed GO medical term vector that combines individual semantic weights and vectors based on the Word2Vec model; * represents the product of a scalar value and a vector.

[0053] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for representing gene ontology medical term vectors, characterized in that: The following steps are involved: S1: data collection, training, and preprocessing; S2: Development of gene ontology based on medical term vectors of Word2Vec model: GO medical term vectors; S3: Calculation of individual semantic weights of GO medical terms; S4: Vector representation of GO medical terms based on semantic weights.

2. A method for representing gene ontology medical term vectors according to claim 1, characterized in that: Step S2 is specifically as follows: The Word2Vec model is trained on the PMC open access dataset: OpenAccess Subset dataset to generate GO medical term vectors as shown in formula (1): Among them: MedTerm represents the medical term in GO; the function Word2Vecmodel calculates the medical term vector after training on the PMC open access dataset: OpenAccess Subset dataset 3. A method for representing gene ontology medical term vectors according to claim 1, characterized in that: Step S3 is specifically as follows: Gene Ontology: The graphical structure of the GO graph is used to calculate the individual semantic weights of medical terms, as shown in formula (2): Among them: Function SW (MedTerm) GO represents the individual semantic weight of the medical term in the GO graph; the function |Des(Term)| represents the number of descendants of the medical term in the GO graph; N represents the total number of medical terms in the GO graph; the function Depth(Term) represents the depth of the medical term in the GO graph; α and β are linear contribution parameters.

4. A method for representing gene ontology medical term vectors according to claim 1, characterized in that: Step S4 is specifically as follows: By integrating the individual semantic weights of GO medical terms, they are transformed into vectors based on the Word2Vec model as shown in formula (3): in: The function represents the newly developed GO medical term vector that combines individual semantic weights and vectors based on the Word2Vec model; * represents the product of a scalar value and a vector.

Images