Towards Olfactory Information Extraction from Text: A Case Study on Detecting Smell Experiences in Novels

TL;DR

This paper introduces a semi-supervised method for extracting smell experiences from English literature, revealing societal influences on smell perception and improving over keyword-based approaches.

Contribution

It presents the first semi-supervised approach for identifying smell experiences in text, enhancing extraction accuracy compared to baseline methods.

Findings

Significantly improved performance over keyword-based baseline

Effective identification of smell experiences in literary texts

Demonstrates societal factors influence smell descriptions

Abstract

Environmental factors determine the smells we perceive, but societal factors factors shape the importance, sentiment and biases we give to them. Descriptions of smells in text, or as we call them `smell experiences', offer a window into these factors, but they must first be identified. To the best of our knowledge, no tool exists to extract references to smell experiences from text. In this paper, we present two variations on a semi-supervised approach to identify smell experiences in English literature. The combined set of patterns from both implementations offer significantly better performance than a keyword-based baseline.