Decision Support System for House Purchasing using KNN (K-Nearest Neighbor) Method.

Abstract

Every prospective home buyers has their own criteria in selecting home that suit their needs. But many of these criteria are followed by the availability of more than one choice of homes that meet those criteria. Therefore, we need a computerized system that can help them to obtain a good home selection decisions based on their needs. Method used in this research is K-nearest Neighbor (KNN). This paper discusses about recommendation system to select house using 11 kinds of criteria, such as price, location, land area, building area, floor, bedroom, bathroom, maid’s room, garage/carport, electric power and water resources. This system also used a geographical information system to display the results in the form of a map. The testing was conducted by 25 users, they tried out the system and filled the satisfaction questionnaire system. The level of satisfaction obtained from two main aspects : result satisfaction and user interface design satisfaction, which is including the usability and the attractivess system. The results of this tests indicate the level of user satisfaction by 38% very satisfied, 58% satisfied and 4% not satisfied.