Saved in:
Bibliographic Details
Main Authors: Ng, Lynnette Hui Xian, Jaidka, Kokil, Tay, Kaiyuan, Ahuja, Hansin, Chhaya, Niyati
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2503.21000
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916762522484736
author Ng, Lynnette Hui Xian
Jaidka, Kokil
Tay, Kaiyuan
Ahuja, Hansin
Chhaya, Niyati
author_facet Ng, Lynnette Hui Xian
Jaidka, Kokil
Tay, Kaiyuan
Ahuja, Hansin
Chhaya, Niyati
contents Supervised machine-learning models often underperform in predicting user behaviors from conversational text, hindered by poor crowdsourced label quality and low NLP task accuracy. We introduce the Metadata-Sensitive Weighted-Encoding Ensemble Model (MSWEEM), which integrates annotator meta-features like fatigue and speeding. First, our results show MSWEEM outperforms standard ensembles by 14% on held-out data and 12% on an alternative dataset. Second, we find that incorporating signals of annotator behavior, such as speed and fatigue, significantly boosts model performance. Third, we find that annotators with higher qualifications, such as Master's, deliver more consistent and faster annotations. Given the increasing uncertainty over annotation quality, our experiments show that understanding annotator patterns is crucial for enhancing model accuracy in user behavior prediction.
format Preprint
id arxiv_https___arxiv_org_abs_2503_21000
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Improving User Behavior Prediction: Leveraging Annotator Metadata in Supervised Machine Learning Models
Ng, Lynnette Hui Xian
Jaidka, Kokil
Tay, Kaiyuan
Ahuja, Hansin
Chhaya, Niyati
Machine Learning
Artificial Intelligence
Supervised machine-learning models often underperform in predicting user behaviors from conversational text, hindered by poor crowdsourced label quality and low NLP task accuracy. We introduce the Metadata-Sensitive Weighted-Encoding Ensemble Model (MSWEEM), which integrates annotator meta-features like fatigue and speeding. First, our results show MSWEEM outperforms standard ensembles by 14% on held-out data and 12% on an alternative dataset. Second, we find that incorporating signals of annotator behavior, such as speed and fatigue, significantly boosts model performance. Third, we find that annotators with higher qualifications, such as Master's, deliver more consistent and faster annotations. Given the increasing uncertainty over annotation quality, our experiments show that understanding annotator patterns is crucial for enhancing model accuracy in user behavior prediction.
title Improving User Behavior Prediction: Leveraging Annotator Metadata in Supervised Machine Learning Models
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2503.21000