A Case Study on Flat Rate Estimation Based on Fuzzy Logic

The main objective of this research is to develop a system based on fuzzy logic to estimate flat rent in Bangladesh. The data set consists of 63 individual flats in different blocks of Bashundhara Residential Area, Dhaka, Bangladesh. A number of factors influence the decision of a tenant to rent a flat. Since it is not desirable to work with a large number of variables, we used one of the dimensionality reduction techniques, known as Principle Component Analysis (PCA). PCA helps in reducing the number of variables from the set of data, as well keeping hold the most of the variability in data. This paper describes the implementation of an adaptive neuro-fuzzy inference system (ANFIS)-based approach to estimate the rent of the flat. The Sugeno ANFIS model is proposed in order to develop a systematized approach of generating fuzzy rules and membership function parameters for fuzzy sets from a given set of input and output data.