News and Articles

Logical Fallacy Detector and Dataset

A new paper published in Feb 2022 (v1) - May 2022 (v2) “Logical Fallacy Detector” by Zhijing Jin and others, has a dataset “Logic” of annotated logical fallacy examples as an open-source appendix. This dataset is available on https://github.com/causalNLP/logical-fallacy . In case the original repo is not online - the Sep 2022 copy of the repository is available on https://github.com/tmakesense/logical-fallacy/tree/main/original-logical-fallacy-by-causalNLP .

This publication arxiv:2202.13758 has on the top of the annotation page

Appeal to Novelty Fallacy is from the Latin argumentum ad novitatem - claiming that some idea or product is better than previous ones just because it is new.

This fallacy is frequently used in marketing of new products, fashion, political advertising, and other areas. Some keywords used to describe and praise novelty are “the next new thing”, “pushing the envelope”, “new and improved”, “cutting-edge”.

How We Choose

We’ve selected the top most engaging news articles on Twitter from Australian news media. The selection criteria are based on logical fallacies statistics found in the retweets and comments.

Critical Concentration

“Senator Jacinta Nampijinpa Price has accused journalist Peter FitzSimons of being “very aggressive” during a passionate interview about the Indigenous Voice…”

The response to this tweet contains approximately 45% of comments that look a lot like fallacious reasoning of any type fallacy finder can recognise. Spread between reasoning type groups:

Intro

There are several logical fallacies detectors and finders available on the internet as an open source projects. Some of these are quite stale and some are not detectors at all. Here is the list of those available and a short description for each one with their status.

Android App

Logical Fallacy Detector App is available in the Android Play Store: Logical Fallacy Detector App

How We Select Top Tweets

We’ve selected the top most engaging news articles on Twitter from Australian news media. The selection criteria are based on logical fallacies statistics detected in the retweets and comments to original news tweet.

Critical Concentration

“Just another day in the office. Video coming soon…”

The response contains approximately 48% of comments that look a lot like fallacious reasoning of any type our detector can recognise. Spread between reasoning type groups: