• Tanay Agarwal
• Akshay Sethia
• M. Indumathy

In International sports and games, basketball is one of the most engaged sports across the world. Also, in recent years, the evolution of fantasy leagues have exercised the sporting fans of these games to show more enthusiasm and analyze the outcomes more intensely. In the world of basketball, the National Basketball Association (NBA) holds a special place in terms of fans, media, skills, and the value marked with each player. Throughout history, the volume of data and statistics that are collected each year for every player has increased drastically. With the availability of such enormous data, it is quite challenging to analyze and predict the outcomes of future games. The study adopts the unsupervised learning approach applying the K-Nearest Neighbors (KNN) algorithm to predict the statistics of the players in the forthcoming season. The projected statistics converted to fantasy points are used in the fantasy leagues. The comparison result confirms that the approach we have adopted display great ability in predicting the statistics regardless of the complexity in the data features.

Statistics, NBA, Fantasy Points, Similarity, Value-Based Drafting

"Unsupervised Learning Approach to Compete in Fantasy Leagues ", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.7, Issue 5, page no.56-58, May-2020, Available :http://www.jetir.org/papers/JETIRDV06014.pdf

"Unsupervised Learning Approach to Compete in Fantasy Leagues ", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.7, Issue 5, page no. pp56-58, May-2020, Available at : http://www.jetir.org/papers/JETIRDV06014.pdf