Trend Detection and Popular Topics on Social-Media Using a clustering algorithm to find patterns and topics that are going viral on the Instagram platform

Abstract

This study aims to cluster Instagram posts based on hashtags and the number
of likes using the K-Means Clustering algorithm. The data used is data that
represents various popular topics on social media, such as travel, culinary,
fashion, and local coffee. The analysis process involves data preprocessing,
clustering algorithm implementation, and result evaluation to identify patterns
and trends among users. The results successfully grouped posts into three
main clusters, namely clusters with low engagement, clusters related to local
food and coffee, and clusters with high engagement on travel and fashion
topics. This clustering provides useful insights for marketers, content creators,
and researchers in understanding social media user behavior and designing
more effective marketing strategies. This research confirms the importance of
data analysis as a tool to uncover hidden patterns and support data-driven
decision-making.