Rivindu Perera


IPedagogy: Question Answering System based on Web Information Clustering

Abstract
As with the excessive information growth in the web, retrieving the exact segment of information even for a simple query, has transformed to a difficult and resource expensive state. Specially, in e-learning domain it is vital to search knowledge frequently and focusing on a limited well defined search space. IPedagogy is a question answering system which works with natural language powered queries and retrieve answers from selected information clusters by reducing the search space of information retrieval. In addition, IPedagogy is empowered by several natural language processing techniques which direct the system to extract the exact answer for a given query. System is evaluated with the use of mean reciprocal rank and it is noted that system has 0.73 of average accuracy level for 10 sets of questions where each set is consisted of 35 questions.