An Advent of Data Mining to Prognosticate Users’ Successive Questions using Association Rules

Abstract

In this paper regarding to question answering system, the user herein submits the question and will be waiting for the answer as the response. If the respective system is potential at prognosticating the interest of the users as the successive questions, its performance does ameliorate greatly. This paper does predict the user’s multiple future questions, which is reckoned on the previous interaction records of the user with those of the system. Their current interactions are managed in the form of the log of questions from, which the respective user in the sessions is actually extracted. Multiple administrations are created for different types of questions. For an instance, like Finance and Sports like wise.  Reckoned on the user which may become interested in the near future. A sample questions as log is chosen as required for the purpose of the performing at the explorations. The model of this association rule at the mining is actually applied primarily to predict the multiple future question of the user.