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.
Logical Fallacy Bot
Available as Open Source on github: https://github.com/hovancik/logical-fallacies-bot
Consists of two parts: bot service subscribing and posting to reddit and a website containing logical fallacy definitions and examples
The main idea of this bot is to help to explain logical fallacies on Reddit, when invited to do so by some Reddit user who is using specific wording in their comments, for example when user types
explain base rate fallacy
Bot will reply to such comment with the fallacy description:
Base rate fallacy
If presented with related base rate information (i.e. generic, general information) and specific information (information only pertaining to a certain case), the mind tends to ignore the former and focus on the latter.
Example: Something can go wrong (premise). Therefore, something will go wrong (invalid conclusion).
Serves definitions and examples of
- Informal Fallacies: Appeal to the stone, Argument from ignorance, Argument from repetition, Argument from silence, Argument to moderation, Begging the Question, Burden of proof, Circular cause and consequence, Circular Reasoning, Continuum fallacy
- Formal Fallacies: Affirmative conclusion from a negative premise, Affirming a disjunct, Affirming the consequent and others
Implemented in python, last activity on the project: 2017
Developed and used by www.logical-fallacy.com, this apart from internal services currently consists of publicly available:
- Twitter account @Makesensenews1 posting most triggering piece of news from Australian news media daily, based on number of fallacies identiied in the replies.
- Alexa Make Sense Skill - analyzes voice input to find logical fallacies in the speech.
- Logical Fallacy Checker Tweet Bot @falldetector1
In details described in Online Logical Fallacy Detectors
Logical Fallacy Detection
This paper by Zhijing Jin and others is published in https://arxiv.org/abs/2202.13758, [v2] Tue, 24 May 2022.
One of the most interesting parts of this paper is the Logical Fallacy Dataset of scraped from the internet annotated logical fallacy examples, available in causalNLP’s repo for this paper is: https://github.com/causalNLP/logical-fallacy The copy of this repo of date: 6 September 2022 is stored in: https://github.com/tmakesense/logical-fallacy/tree/main/original-logical-fallacy-by-causalNLP
This dataset could be a very good starting point for the experiments with AI. There is A LOT of cleansing and fixing required in the dataset to improve detection rate. Please see what needs to be fixed in this dataset to increase the detection rate: Logical Fallacy Dataset - Review. The fixed version of the dataset is published on github: Improved Logical Fallacies Dataset
Python, last activity: April 2022.
Logical Fallacy Classifier
Python notebook using mini BERT - published on github https://github.com/mlopezgez/fallacy-classifier . It is also using the same open source fallacy dataset from Logical Fallacy Detection project described above. Not cleansed though.
Ref: Victor Sanh. 2019. “Smaller, faster, cheaper, lighter: Introducing DistilBERT, a distilled version of BERT” Medium.: https://medium.com/huggingface/distilbert-8cf3380435b5
This project might be even better start for coding automatic logical fallacy finder.
Python notebook, last activity: August 2022.
There was no mobile apps available on android store that are able to detect logical fallacies, just two pocket reference type applications. We decided to add them to this review anyway so when you see them on the app/play store you will not have to spend your time searching for the button “ckeck”, “analyze” or “detect”.
Android app contains several pages - list of logical fallacies with their descriptions and examples. Identification of the fallacies is left to the reader.
This is also a reference android, not the fallacy checker application with description pages, very basic interface.
Last updated: 2017
If we missed an application or project detecting logicall fallacies please let us know, our email is on the bottom of this page.