Improving sentiment analysis using containerized microservices approach

Goyal, Bharat (2022) Improving sentiment analysis using containerized microservices approach. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (2MB) | Preview
Preview	PDF (Configuration manual) Download (2MB) | Preview

Abstract

Disaster management heavily relies on the monitoring of social media data, but the number of users growing is exponential, which generates a very large amount of data when a disaster happens. It is very difficult to monitor the data manually, so to monitor it requires a highly available system that can be scaled up and down easily according to the demand and perform analytics in real-time. This work aims to develop and evaluate a system that solves the issues of collecting such huge data, performs analysis in real-time, and can be set up and scaled very easily. Performance of the system will be measured by comparing memory usage, execution time, the accuracy of the models, and response time of the system. After evaluating the performance of the system, storage requirement was reduced by 31 percent when a parquet file format is used compared to a text file to store, the accuracy of the model was increased by approximately 5 percent and there was almost a 75 percent reduction in response time of the system. However, this system has some limitations which can be addressed and further research can be done to improve the system.

Item Type:	Thesis (Masters)
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Cloud computing
Divisions:	School of Computing > Master of Science in Cloud Computing
Depositing User:	Tamara Malone
Date Deposited:	05 Dec 2022 13:12
Last Modified:	05 Dec 2022 13:12
URI:	https://norma.ncirl.ie/id/eprint/5965

Actions (login required)

View Item