情報技術およびソフトウェア工学ジャーナル
オープンアクセス

概要

Provenance Detection of Online News Article

Ruba Ali Alsuhaymi

At present, with the current wide spread of information on the social media, the recipient or the researcher needs more details about the received information or spread, including the provenance. With the current explosion of the news websites, there is a question of credibility of news articles on the internet. It is important to know whether the news is correct or not. This paper focuses on identifying the provenance of news articles. Also, trace the provenance of news articles often to see where did the first publication of such news appear. Is the news publication true (the credibility of the news), or is the news quoting from the provenance of the news on the news website or is plagiarism and redistributed on news websites on the Internet? In this paper, we will answer these questions through the design and implementation of two techniques Google Search API and Google Custom Search that will define the provenance of news articles through the technique Topic Detection and Tracking (TDT). Therefore, verifies the proposed technical quality in terms of performance metrics through several different experiments. Based on these experiments and tests it were discovered that the technique Google Search API is better performance than Google Custom Search in detecting the provenance of news articles. The Google Search API is the best technique, depending on the user satisfaction, the time it takes to view the results and the accuracy and validity. So, the result of the Google Search API is 90% while Google Custom Search 70%.