Longitudinal Sentiment Classification of Reddit Posts

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Nwaoha, Fabian, Gaffar, Ziyad, Chun, Ho Joon, Sokolova, Marina
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916103216693248
author Nwaoha, Fabian
Gaffar, Ziyad
Chun, Ho Joon
Sokolova, Marina
author_facet Nwaoha, Fabian
Gaffar, Ziyad
Chun, Ho Joon
Sokolova, Marina
contents We report results of a longitudinal sentiment classification of Reddit posts written by students of four major Canadian universities. We work with the texts of the posts, concentrating on the years 2020-2023. By finely tuning a sentiment threshold to a range of [-0.075,0.075], we successfully built classifiers proficient in categorizing post sentiments into positive and negative categories. Noticeably, our sentiment classification results are consistent across the four university data sets.
format Preprint
id arxiv_https___arxiv_org_abs_2401_12382
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Longitudinal Sentiment Classification of Reddit Posts
Nwaoha, Fabian
Gaffar, Ziyad
Chun, Ho Joon
Sokolova, Marina
Computation and Language
Machine Learning
Social and Information Networks
I.2.6
We report results of a longitudinal sentiment classification of Reddit posts written by students of four major Canadian universities. We work with the texts of the posts, concentrating on the years 2020-2023. By finely tuning a sentiment threshold to a range of [-0.075,0.075], we successfully built classifiers proficient in categorizing post sentiments into positive and negative categories. Noticeably, our sentiment classification results are consistent across the four university data sets.
title Longitudinal Sentiment Classification of Reddit Posts
topic Computation and Language
Machine Learning
Social and Information Networks
I.2.6
url https://arxiv.org/abs/2401.12382