Anime Popularity Prediction Before Huge Investments: a Multimodal Approach Using Deep Learning

TL;DR

This paper introduces a multimodal deep learning approach using text and image data to predict anime popularity, demonstrating significant improvements over traditional methods and highlighting the value of incorporating visual information.

Contribution

It presents the first multimodal dataset and deep learning model for anime popularity prediction, combining text and images to improve accuracy.

Findings

Deep neural network achieved MSE of 0.011

Multimodal approach outperformed traditional text-only methods

Incorporating images significantly enhances prediction accuracy

Abstract

In the japanese anime industry, predicting whether an upcoming product will be popular is crucial. This paper presents a dataset and methods on predicting anime popularity using a multimodal textimage dataset constructed exclusively from freely available internet sources. The dataset was built following rigorous standards based on real-life investment experiences. A deep neural network architecture leveraging GPT-2 and ResNet-50 to embed the data was employed to investigate the correlation between the multimodal text-image input and a popularity score, discovering relevant strengths and weaknesses in the dataset. To measure the accuracy of the model, mean squared error (MSE) was used, obtaining a best result of 0.011 when considering all inputs and the full version of the deep neural network, compared to the benchmark MSE 0.412 obtained with traditional TF-IDF and PILtotensor vectorizations. This is the first proposal to address such task with multimodal datasets, revealing the substantial benefit of incorporating image information, even when a relatively small model (ResNet-50) was used to embed them.