PROCESSING COMPLEX SENTENCES FOR INFORMATION EXTRACTION APPROVED:

Genomic research in the last decade has resulted in a huge production of data in the form of microarray experiments, sequence information and publications. The data generated by these experiments are highly connected; the results from sequence analysis and microarrays depend on functional information and signal transduction pathways cited in peer-reviewed publications for evidence. Though scientists in the field are aided by many databases of biochemical interactions available for use online, a majority of these databases are curated by domain experts, often from literature. Information extraction from text has therefore been pursued actively as an attempt to present knowledge from published material in a computer readable format. An automated extraction tool would not only save time and efforts, but also pave the way to discover hitherto unknown information implicitly conveyed in text. This thesis introduces a way of processing information in biomedical text to result in simple syntactic constructs. The aim is to break up complex sentence structures to processable chunks. Information extraction is made easier by the inherent simplicity and dependencies between the chunks. The modul