EURO 2024 Copenhagen
Abstract Submission

EURO-Online login

367. Applying Mathematical Programming to Visualize Text Summarization from Multiple Perspectives

Invited abstract in session MB-3: Optimization in Online Environments, stream Data Science Meets Optimization.

Monday, 10:30-12:00
Room: 1005 (building: 202)

Authors (first author is the speaker)

1. Li-Ching Ma
Department of Information Management, National United University

Abstract

The widespread adoption of the Internet and social networks has resulted in an explosion of data from diverse sources, often causing information overload. Consequently, a growing need for text summarization has emerged to assist users in quickly digesting information. Additionally, many practical scenarios require comparative opinions from multiple perspectives. However, most research tends to address issues from a single viewpoint.
The aim of this study is to propose a graphical text summarization approach that integrates multiple perspectives through the utilization of text mining and goal-programming optimization methods. From the idea of a data warehouse schema, this study constructs three-layer heterogeneous graphs to represent entities, nouns, adjectives, and their interrelationships. Star and Snowflake diagrams are constructed to show relationships between nouns and adjectives. Additionally, a constellation diagram is introduced to visualize the relationships among different entities.
The proposed approach offers several significant advantages. It provides a straightforward and easily understandable method to summarize a large amount of text content on a single graph. Users can, therefore, significantly reduce the time spent individually browsing numerous documents or reviews. Decision-makers can compare the strengths and weaknesses of different entities from an integrated graph or understand the differences in user opinions from various perspectives.

Keywords

Status: accepted

Back to the list of papers