A Framework for Extracting Biological Relations from Different Resources

Enas M.F.ElHouby
2015 International Journal of Computer Applications  
The World Wide Web provides a vast source of information of almost all types. Biological data specifically have increased dramatically in the past years because of the exponential growth of knowledge in biological domain. It is very difficult to search for the required data in unstructured documents. Text documents often hide valuable structured data. This data can be exploited if available as a relational table that could be used to answer queries or to perform data mining tasks. Manually
more » ... cting biological relations from published literature and transforming them into machine-understandable knowledge is a difficult task because biological domain comprises huge, dynamic, and complicated knowledge. Automatic extraction of semantic relation between biological terms from unstructured documents is challenging in information extraction and important task for deep information processing and management. In this research, a framework has been developed to extract different relations between various biological entities from documents. Semi supervised approach has been used to develop the framework. It requires the user to just provide a handful of valid pairs as initial seeds of the target relation, with no other training. Different patterns can be generated from initial seeds, and then from these patterns additional relation pairs can be extracted. The results has showed that different relations can be extracted such as gene-disease, protein-protein.
doi:10.5120/21044-3675 fatcat:u5neftbmmngrrpzlany6ssiy2y