Saved in:
Bibliographic Details
Main Author: Hu, Jinfan Frank
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2509.14238
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911181085605888
author Hu, Jinfan Frank
author_facet Hu, Jinfan Frank
contents Tokenization plays a critical role in processing agglutinative languages, where a single word can encode multiple morphemes carrying syntactic and semantic information. This study evaluates the impact of various tokenization strategies - word-level, character-level, n-gram, and Byte Pair Encoding (BPE) - on the quality of static word embeddings generated by Word2Vec for Turkish and Finnish. Using a 10,000-article Wikipedia corpus, we trained models under low-resource conditions and evaluated them on a Named Entity Recognition (NER) task. Despite the theoretical appeal of subword segmentation, word-level tokenization consistently outperformed all alternatives across all tokenization strategies tested. These findings suggest that in agglutinative, low-resource contexts, preserving boundaries via word-level tokenization may yield better embedding performance than complex statistical methods. This has practical implications for developing NLP pipelines for under-resourced languages where annotated data and computing power are limited.
format Preprint
id arxiv_https___arxiv_org_abs_2509_14238
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Tokenization Strategies for Low-Resource Agglutinative Languages in Word2Vec: Case Study on Turkish and Finnish
Hu, Jinfan Frank
Computation and Language
Tokenization plays a critical role in processing agglutinative languages, where a single word can encode multiple morphemes carrying syntactic and semantic information. This study evaluates the impact of various tokenization strategies - word-level, character-level, n-gram, and Byte Pair Encoding (BPE) - on the quality of static word embeddings generated by Word2Vec for Turkish and Finnish. Using a 10,000-article Wikipedia corpus, we trained models under low-resource conditions and evaluated them on a Named Entity Recognition (NER) task. Despite the theoretical appeal of subword segmentation, word-level tokenization consistently outperformed all alternatives across all tokenization strategies tested. These findings suggest that in agglutinative, low-resource contexts, preserving boundaries via word-level tokenization may yield better embedding performance than complex statistical methods. This has practical implications for developing NLP pipelines for under-resourced languages where annotated data and computing power are limited.
title Tokenization Strategies for Low-Resource Agglutinative Languages in Word2Vec: Case Study on Turkish and Finnish
topic Computation and Language
url https://arxiv.org/abs/2509.14238