Look Ma, Only 400 Samples! Revisiting the Effectiveness of Automatic N-Gram Rule Generation for Spelling Normalization in Filipino

TL;DR

This paper demonstrates that a simple N-Gram and Damerau-Levenshtein distance model with automatic rule extraction can effectively perform Filipino spelling normalization with limited data, outperforming complex deep learning methods.

Contribution

It introduces a data-efficient, interpretable spelling correction model for Filipino that requires minimal compute and training time, challenging the dominance of deep learning in low-data scenarios.

Findings

Achieves high accuracy with only 300 training samples

Outperforms deep learning approaches in accuracy and edit distance

Requires minimal compute and training time

Abstract

With 84.75 million Filipinos online, the ability for models to process online text is crucial for developing Filipino NLP applications. To this end, spelling correction is a crucial preprocessing step for downstream processing. However, the lack of data prevents the use of language models for this task. In this paper, we propose an N-Gram + Damerau Levenshtein distance model with automatic rule extraction. We train the model on 300 samples, and show that despite limited training data, it achieves good performance and outperforms other deep learning approaches in terms of accuracy and edit distance. Moreover, the model (1) requires little compute power, (2) trains in little time, thus allowing for retraining, and (3) is easily interpretable, allowing for direct troubleshooting, highlighting the success of traditional approaches over more complex deep learning models in settings where data is unavailable.