Abstract
The interactions between subscribers of the health-related social networking (HSNs) platforms rise the production and sharing of a huge amount of multimedia content, daily, by permitting them to upload their medical images. These images become the centre of communication in various multilingual expressions immediately describing observations, comments and health checkups. As a part of this exchange, it is clear that these spaces are a valuable source of subscribers-generated information. Besides, it is still an open question to enable subscribers to investigate relevant information, due to the diversity of the available content. So, it is vital to engage new mechanisms in order to pull out information and acquaintance from this content. For this purpose, we have implemented a content analysis model of health-related information to get an overview of the medical content available. We present a semantic terms-based approach to pull out pertinent terms and concepts from the text material. As a result, notable extracted terms and keywords will be applied, subsequently, to present to annotate medical images, to direct users to an appropriate seeking task, through the SN site. So, the analysis method concentrates on algorithms based on statistical methods and external multilingual semantic resources to cover and treat this situation. It is essential also to deal with such ambiguities causing the efficacy decreasing of the search function. Our study is validated by a set of experiments and compared with some existing models. Experimental results have ensured that the presented model has better findings, in terms of performance and satisfaction.