Natural language processing for participatory corporate foresight: The participant input analyzer for identifying biases and fallacies

Corporate Foresight (CF) relies heavily on expert opinions collected through participatory processes, underpinning its results with a wide array of expertise. However, the efficient analysis and processing of comprehensive information shared by participants, which are often influenced by heuristics, biases, and fallacies, remains a challenge. Natural Language Processing (NLP) techniques offer an untapped potential for efficient, objective analysis of textual input, including identifying fallacies and biases. Yet, their application remains limited within foresight literature. This article explores the use of NLP to enhance CF. Specifically, it proposes application of NLP to help facilitators of participatory processes identify biases – e.g., desirability and undesirability – and common fallacies, like the assumptions of ceteris paribus environments and linear trends. We introduce a novel NLP tool, the ‘Participant Input Analyzer’, which by enabling analysis and visualization of sentiment, tense, and topic in textual statements, demonstrates the power of NLP in addressing those challenges. The use of NLP, as exhibited with proposed tool, has the potential to significantly advance the CF field. It can lead to more comprehensive and debiased analyses, contributing to better decision-making for corporates. This paper, therefore, serves as a steppingstone towards more rigorous, data-driven approaches in CF research and practice.