Microaggressions describe subtle often offensive comments or actions made by one individual to another. Typically, such comments or actions are made subconsciously with the offender potentially unaware of the impacts on the recipient. Currently, machine learning methods for racial microaggression detection are sparse with no, one, comparable approach to the one we propose further on. Automated detection in this work describes the method of finding microaggressions through the use of machine-learning algorithms. Efforts have been made for the detection of hate speech and harassment; providing us with a rather humble place to begin, as we explore further, such methods are proven ineffective. We propose a step forward in solving this problem with the demonstration of an automated racial microaggression detection method. Whilst racial hatespeech detection method provides us with an idea as to where we can start, we find further on, that microaggressions and hatespeech use very different features to portray their sentiment. This work aims to provide a technical review which explores the understanding of the automated racial microaggression detection, outlining the definitions of microaggressions currently described in the literature, with the presentation and assessment (through precision, callback and F-measure) of a promising approach in regards to racial microaggression detection. We also intend to analyse a case study in which we detect the presence of racial microaggressions within new reports with the intention of laying a foundation for the potential security-related applications of this work. Our further works begin to discuss this notion in a higher level of detail.
|Name||2020 IEEE Symposium Series on Computational Intelligence, SSCI 2020|
|Conference||2020 IEEE Symposium Series on Computational Intelligence, SSCI 2020|
|Period||1/12/20 → 4/12/20|
- Hate Speech
- Human Modeling
- Online Content Moderation
- Sentiment Analysis