Home // WEB 2025, The Thirteenth International Conference on Building and Exploring Web Based Environments // View article

How to Interpret an Economic Index? Generating Reports with Topic Sentiment Analysis

Authors:
Kazuhiro Seki

Keywords: topic analysis; summarization; large language models; sentiment analysis

Abstract:
This paper presents our ongoing work on generating a monthly economic report to enhance the interpretability of a news-based business sentiment index. In our previous work, we developed a framework to provide a timely and quantitative measure of economic confidence. However, it lacked clarity on the underlying factors influencing the index values. To tackle the issue, we propose a topic-centered summarization approach which explains key events and trends during a specified period. Preliminary experiments show a promising result, producing a concise and comprehensive summary of the economic situation.

Pages: 9 to 10

Copyright: Copyright (c) IARIA, 2025

Publication date: March 9, 2025

Published in: conference

ISSN: 2308-4421

ISBN: 978-1-68558-243-2

Location: Lisbon, Portugal

Dates: from March 9, 2025 to March 13, 2025