An Automated Toxicity Classification on Social Media Using LSTM and Word Embedding
Why is this work in the frame?
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Post-publication record
- Nature
- Retraction
- Reason
- Concerns/Issues about Peer Review;Investigation by Journal/Publisher;Objections by Author(s);
- Date
- 2/9/2023 0:00
- Flagged by OpenAlex?
- Yes
Source: Retraction Watch, joined by DOI. OpenAlex records retraction as is_retracted, a boolean over a state space with at least four values, so it cannot express an expression of concern, a correction or a reinstatement — it reports them as false, which reads as “fine”.
Abstract
The automated identification of toxicity in texts is a crucial area in text analysis since the social media world is replete with unfiltered content that ranges from mildly abusive to downright hateful. Researchers have found an unintended bias and unfairness caused by training datasets, which caused an inaccurate classification of toxic words in context. In this paper, several approaches for locating toxicity in texts are assessed and presented aiming to enhance the overall quality of text classification. General unsupervised methods were used depending on the state-of-art models and external embeddings to improve the accuracy while relieving bias and enhancing F1-score. Suggested approaches used a combination of long short-term memory (LSTM) deep learning model with Glove word embeddings and LSTM with word embeddings generated by the Bidirectional Encoder Representations from Transformers (BERT), respectively. These models were trained and tested on large secondary qualitative data containing a large number of comments classified as toxic or not. Results found that acceptable accuracy of 94% and an F1-score of 0.89 were achieved using LSTM with BERT word embeddings in the binary classification of comments (toxic and nontoxic). A combination of LSTM and BERT performed better than both LSTM unaccompanied and LSTM with Glove word embedding. This paper tries to solve the problem of classifying comments with high accuracy by pertaining models with larger corpora of text (high-quality word embedding) rather than the training data solely.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
The record
- Venue
- Computational Intelligence and Neuroscience
- Topic
- Hate Speech and Cyberbullying Detection
- Field
- Computer Science
- Canadian institutions
- École de Technologie SupérieureUniversité du Québec à Montréal
- Funders
- Taif University
- Keywords
- Computer scienceWord (group theory)Artificial intelligenceWord embeddingBinary classificationNatural language processingEmbeddingEncoderContext (archaeology)Social mediaMachine learningSpeech recognitionSupport vector machineWorld Wide WebLinguistics
- Has abstract in OpenAlex
- yes