From text to insights: building a dynamic dashboard for sentiment analysis - A Twitter case study

Adekoya-Cole, Temitope and Fernando, Sandra (2023) From text to insights: building a dynamic dashboard for sentiment analysis - A Twitter case study. International Journal of Data Science and Analytics. ISSN 2364-4168 (In Press)

[img] Text
From Text To Insights.pdf - Accepted Version
Restricted to Repository staff only until 8 January 2026.

Download (2MB) | Request a copy
Official URL: https://link.springer.com/journal/41060/volumes-an...

Abstract / Description

Platforms like Facebook, Instagram, Snapchat, and Twitter have offered individuals and groups a public forum to share their ideas on numerous social issues and themes in the age of social media. Twitter, a popular social media, and networking tool has evolved into an important source of real-time news and a hub for public opinion. Using Twitter data, this study investigates the possibility applying of sentiment analysis to automatically identify tweets as positive, negative, or neutral and then presenting the results visually, resulting in an in-depth analysis of public opinion. This research project compares two pre-trained sentiment analysis models, VADER and TweetNLP, to determine which model best suits this task. The project concludes with the creation of a dynamic dashboard for sentiment analysis based on Twitter data. This dashboard provides users with real-time insights from text data in a user-friendly way. The dashboard's architecture and style are proficient in analysing sentiment across a wide range of topics and effectively expressing findings using visual aids such as bar charts, time graphs, and word clouds.
This research project provides evidence for practical applications of sentiment analysis techniques to organisations seeking to understand public sentiment, governments monitoring public opinion, and researchers studying sentiment across multiple topics, serving as a prototype for future development in these domains.

Item Type: Article
Uncontrolled Keywords: Sentiment analysis, Social media, Twitter, Dynamic dashboard, VADER, TweetNLP, Data visualization
Subjects: 300 Social sciences
300 Social sciences > 380 Commerce, communications & transportation
Department: School of Computing and Digital Media
Depositing User: Sandra Fernando
Date Deposited: 08 Jan 2024 11:07
Last Modified: 08 Jan 2024 11:07
URI: https://repository.londonmet.ac.uk/id/eprint/9047

Actions (login required)

View Item View Item