A Computational Analysis of Lexical and Form-Level Variation between the Lutenga and Lupakoyo Varieties of Lusoga

Differences between closely related language varieties are not always apparent from general measures of similarity, particularly when they involve alternative words or recurring changes in written forms. This study examined lexical and written-form variation between Lutenga and Lupakoyo, two regional varieties of Lusoga spoken in eastern Uganda. Using 14,597 aligned English–Lutenga–Lupakoyo sentence triplets, the analysis combined sequence-based word alignment, character-similarity analysis, character-edit analysis, contextual screening, and bidirectional correspondence-consistency measures. It identified 5,644 distinct candidate word-form correspondences, of which 422 occurred at least five times. Among these recurrent pairs, 41 had character-similarity scores below 0.40. Examination of the corresponding translation contexts suggested that eight of these pairs could represent alternative lexical choices, although their linguistic status was not independently established. Most extracted correspondences (75.6%) involved closely resembling written forms, although this resemblance does not necessarily imply a shared linguistic structure. The analysis also identified 52 character-edit families. Insertion of o was the most frequent, occurring 2,699 times across ten distinct word pairs, whereas deletion of g occurred 526 times across 31 pairs. This difference illustrates how a transformation may occur frequently without being widely distributed across word forms. Of the 422 recurrent correspondences, 369 (87.4%) were mutual top counterparts, while 266 (63.0%) achieved at least 80% bidirectional consistency.These measures indicate recurring and relatively consistent associations within the extracted inventory, but do not independently establish that the correspondences represent systematic linguistic differences between the two varieties. The study provides a corpus-based account of these associations, offering a basis for investigating lexical differentiation and written-form variation, and for identifying patterns that warrant further linguistic validation.

Authors

Institutions

Publication Details

Journal
Researchers Journal of Science and Technology
Published
2026-11-21
DOI
https://doi.org/10.83080/rejost.vol6no10.366
Primary Topic
Natural Language Processing Techniques
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

A Computational Analysis of Lexical and Form-Level Variation between the Lutenga and Lupakoyo Varieties of Lusoga

Adeleke Raheem Ajiboye, Azizi Wasike, Ismail Kamukama, Muriithi Irene Wanjiru et al.
Researchers Journal of Science and Technology
Natural Language Processing Techniques
article

A Computational Analysis of Lexical and Form-Level Variation between the Lutenga and Lupakoyo Varieties of Lusoga

Adeleke Raheem Ajiboye, Azizi Wasike, Ismail Kamukama, Muriithi Irene Wanjiru, Awadallah M. Ahmed
article en

Abstract

Differences between closely related language varieties are not always apparent from general measures of similarity, particularly when they involve alternative words or recurring changes in written forms. This study examined lexical and written-form variation between Lutenga and Lupakoyo, two regional varieties of Lusoga spoken in eastern Uganda. Using 14,597 aligned English–Lutenga–Lupakoyo sentence triplets, the analysis combined sequence-based word alignment, character-similarity analysis, character-edit analysis, contextual screening, and bidirectional correspondence-consistency measures. It identified 5,644 distinct candidate word-form correspondences, of which 422 occurred at least five times. Among these recurrent pairs, 41 had character-similarity scores below 0.40. Examination of the corresponding translation contexts suggested that eight of these pairs could represent alternative lexical choices, although their linguistic status was not independently established. Most extracted correspondences (75.6%) involved closely resembling written forms, although this resemblance does not necessarily imply a shared linguistic structure. The analysis also identified 52 character-edit families. Insertion of o was the most frequent, occurring 2,699 times across ten distinct word pairs, whereas deletion of g occurred 526 times across 31 pairs. This difference illustrates how a transformation may occur frequently without being widely distributed across word forms. Of the 422 recurrent correspondences, 369 (87.4%) were mutual top counterparts, while 266 (63.0%) achieved at least 80% bidirectional consistency.These measures indicate recurring and relatively consistent associations within the extracted inventory, but do not independently establish that the correspondences represent systematic linguistic differences between the two varieties. The study provides a corpus-based account of these associations, offering a basis for investigating lexical differentiation and written-form variation, and for identifying patterns that warrant further linguistic validation.

Researchers Journal of Science and Technology
Kampala International University (UG), Muni University (UG), Kyambogo University (UG)
Natural Language Processing Techniques
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon