Multi-dimensional Features for Prediction with Tweets

TL;DR

This paper demonstrates that combining text and location-based features from Twitter data enhances the prediction of new HIV diagnoses across US counties, offering a social media-driven approach for public health surveillance.

Contribution

It introduces a novel multi-dimensional feature construction method that integrates text and location data from Twitter to improve HIV diagnosis prediction accuracy.

Findings

Multi-dimensional features outperform text-only features in prediction.

Location-based smoothing features significantly enhance model performance.

Twitter data can serve as an effective tool for public health monitoring.

Abstract

With the rise of opioid abuse in the US, there has been a growth of overlapping hotspots for overdose-related and HIV-related deaths in Springfield, Boston, Fall River, New Bedford, and parts of Cape Cod. With a large part of population, including rural communities, active on social media, it is crucial that we leverage the predictive power of social media as a preventive measure. We explore the predictive power of micro-blogging social media website Twitter with respect to HIV new diagnosis rates per county. While trending work in Twitter NLP has focused on primarily text-based features, we show that multi-dimensional feature construction can significantly improve the predictive power of topic features alone with respect STI's (sexually transmitted infections). By multi-dimensional features, we mean leveraging not only the topical features (text) of a corpus, but also location-based information (counties) about the tweets in feature-construction. We develop novel text-location-based smoothing features to predict new diagnoses of HIV.