Predictive Mining of Comparable Entity from Comparative Question

Comparing entities are an important part of decision making process. To assist decision making it is useful to compare entities that share common utility but have distinguishing peripheral features. One possible approach is comparable entity mining from comparative questions. The technique used is weakly supervised bootstrapping approach which identifies the comparative question and extract the comparable entity. This is done by detecting whether a given question is comparative or not. A sequential pattern is generated and is called an indicative extraction pattern (IEP) if it can be used to identify comparative questions and extract comparator pairs with high reliability. This method achieves the F1-measure of 82.5 percent in comparative question identification and 83.3 percent in comparable entity extraction. In proposed system Clique grow analysis is used in which the comparable relations are extended and it is used to manipulate various query logs from users. Ranking method is used to rank the comparable entities for user’ input entity and the results show highly relevance to user’s