Document Type



This study proposes a system for an automatic question analysis and answering using data mining techniques and OLAP. By applying information extraction techniques, relevant information is taken out from of the Internet connection questions provided by users. By using data mining, user profiles and their questions on the Internet usage are used as input for the system. Two steps clustering by SOM and K-Means algorithms are used to segment user data on their characteristics which obtained from user profile. Then output from clustering and question extraction are used for OLAP (Online Analytical Processing) to analyze the cause of the Internet usage problems. After internet usage questions have been asked to the systems, the answers for solving the Internet connections for cases are replied interactively. Results from this study reveal that it is practical to develop an automatic question answering system. The proposed automatic system had been tested with sample questions from the Internet Service Provider, Call Center in Thailand. Precision of the automatic system is about 70% which is good. This study offers useful information regarding the areas of data mining for Call Center or Customer Relationship Management Center.