Абстрактный

Domain-Independent Text Mining Framework based on Open Information Extraction

Mohamed Lotfy, Mohamed H Haggag, Ensaf H Mohamed

We propose a robust hybrid text mining solution that combines Open Information Extraction (OIE), Knowledge Discovery from Databases (KDD) and Association Rules Mining (ARM) methods to perform domain independent text mining task. For OIE, we used ClausIE system with extra added features like "Co-Reference Resolution" and "Acronyms Detection" that allow the system to extract more information and can help later in the mining process. The added features in the proposed solution will enhance both the quality and the quantity of the extracted propositions regarding both precision and recall.