Information Extraction Using RDBMS and Stemming Algorithm

Information extraction systems are traditionally implemented as a pipeline of special-purpose processing modules targeting the extraction of a fussy form of data. A major drawback of such an attack is that whenever a new extraction goal emerges or a module is improved, extraction has to be replayed from scratch to the entire text corpus even though but a minor portion of the corpus might be involved. In this report, we report a novel approach for information extraction in which extraction needs are extracted in the configuration of database inquiries, which are measured and optimized by database systems. Using database queries for data extraction enables generic extraction and minimizes reprocessing of data by performing incremental extraction to identify which part of the data are affected by the change of components or goals. Further, our approach provides automated query generation components and stemming algorithm so that casual users do not have to learn the query language in parliamentary procedure to do the extraction. To show the feasibility of our incremental extraction approach, we performed couple of experiments to highlight two significant aspects of an information extraction system: quality and efficiency of the extraction results. Our experiments show that in the outcome of deployment of a new module, our incremental extraction approach minimizes the processing time by 92 percent as compared to a traditional pipeline approach. By using our methods to a corpus of 20 million biomedical abstracts, our experiments indicate that the query performance is efficient for real-time applications. Our experiments also uncovered that our approach achieves high quality extraction results.