WEKA-Based: Key Features and Classifier for French of Five Countries

TL;DR

This paper presents a dialect recognition system for French across five regions, utilizing a corpus and machine learning tools to distinguish regional variations based on thematic content.

Contribution

It introduces a French dialect classification approach using a new regional corpus and machine learning with WEKA, tailored for regional dialect identification.

Findings

Effective differentiation of regional French dialects achieved

Utilized WEKA classifiers with thematic corpus for dialect recognition

Demonstrated feasibility of machine learning in dialect classification

Abstract

This paper describes a French dialect recognition system that will appropriately distinguish between different regional French dialects. A corpus of five regions - Monaco, French-speaking, Belgium, French-speaking Switzerland, French-speaking Canada and France, which is targeted forconstruction by the Sketch Engine. The content of the corpus is related to the four themes of eating, drinking, sleeping and living, which are closely linked to popular life. The experimental results were obtained through the processing of a python coded pre-processor and Waikato Environment for Knowledge Analysis (WEKA) data analytic tool which contains many filters and classifiers for machine learning.