A Model for Multilingual Terminology Extraction via a Medical Social Network

A growing majority of healthcare professionals, even patients, are seeking out medical social networks (MSNs) to acquire health information. The influence of MSN grows and changes daily, generating a huge volume of medical images, which are diagnosed and commented, in different languages, by several specialists. Moreover, it is necessary to employ new techniques, in order to automatically extract information and knowledge from these comments. For this reason, we propose a terms based method in order to extract the relevant concepts which can describe medical images. Significant extracted terms will be used later to facilitate their search through the medical social network site. In fact, we need to take account that existing comments are expressed in different languages to eliminate the ambiguity causing of the effectiveness's reduction of the search function. Our methodology concentrates on the harmony between statistical methods and external multilingual semantic resources. The use of external resources will improve the efficiency of the indexing process. We evaluated our methodology by a set of experiments and a comparison study with some existing approaches in literature.