IRJET - Movie Genre Prediction from Plot Summaries by Comparing Various Classification Algorithms

This document discusses predicting movie genres from plot summaries using various classification algorithms. It analyzes Multinomial Naive Bayes, Logistic Regression, Random Forest, and Stochastic Gradient Descent algorithms on a dataset of over 40,000 movie plots labeled with 12 genres. It finds that Multinomial Naive Bayes performs best, achieving the highest AUC scores for predicting each genre individually. It then builds separate classifiers for each genre using the best performing algorithm to predict the genre of new movie plots.