A Dynamic Topic Identification and Labeling Approach of COVID-19 Tweets
- URL: http://arxiv.org/abs/2109.02462v1
- Date: Fri, 13 Aug 2021 16:51:04 GMT
- Title: A Dynamic Topic Identification and Labeling Approach of COVID-19 Tweets
- Authors: Khandaker Tayef Shahriar, Iqbal H. Sarker, Muhammad Nazrul Islam and
Mohammad Ali Moni
- Abstract summary: The COVID-19 epidemic has affected the use of social media by many people across the globe.
This paper formulates the problem of dynamically identifying key topics with proper labels from COVID-19 Tweets to provide an overview of wider public opinion.
- Score: 3.097385298197292
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: This paper formulates the problem of dynamically identifying key topics with
proper labels from COVID-19 Tweets to provide an overview of wider public
opinion. Nowadays, social media is one of the best ways to connect people
through Internet technology, which is also considered an essential part of our
daily lives. In late December 2019, an outbreak of the novel coronavirus,
COVID-19 was reported, and the World Health Organization declared an emergency
due to its rapid spread all over the world. The COVID-19 epidemic has affected
the use of social media by many people across the globe. Twitter is one of the
most influential social media services, which has seen a dramatic increase in
its use from the epidemic. Thus dynamic extraction of specific topics with
labels from tweets of COVID-19 is a challenging issue for highlighting
conversation instead of manual topic labeling approach. In this paper, we
propose a framework that automatically identifies the key topics with labels
from the tweets using the top Unigram feature of aspect terms cluster from
Latent Dirichlet Allocation (LDA) generated topics. Our experiment result shows
that this dynamic topic identification and labeling approach is effective
having the accuracy of 85.48\% with respect to the manual static approach.
Related papers
- Towards Scalable Topic Detection on Web via Simulating Levy Walks Nature of Topics in Similarity Space [55.97416108140739]
We present a novel, yet very powerful Explore-Exploit (EE) approach to group topics by simulating Levy walks nature in the similarity space.
Experiments on two public data sets demonstrate that our approach is not only comparable to the state-of-the-art methods in terms of effectiveness but also significantly outperforms the state-of-the-art methods in terms of efficiency.
arXiv Detail & Related papers (2024-07-26T07:19:46Z) - Exploring a Hybrid Deep Learning Framework to Automatically Discover
Topic and Sentiment in COVID-19 Tweets [2.3940819037450987]
COVID-19 has created a major public health problem worldwide and other problems such as economic crisis, unemployment, mental distress, etc.
The pandemic is deadly in the world and involves many people not only with infection but also with problems, stress, wonder, fear, resentment, and hatred.
Twitter is a highly influential social media platform and a significant source of health-related information, news, opinion and public sentiment.
arXiv Detail & Related papers (2023-12-02T16:58:17Z) - HuBERTopic: Enhancing Semantic Representation of HuBERT through
Self-supervision Utilizing Topic Model [62.995175485416]
We propose a new approach to enrich the semantic representation of HuBERT.
An auxiliary topic classification task is added to HuBERT by using topic labels as teachers.
Experimental results demonstrate that our method achieves comparable or better performance than the baseline in most tasks.
arXiv Detail & Related papers (2023-10-06T02:19:09Z) - ManiTweet: A New Benchmark for Identifying Manipulation of News on Social Media [74.93847489218008]
We present a novel task, identifying manipulation of news on social media, which aims to detect manipulation in social media posts and identify manipulated or inserted information.
To study this task, we have proposed a data collection schema and curated a dataset called ManiTweet, consisting of 3.6K pairs of tweets and corresponding articles.
Our analysis demonstrates that this task is highly challenging, with large language models (LLMs) yielding unsatisfactory performance.
arXiv Detail & Related papers (2023-05-23T16:40:07Z) - Changes in European Solidarity Before and During COVID-19: Evidence from
a Large Crowd- and Expert-Annotated Twitter Dataset [77.27709662210363]
We introduce the well-established social scientific concept of social solidarity and its contestation, anti-solidarity, as a new problem setting to supervised machine learning in NLP.
We annotate 2.3k English and German tweets for (anti-)solidarity expressions, utilizing multiple human annotators and two annotation approaches (experts vs. crowds)
Our results show that solidarity became increasingly salient and contested during the COVID-19 crisis.
arXiv Detail & Related papers (2021-08-02T17:03:12Z) - COVID-19 and Big Data: Multi-faceted Analysis for Spatio-temporal
Understanding of the Pandemic with Social Media Conversations [4.07452542897703]
Social media platforms have served as a vehicle for the global conversation about COVID-19.
We present a framework for analysis, mining, and tracking the critical content and characteristics of social media conversations around the pandemic.
arXiv Detail & Related papers (2021-04-22T00:45:50Z) - Pulse of the Pandemic: Iterative Topic Filtering for Clinical
Information Extraction from Social Media [1.5938324336156293]
The rapid evolution of the COVID-19 pandemic has underscored the need to quickly disseminate the latest clinical knowledge during a public-health emergency.
We present an unsupervised, iterative approach to mine clinically relevant information from social media data.
This approach identifies granular topics and tweets with high clinical relevance from a set of about 52 million COVID-19-related tweets.
arXiv Detail & Related papers (2021-02-13T01:01:04Z) - Understanding the Spatio-temporal Topic Dynamics of Covid-19 using
Nonnegative Tensor Factorization: A Case Study [1.6328866317851185]
This paper proposes a representation of social media data and Non-negative Factorization (NTF) to identify the topics discussed in social media data.
A case study on the Australia Twittersphere is presented to identify visualize the topic dynamics on and off the Covid-19.
arXiv Detail & Related papers (2020-09-19T15:16:28Z) - Analyzing COVID-19 on Online Social Media: Trends, Sentiments and
Emotions [44.92240076313168]
We analyze the affective trajectories of the American people and the Chinese people based on Twitter and Weibo posts between January 20th, 2020 and May 11th 2020.
By contrasting two very different countries, China and the Unites States, we reveal sharp differences in people's views on COVID-19 in different cultures.
Our study provides a computational approach to unveiling public emotions and concerns on the pandemic in real-time, which would potentially help policy-makers better understand people's need and thus make optimal policy.
arXiv Detail & Related papers (2020-05-29T09:24:38Z) - Privacy-Aware Recommender Systems Challenge on Twitter's Home Timeline [47.434392695347924]
RecSys 2020 Challenge organized by ACM RecSys in partnership with Twitter using this dataset.
This paper touches on the key challenges faced by researchers and professionals striving to predict user engagements.
arXiv Detail & Related papers (2020-04-28T23:54:33Z) - COVID-19 on Social Media: Analyzing Misinformation in Twitter
Conversations [22.43295864610142]
We collected streaming data related to COVID-19 using the Twitter API, starting March 1, 2020.
We identified unreliable and misleading contents based on fact-checking sources.
We examined the narratives promoted in misinformation tweets, along with the distribution of engagements with these tweets.
arXiv Detail & Related papers (2020-03-26T09:48:24Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.