Predicting Talent Breakout Rate using Twitter and TV data

TL;DR

The paper tackles predicting rising Japanese entertainment talents (talent breakouts) by fusing Twitter mentions with TV signals to forecast future Twitter activity three months ahead. It compares autoregressive baselines (VAR/VARMA) with ensemble and neural network approaches, finding that neural networks excel in breakout-specific precision and recall, while ensemble methods offer strong standard regression accuracy. TV data provides modest improvements and helps stabilize predictions when combined with Twitter data. The study highlights the need for additional social signals to improve volatility-driven forecasting and suggests that traditional methods can match ML performance with proper tuning, offering practical interpretability through confidence intervals.

Abstract

Early detection of rising talents is of paramount importance in the field of advertising. In this paper, we define a concept of talent breakout and propose a method to detect Japanese talents before their rise to stardom. The main focus of the study is to determine the effectiveness of combining Twitter and TV data on predicting time-dependent changes in social data. Although traditional time-series models are known to be robust in many applications, the success of neural network models in various fields (e.g.\ Natural Language Processing, Computer Vision, Reinforcement Learning) continues to spark an interest in the time-series community to apply new techniques in practice. Therefore, in order to find the best modeling approach, we have experimented with traditional, neural network and ensemble learning methods. We observe that ensemble learning methods outperform traditional and neural network models based on standard regression metrics. However, by utilizing the concept of talent breakout, we are able to assess the true forecasting ability of the models, where neural networks outperform traditional and ensemble learning methods in terms of precision and recall.