Document Management System Using Text Mining for Information Acquisition of International Construction

Abstract

Acquiring timely and proper information of host countries is a crucial element to lead successful and lucrative delivery of international construction projects. This information, however, commonly exists in forms of unstructured text data such as news articles and reports, which calls for the need of text mining. The aim of this research is to develop a prototype of construction document management system for global contract, which provides the user-needed information in a timely manner. The system named UNI (User Needed Information)-Tacit collects text data containing the recent information of the global construction market by using the web crawling algorithm, automatically allocates tags for each document with the most representative keywords based on Natural Language Processing, and eventually visualizes the results in the form of word clouds. The developed system, the survey validated its usefulness, is expected to contribute to collecting and organizing the latest issues on the global construction market, which provides better understanding of the target countries for decision makers.