More information
Description
Athens Technology Center (ATC) is an Information Technology Company offering solutions and services in specific business sectors, including media, banking and retail, utilities and the public sector.
The exposure of citizens to large scale disinformation, including misleading or outright false information, has been acknowledged as a major challenge for Europe. The spread of disinformation is also facilitated by the increasing trend of algorithmic-driven news distribution channels, such as news aggregators and social media.
Their R&D-project UNBIASED aims to combat disinformation online and provide automated tools for fact-checking and identification of reporting bias. The tools developed will be integrated into a news aggregator that will incorporate a large number of different news sources with functionalities that will contribute to combating disinformation and the identification of fake news. Specifically, UNBIASED will offer tools for:
- real-time content organization and classification
- analyzing and estimating the various forms of reporting bias
- discovery of debated topics and polarity of news sources - provision of external evidence for fact checking
Athena Research and Innovation Center is one out of two project partners working on UNBIASED. They conducts research in Information, Communication, and Computational Science and Technology, including all areas of data science. In this project, Athena Research and Innovation Center participates with the Institute for Speech and Language Processing - ILSP. ILSP will develop/adapt the modules for real-time data news collection from different news sources, discovery of debated topics, sentiment analysis and polarity estimation, as well as the provision and analysis of external evidence from authoritative sources.
The project will result in 4 new innovative ICT solutions, 1 IPR and 4 jobs created. Also, increased revenues and competitiveness for the entities in the beneficiary country.
Summary of project results
The project aimed to address the well-known and recognised issue of the lately increasing exposure of citizens to large scale disinformation, including misleading or outright false information. Disinformation, also known as ‘fake-news’, erodes trust in institutions and mass media, impairs freedom of expression, manipulates public opinion and detracts attention from science and empirical evidence. The spread of disinformation is also facilitated by the increasing trend of algorithmic-driven news distribution channels, such as news aggregators and social media. Such channels, preferred by younger news readers, have separated the role of content editors and curators of news distribution and weakened the role of trusted editors as quality intermediaries.
Even though some methods for the detection of fake news exist, their efficiency in combating the actual phenomenon of disinformation is questionable. The project worked in this direction by providing automated web tools for conveying scientific evidence on debated topics, and estimating the polarity of news sources, as an indication of the “media bias” of a news source. The project also increased the automatization of fact-checking on debated topics. Fact-checkers today are mostly human-operated, which generally results in an intellectually demanding, laborious and time-consuming process. There is, therefore, an increasing demand for automated fact checking, which can keep up with the rate at which false claims or hoaxes are produced on news sites and social media.
The project developed software components for real-time collection and classification of factual information, particularly news, but also statements and opinions contained in user generated content and propagated in social media. The tool provides an analysis and assessment of the various forms of reporting bias, e.g. the publication or nonpublication of news items, the selective reporting of items, the rapid or delayed publication of a news item with respect to its first publication, the multiple or singular publication of a news item, etc. The tools assist the user in finding out debated topics (existence of different opinions on a certain topic), and provide a measure of the polarity of news sources. They also assist in fact-checking, by providing external evidence from surveys, authoritative sources, and scientific papers that can either support or refute an expressed opinion.
The tools are showcased in a news aggregator, which aggregates content using either an online crawler, or feed readers like RSS, Atom and other formats. It incorporates a large number of different news sources, such as online news agencies, newspapers, blogs, etc., with functionalities for:
- content organization and classification, according to author, topic, location, etc.
- analyzing and estimating the various forms of reporting bias, e.g. the publication or non-publication of news items, the selective reporting of items, the immediate and fast spread of a news item with respect to its first publication, the publication frequency of a news item, etc.
- discovery of debated topics and polarity of news sources
- provision of external evidence from surveys, authoritative sources, and scientific papers for fact checking
The project output was the development of 1 news aggregating and fact checking platform with 3 separate add-on tools (bias reporting, discovery of debated topics and polarity, provision of external evidence).
The project will contribute to an increased turnover and net operating profit for the Project Promoter, as well as in the creation of 4 new permanent full-time jobs in the company, which are planned to be retained even after the project is completed.