Geometric lexical representative perspectives: The impact of threshold values through #LancsBox software
Year: 2022
Type: conference-paper
Abstract: This paper is an investigation of the impact of threshold values in geometric representations of lexical plots. The study uses MI-scores to test lexical representations in the Household Building Corpus (KPMR) data. MI-scores are a form of collocation of the assessment of lexical co-occurrence with values of > 3.0. However, changes in threshold values can provide an additional representative value to the nodes involved. The findings show that the threshold effect can contribute to lexical presence such as hapaxes as lexical density. Rational lexical expressions can be evaluated and viewed through the impact of its lexical wealth, in addition to examining the capabilities of #LancsBox 4.5 as a machine learning processer.
Source: AIP conference proceedings
DOI: 10.1063/5.0104817
Institution: Sultan Idris Education University
Language: English
Subfield: General Social Sciences
Field: Social Sciences
Domain: Social Sciences
Open Access status: hybrid
OpenAlex ID: W4312789136
