Abstract
Recommendation systems are brilliant supplications to guide the users in making a decision where the users have to choose one entity amongst a potentially overwhelming set of alternative applications. This study describes the different soft computing approaches known, along with the reason for further scrutinizing about Fuzzy Soft Computing method. The experiments and analysis are done for movie recommendation systems by implementing two recommendation systems, one using traditional approach and the other using the fuzzy soft computing approach, for comparison purpose. Traditional system is implemented using cosine similarity whereas the fuzzy system is implemented using fuzzy distance. Fuzzy Recommendation System provides more accurate results shown in lower Root Mean Square Error (RMSE) and Mean Absolute Error (MAE) values. Additionally, this study has found that Fuzzy Recommendation System is able to provide more diverse results compared to Conventional. Recommendation Systems. This study demonstrates that fuzzy soft computing is capable of providing recommendation systems that are more efficient and more user friendly. Because, human behavior plays a vital role in recommendation system, I conducted a survey to collect real world data of what exactly end users think of the existing recommendation systems and their closure points of integrating human aspects into the systems. This survey was conducted for a population including both of technical as well as non-technical backgrounds, and almost all age groups. The results of this survey is analyzed and discussed in this study as well.