Karlova-Bourbonus, NataliNataliKarlova-Bourbonus2023-03-282019-04-262023-03-282018http://nbn-resolving.de/urn:nbn:de:hebis:26-opus-144470https://jlupub.ub.uni-giessen.de/handle/jlupub/15796http://dx.doi.org/10.22029/jlupub-15178The main purpose of news is to inform the reader about the current political, economic, and cultural events in the world. By that, the main requirements for the process of news produc-tion is an objective, uninvolved news reporting and an accurate, i.e. correct and consistent (contradiction-free) use of facts. A violation of the latter leads to the misinformation of the reader and, if detected, to a negative impact on the credibility and trustworthiness of the newspaper. The recognition of contradictions in a (news) text is a challenging task for a human as it presupposes concentrated reading and requires world knowledge and the ability to analyti-cally process the information obtained. Also, the age and mental capability of the reader plays an important role. Further, the task of contradiction recognition becomes even more difficult when dealing with contradictory facts occurring in texts that are separated by space and time. For this reason, the main aim of the present study was to propose a system for the automatic detection of contradictions occurring in news texts written in English.enIn Copyrightcontradictionmachine learningnatural language processingtextual entailmentcorpusddc:400