The term “fake news” has been popularized by Donald Trump, the President of the United States of America, since 2016. Since then studies on “fake news” are booming as there are more than 7,000 academic research articles published in 2016. Yet, a more astonishing fact is that more than 44% of people treat Facebook and the like social media as the ONLY news source. In USA, there are more than 67% people read social media and 20% or more read them very frequently. The most surprising study shows that fake news are more popular than true news, among the top 20 news articles.

The first step to combat “fake news”, obviously is to scrutinize the news itself, for its data quality such as data integrity, authentication, authorization, confidentiality, and privacy etc., as is done by CDO (Chief Data Office). An easiest way to identify “fake news” is to determine the news author, publication source, date, and references. Unfortunately, in social media, such as Line, PTT, and WeChat, those features are almost always missing and are extremely difficult to be identified “automatically”.

TICC (Taiwan Information Care Center) project takes a completely different but maybe the most practical approach. It creates a database such that: it contains the most comprehensive proven fake news, and to fake news checking, with the fastest response, the most accurate, the least un-relevant, and the most timely fake news search results.

TICC will be the best tool for researches on text-based data-mining such as frequent patterns, clustering, classification, and data analytics. TICC is also a key breakthrough for researches with Natural Language Processing (NPL), Artificial Intelligence, Machine Learning, as well as applications to text-intensive, political, sociological, and economical sciences, and law, accounting and liberal arts.

 

 

作者

陳博修博士，美國馬利蘭大學電腦系碩士、博士及台灣大學電機系學士。曾任職於美國AT&T貝爾實驗室高級研究員及美國休斯網路公司遠東區業務主管，在台灣曾任職中國時報總管理處，也是時報資訊公司創始人之一。目前擔任順通資訊公司總經理、大同大學資訊經營研究所兼任副教授、科學園區數家上市公司波若威等策略性發展顧問。專精於分散式系統、軟體工程、多媒體通信、中文全文檢索等領域。

陳博士發表過許多IEEE/ACM論文，也在工商時報、自由時報等發表文章關心時政。陳博士擁有多項專利，涵蓋網路電話、高速中文全文檢索、分散式資料庫等等領域。目前研究領域包括區塊鍊及加密貨幣、自動化假新聞檢測及假新聞資料庫檢索系統、5G/資料中心、光纖網路等等。