Seminars on “Text Mining Research in Business: Methods and Insights, and Brand Management using Text Mining”

Posted on January 9, 2019 by administrator

By: Prof. Joseph Johnson, Associate Professor of Marketing, Miami Business School, Florida & Visiting Fellow, IIM Kozhikode

Abstract of the Talks:

Text mining Research in Business:–Methods and Insights

The availability of large volumes of virtually free data such as social media posts and advances in computer algorithms have made it possible today to analyze unstructured text data to solve business problems arising in marketing, finance, accounting and supply chain management. That was a  two part seminar examined the main text mining methods used in business research and then showed the use of theose tools in a specific marketing context: brand management.

Brand Management using Text Mining

No brand is the market leader uniformly across geographic locations. Brands have strong and weak markets. Therefore, analysts need to track brand preferences by location. However, current methods of tracking geographic brand preferences (e.g., surveys, company sales, scanner panel data, Google Analytics, etc.) have limitations: limited sampling of brands, stores, and categories with time and region aggregation. They addressed these limitations through a spatial, nonparametric approach to track brand preferences at the micro-temporal and micro-geographic level using Twitter. Their approach inherited the intrinsic strengths of social media: micro-temporal, micro-geographic, faster than news media, higher frequency than sales, low cost, competitive monitoring, sentiment inclusion, and social network information. A unique feature of their proposed approach was the addition of temporal geographic influence of social connections based on the number of followers of each post. They estimated their model for carbonated soft drinks brands. They validated the model using Pepsi sales, an apparel fashion brand and the 2016 US presidential election (a domain where marketing is critical).
Their  study implied that continuous brand tracking using social media across geographic units like cities or districts provides actionable insights for marketing brands and political candidates.

***