IU Incubator

Fan Trend Tracker

The rapid proliferation of various digital platforms has resulted in an unprecedented volume of structured and unstructured data available through API’s concerning consumer attitudes, behaviors, preferences, and sentiments. As consequence, companies, organizations and institutions across almost all sectors face one similar problem: how to deal with the abundance of structured and unstructured data available online relevant to their sectors, how to integrate such heterogeneous data, and analyze them in a manner that allows for a clear understanding and actionable solutions. 

Therefore, the main goal of our incubator project is to develop a novel framework and create an AI-powered system to make sense of the vast amount of structured and unstructured consumer data available and transform it into clear and understandable insights and actionable recommendations. Importantly, it will track the evolution of sentiments, attitudes, trends, and consumer behavior over time while maintaining consumer privacy.

In specific, the central aim of the proposed incubator project is twofold:

(1) develop a novel advanced analytic and predictive framework and system using artificial intelligence to identify, crawl and process large-scale structured and unstructured data systematically. The system will focus on centralizing and converting diverse publicly available datasets on several API’s (e.g. engagement metrics, social media comments, behavioral meta-data) into actionable insights, enabling the identification, interpretation, and forecasting of trends in consumer attitudes, preferences, and behaviors (e.g. sentiment and topic analysis, predictive analysis, cross-consumer group comparisons).

(2) apply, test and refine the framework and smart system through a “case project” in the sports sector (e.g. professional German football club or competition, such as the Bundesliga), revealing relevant consumer (fans) insights to stakeholders in the sector, and academics in the field of consumer behavior. 

Specifically, the project seeks to address the following research questions:

• How do sentiment patterns related to specific entities (e.g., individuals, brands, products, or teams) evolve in response to external events, and can these patterns reliably predict future shifts in public opinion and consumption behaviors?

• What hidden or previously undetected connections exist between seemingly unrelated topics or events across different digital platforms, and how can these connections reveal deeper insights into underlying attitudes and consumption patterns?

• How can predictive analytics effectively categorize fan preferences and behaviors, and what implications do these preference clusters have for targeted communication strategies and consumer engagement?