Indication Similarity Of Drugs Based On Chinese Word Segmentation Technology

Drug indications similarity research is mining a variety of drugs with the same or similar indications of information retrieval technology, it can provide the technical supports foe users to quickly find alternative medicines. The core task of the drugs indications similarity research is to compute the drug indications similarity and construct the drugs similarity knowledge base, which provides the basis for retrieving similar drug. In this paper, we have integrated use the Chinese word segmentation technology, term extraction technology and method of building knowledge base, and establish the similar indications automatic detection model. The main research content is as follows:(1)Drug term extraction algorithm is proposed. According to fetching the data source, we preprocess the collected drugs data and use the Chinese word segmentation technology effectively to extract drug indications term, at the same time build user dictionary and thesaurus.(2)The similar knowledge base is established. We building the weighting database on the basis of indications term, give different weight value to drug indications according to the different importance of indications to the different drug, which provides the technical support for similarity calculation and provide a basis for the system to improve computing drug similarity.(3)Put forward drug similarity calculation method. The system will search out the drugs with the same main indication from the similar knowledge database according to the input of drug name. The system will computes the drug similarity by drug similarity calculation method by different weights between drug indications. The algorithm will Filter and give the drug sorted subset, the system will recommends some reasonable drug according to the needs of users.In this paper, the experimental results show that the drug similarity computing research is in full compliance with the query results of expectations in particular sample library experiments. Retrieval result is consistent with people’s expectations, which proves the feasibility and effectiveness of the proposed method. In this paper, computing semantic similarity is applied to the calculation of similarity between drugs creatively, which provides a quick and convenient way of similar drug retrieval for medical staff and patients, and it effectively reduce the cognitive burden of people. This article research results can provide theoretical basic and technical support for the development of medical information retrieval in the field of intelligent such as similar drug retrieval and online recommendation.