Mining social media data from sparse text: an application to diplomacy

Chua, Cecil; Li, Xiaolin (David); Kaul, Mala; Storey, Veda C.

Mining social media data from sparse text: an application to diplomacy

Files

Paper 7.pdf(618.86 KB)

Published Version

Date

2016-05

Authors

Chua, Cecil

Li, Xiaolin (David)

Kaul, Mala

Storey, Veda C.

Publisher

DESRIST 2016

Abstract

Publicly available data in social media provides a wealth of unstructured data for applications, such as sentiment analysis and location-based services. This research analyzes a specific application of diplomats, who seek to understand the people with whom they must negotiate. Social media data about a negotiating partner can, potentially, be used to build a profile of that partner. However, such data is difficult to mine effectively because it has sparse text with high dimensionality. This research uses a design science approach to develop a method for extracting critical information from sparse text. The method mines sparse text from publically available Facebook data to extract patterns from individual communications. The method is applied to Facebook posts of a political figure to identify meaningful categories of information for insightful inferences. Preliminary evaluation shows support for the method.

Keywords

Design application , Diplomat , Natural language processing , Sentence clustering , Sparse text mining , Semantic chain

Citation

Chua, C., Li, X. , Kaul, M., Storey, V. C. 2016. Mining social media data from sparse text: an application to diplomacy. In: Parsons, J., Tuunanen, T., Venable, J. R., Helfert, M., Donnellan, B., & Kenneally, J. (eds.) Breakthroughs and Emerging Insights from Ongoing Design Science Projects: Research-in-progress papers and poster presentations from the 11th International Conference on Design Science Research in Information Systems and Technology (DESRIST) 2016. St. John, Canada, 23-25 May. pp. 51-58