Claim Missing Document
Check
Articles

Found 1 Documents
Search
Journal : JBIAM

Hyperlocal Business Patterns in Indonesia: A Hashtag-Based Machine Learning Study of Top Metropolitan Areas Koswara, Asep
Journal Business Inovation, Accounting and Management Vol 1 No 1 (2025): April 2025
Publisher : Lembaga Riset Penelitian dan PKM Aliansi Manajemen Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

This study explores how hyperlocal businesses in Indonesia's metropolitan areas construct their digital identities through hashtags on social media, specifically Instagram. The research aims to identify spatial and sectoral variations of digitally expressed business activity by analyzing the top 30 hashtags from five major cities: Jakarta, Bandung, Surabaya, Yogyakarta, and Medan. Data was collected via Tagsfinder.com and classified using a supervised machine learning model based on the frequency-inverse document frequency (TF-IDF) to distinguish business-related hashtags from non-business ones. The results show that out of 150 total hashtags, 42.7% were classified as business-related, with Bandung showing the highest proportion (56.6%) and Yogyakarta the lowest (20%). Sectorally, Bandung stands out in fashion and apparel, while Medan shows strong presence in the jewellery sector (e.g., #cincinnikah, #cincinmedan) and Surabaya with culinary sector. These findings affirm previous literature on hashtags as proxies for economic activity and address a significant research gap in the Global South context. In conclusion, social media hashtags serve not only as tools for promotion but also as spatial and cultural markers of urban business identity, providing new insights for digitally mapping hyperlocal urban economies in Indonesia.