Social media sentiment analysis

• Main Author: Song, Siong Yew
• Other Authors: Long Cheng
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2023
• Subjects: Engineering::Computer science and engineering
• Online Access: https://hdl.handle.net/10356/166010
• Institution: Nanyang Technological University

Entering the online era, a vast majority of mankind now possesses electronic devices that have the capabilities to access the internet, giving people multitude of tools at their fingertips. It comes as no surprise that of which, some of the most popular tools available from the advent of the internet are social media platforms, with more than half of the world’s population active on one or more of these sites. The ability to connect to anyone globally at anytime, anywhere is massively convenient and is one of the major driving forces of globalization and technological revolution. In 2022, they hold 3 seats in the top 10 most visited sites on the web, namely Facebook, Twitter, and Instagram. The ordinary user spends on average up to 2½ hours daily, accounting for 75% of the time people spent on their phones (Deyan, 2022). This trend is further exacerbated by lockdowns imposed by many countries, causing more people to be increasing active on the webspace and has even broken into the older demographics. (Samet, 2020). With such a large readily accessible and wide-ranging audience, many companies are happily leveraging social media platforms to empower their business: community growth, international reach, customer feedback, networking, tracking competition and many other essential commercial activities. Over 90% of Fortune 500 companies are currently utilizing this golden opportunity to enrich their business (Porteous, 2021). Thus, social media posts for ad campaigns and brand promoting are commonplace nowadays and have shown to be a very effective marketing strategy. But most studies have been focused on the general effectiveness of social media as a marketing tool rather than for specific posts in their own niche markets. Furthermore, given the unstructured nature of social media interactions and data, this wealth of information is also not fully utilized by market services. Therefore, this project aims to create a tool that helps companies properly evaluate the approximate effectiveness of their marketing posts according to a set of indexes and possibly how to better improve future posts to attract more users.