Published on

Introduction to Sentiment Scope

Authors

Aims

Over the next few months, I will be sharing my research concerning the development of a method to rapidly analyse anti-immigrant sentiments with Natural Language Processing (NLP) techniques. The goal of this research is to develop an efficient and reliable method for quantifying intergroup disgust intensity online. Unlike an in-depth qualitative analysis, a quantitative analysis of disgust will be in less danger of falling prey to ideological and political biases, and, just as importantly, be less likely to be dismissed as biased towards a political agenda. This methods research will be the foundation for further study into whether such online disgust narratives are predictive of grassroots intergroup conflict.

The focus of the study will be disgust narratives levied online by local ingroups towards outgroups of immigrants and the descendents of immigrants. This is due to the pressing importance of native/immigrant tension in our times, its international and widespread prevalence during the European migrant crisis and Russia/Ukraine war, and the resultant ease of data collection regarding intragroup narratives and intergroup conflict in the case of local and immigrant groups. If further study demonstrates a predictive relationship between this initial project's measurement of disgust and conflict between locals and immigrant groups, then work will be expanded to measuring grassroots disgust narratives to predict other areas of intergroup conflict, such as in the case of interethnic, interclass, or interreligious tensions.

If all three projects prove successful, it will result in a cost-effective tool which, when used in tandem with existing methods, will help NGOs, government bodies, and other stakeholders to better predict and manage grassroots conflict.

Post schedule

The following posts of this blog will outline my methodology, which will proceed after this introduction as follows:

  1. Introduction to Sentiment Scope

  2. Disgust, Morality, and Intergroup Relations

  3. Quantifying Disgust

  4. The Categorical Model of Ingroup Disgust towards Outgroups

  5. Collocation manual selection

  6. Limitations Necessitating the Training of an LLM

  7. LLM Specifications and Development Plan

  8. Targets of Study, Dataset Selection, and Data Collection

  9. Final Considerations and Limitations

Following these, posts will be made reviewing relevant literature on disgust, dehumanisation, and conflict. The Categorical Model will be explored in detail, with each category receiving a post examining and evaluating it in reference to literature. Data collection and analysis will then begin. Updates will be published concerning the progress of each study. Finally, studies will be published in full, containing data, results, and discussion.

Welcome to Sentiment Scope.