# Predicting Movie Genres Based on Plot Summaries

**Authors:** Quan Hoang  
**arXiv:** [1801.04813](https://arxiv.org/abs/1801.04813)  
**Date:** 2018-01-15  
**Summary page:** https://www.summarizepaper.com/en/paper/1801.04813v1/

## Key points

- The paper explores Machine Learning methods to predict movie genres based on plot summaries
- Naive Bayes, Word2Vec+XGBoost, and Recurrent Neural Networks are used for text classification
- K-binary transformation, rank method, and probabilistic classification with learned probability threshold are used for the multi-label problem involved in genre tagging
- Experiments conducted on more than 250,000 movies show that using Gated Recurrent Units (GRU) neural networks for the probabilistic classification with learned probability threshold approach achieves the best results on the test set
- The model attains a Jaccard Index of 50.0%, an F-score of 0.56, and a hit rate of 80.5%
- This study provides valuable insights into how Machine Learning can be effectively used to predict movie genres based on plot summaries
- GRU neural networks perform superiorly in this task

## Summary

In the paper titled "Predicting Movie Genres Based on Plot Summaries" by Quan Hoang, various Machine Learning methods are explored to predict movie genres based on plot summaries. The study employs Naive Bayes, Word2Vec+XGBoost, and Recurrent Neural Networks for text classification. For the multi-label problem involved in genre tagging, K-binary transformation, rank method, and probabilistic classification with learned probability threshold are used. The experiments conducted on more than 250,000 movies show that using Gated Recurrent Units (GRU) neural networks for the probabilistic classification with learned probability threshold approach achieves the best results on the test set. The model attains a Jaccard Index of 50.0%, an F-score of 0.56, and a hit rate of 80.5%. This study provides valuable insights into how Machine Learning can be effectively used to predict movie genres based on plot summaries and highlights the superior performance of GRU neural networks in this task.