Dynamic Sentiment Analysis: A Low-Latency System for Social Media Monitoring

Sanskar Kumar Agrahari, Arjun Kumar Das, Krishna Bhagat, Vivek Kumar Shah, Nikita Sharma, Gayathri Ramasamy

2025

Abstract

Social media sentiment analysis needs real-time tracking of public opinion therefore requires fast processing together with low latency and high accuracy. To achieve this, PySpark is used for the data preprocessing and model training process. A web application that is developed through Django lets users submit tweets that generate instant sentiment predictions whether the tweet is positive, negative, neutral, or irrelevant while Kafka manages real-time streaming of processed results. MongoDB utilizes NoSQL architecture to effectively store sentiment forecasts to- gather with their associated data. Among different trained models, Logistic regression achieved maximum accuracy according to testing while the system showed successful operation through real-time sentiment analysis with high- speed data processing and quick response times and approachable user interface which proved its usefulness for sentiment trend analysis.

Paper Citation

in Harvard Style

Agrahari S., Das A., Bhagat K., Shah V., Sharma N. and Ramasamy G. (2025). Dynamic Sentiment Analysis: A Low-Latency System for Social Media Monitoring. In Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25; ISBN 978-989-758-777-1, SciTePress, pages 641-649. DOI: 10.5220/0013940500004919

in Bibtex Style

@conference{icrdicct`2525,
author={Sanskar Agrahari and Arjun Das and Krishna Bhagat and Vivek Shah and Nikita Sharma and Gayathri Ramasamy},
title={Dynamic Sentiment Analysis: A Low-Latency System for Social Media Monitoring},
booktitle={Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25},
year={2025},
pages={641-649},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013940500004919},
isbn={978-989-758-777-1},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25
TI - Dynamic Sentiment Analysis: A Low-Latency System for Social Media Monitoring
SN - 978-989-758-777-1
AU - Agrahari S.
AU - Das A.
AU - Bhagat K.
AU - Shah V.
AU - Sharma N.
AU - Ramasamy G.
PY - 2025
SP - 641
EP - 649
DO - 10.5220/0013940500004919
PB - SciTePress