An Efficient Dimensionality Reduction Techniques to Data Data Visualization

The aim of this paper, is to gain insights in to dynamics of Kabaddi. This sport has become increasingly popular, by analyzing teams’ performance in national leagues. To do so, the results of each round of matches the participating teams play are used to create a dissimilarity matrix. This matrix is then processed through two algorithms, A visualization of the outcomes of each team was achieved by the utilization of novel dimensionality techniques. This novel approach for dimensionality reduction enables the teams’ performance to be represented in a lower-dimensional manner, facilitating the visualization of complex data connections. The t-SNE algorithm, on the other hand, captures the non-linear relationships in the teams’ performance. This study attempts to discover crucial characteristics that impact team performance and give a greater knowledge of the dynamics of the sport by comparing the findings acquired using a novel dimensionality reduction technique. The data for this study comes from the Kabaddi season that took place during 2017-2018.