Identifying Trendsetters in Online Social Networks – A Machine Learning Approach

Autor:	Martina Fricke, Freimut Bodendorf
Rok vydání:	2020
Předmět:	Data source Ask price Computer science business.industry Principle of maximum entropy Social media Product (category theory) Artificial intelligence Machine learning computer.software_genre business computer
Zdroj:	Advances in Intelligent Systems and Computing ISBN: 9783030510565 AHFE (8)
DOI:	10.1007/978-3-030-51057-2_1
Popis:	Especially in the fashion business consumers nowadays ask continuously for new styles and products, which forces companies to generate new ideas and innovative products faster than ever. Besides, consumers interact with each other on social media platforms and exchange their problems, needs, preferences, and ideas. With 95 Mio. daily postings on Instagram, it becomes obvious that such a platform is a huge data source containing important and valuable information regarding product requirements, customer tastes and needs, and upcoming trends. This paper presents a model to identify trendsetters based on their social media profiles and interactions on Instagram by using multiple machine learning classifiers. The model is trained with data of 665 user accounts, considering 59 features. Maximum Entropy Model performs the best with a F1-score of 66.67%.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::51fc5a8643cc4b61b380ff07d85fe985 https://doi.org/10.1007/978-3-030-51057-2_1 Zobrazit plný text záznamu