Saved in:
Bibliographic Details
Main Authors: Kim, Yunwoo, Hwang, Junhyuk
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2508.21650
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909760169705472
author Kim, Yunwoo
Hwang, Junhyuk
author_facet Kim, Yunwoo
Hwang, Junhyuk
contents We present a machine learning approach for predicting social media engagement (comments and likes) from emotional and temporal features. The dataset contains 600 songs with annotations for valence, arousal, and related sentiment metrics. A multi target regression model based on HistGradientBoostingRegressor is trained on log transformed engagement ratios to address skewed targets. Performance is evaluated with both a custom order of magnitude accuracy and standard regression metrics, including the coefficient of determination (R^2). Results show that emotional and temporal metadata, together with existing view counts, predict future engagement effectively. The model attains R^2 = 0.98 for likes but only R^2 = 0.41 for comments. This gap indicates that likes are largely driven by readily captured affective and exposure signals, whereas comments depend on additional factors not represented in the current feature set.
format Preprint
id arxiv_https___arxiv_org_abs_2508_21650
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Predicting Social Media Engagement from Emotional and Temporal Features
Kim, Yunwoo
Hwang, Junhyuk
Machine Learning
We present a machine learning approach for predicting social media engagement (comments and likes) from emotional and temporal features. The dataset contains 600 songs with annotations for valence, arousal, and related sentiment metrics. A multi target regression model based on HistGradientBoostingRegressor is trained on log transformed engagement ratios to address skewed targets. Performance is evaluated with both a custom order of magnitude accuracy and standard regression metrics, including the coefficient of determination (R^2). Results show that emotional and temporal metadata, together with existing view counts, predict future engagement effectively. The model attains R^2 = 0.98 for likes but only R^2 = 0.41 for comments. This gap indicates that likes are largely driven by readily captured affective and exposure signals, whereas comments depend on additional factors not represented in the current feature set.
title Predicting Social Media Engagement from Emotional and Temporal Features
topic Machine Learning
url https://arxiv.org/abs/2508.21650