Exploring and Predicting Top streamed spotify songs using Lazy learners

Stream prediction holds great significance for analysing and recommending the song based on the viewer or most played category. Prediction involves instances that can be recommended based on playlist presence high viewer count, adding complexity to the prediction process. The paper mainly helps to unfold the convergence of machine learning, song prediction, and multimedia analysis within the context of the Spotify dataset. This paper focuses on the application of data exploration, ML and Data science techniques to unravel patterns and insights within the Spotify dataset. Leveraging the rich and diverse data provided by Spotify, encompassing user interactions, audio features, and textual information, the study aims to develop robust classification models for music-related tasks. The project explores various machine learning algorithms like lazy classifier and lazy regressor a combination of seventy algorithms trained into multiple classes to discern intricate patterns in the data and enhance the accuracy of genre classification. In addition to EDA analysis of features, the project also endeavors to predict Artist Highlights, Playlist Dynamics, Regional Variations. The anticipated outcomes include the development of accurate models. Evaluation metrics such as precision, R-Score, R Sq Score, recall, mean-R Sq Score are employed to assess the performance of the models in handling prediction data associated with diverse music genres.