Article Per Year (5 Year)
p-Index From 2021 - 2026
0.408
P-Index
This Author published in this journals
All Journal Indonesian Journal of Artificial Intelligence and Data Mining Indonesian Journal of Electrical Engineering and Computer Science
Lawrence Muchemi
University of Nairobi
Computer Science & IT

Published : 3 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 3 Documents
Search

 1 
A Review on Machine Translation Approaches Benson Kituku; Lawrence Muchemi; Wanjiku Nganga
Indonesian Journal of Electrical Engineering and Computer Science Vol 1, No 1: January 2016
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijeecs.v1.i1.pp182-190

Abstract

There is need to fill the gap between two entities which use different languages  to communicate which can be done either via human translation or machine translation means. The world has over 7000 living natural languages, thus making it costly if human translation route is taken hence need for machine translation. There is need to know the available machine translation approaches and their requirement in order to decide which one suits for particular languages or not, hence the motivation for this survey. The survey provide overview and architectures of the three major techniques available namely: rule based translation, corpus based translation and hybrid based translation plus their subcategories available in each approach.
Evaluating the Bantu parametric grammar in grammatical framework using Swahili grammar Benson Kituku; Lawrence Muchemi
Indonesian Journal of Electrical Engineering and Computer Science Vol 30, No 2: May 2023
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijeecs.v30.i2.pp817-824

Abstract

Monologue grammar development for under-resourced languages is very slow and laborious (involves creating rules to generate the computational grammar to enable analysis and synthesis of the language(s) in question). However, the need for computational grammar continues to soar in this technology-driven economy for information synthesis and analysis. This paper aims to set up an experiment in the grammatical framework (GF), to evaluate the efficiency and effectiveness of the Bantu parameterized grammar to bootstrap a new grammar for Swahili. The goal is to investigate how this approach of bootstrapping grammar in a multilingual environment is effective and efficient in reducing the development effort. The bootstrapping approach uses the GF morphology-driven approach to develop portable and unique segments of Swahili grammar. The bootstrapped Swahili grammar resulted in a shareability of 100%, 71.11%, 68.75%, and 91.41% at category linearization, paradigms, parameters and syntax rules respectively. The portability was at 15.55%, 18.57%, and 8.59% at paradigms, parameters and syntax rules, respectively. Finally, this paper contributes in: first, provides an approach that leads to an effective and efficient method for developing and bootstrapping computational grammar for the under-resourced Bantu languages. Secondly, the research provided a Swahili grammar.
Hybrid Machine Learning Techniques for Comparative Opinion Mining Bernard Omoi Ondara; Stephen Waithaka; John Kandiri; Lawrence Muchemi
Indonesian Journal of Artificial Intelligence and Data Mining Vol 6, No 2 (2023): September 2023
Publisher : Universitas Islam Negeri Sultan Syarif Kasim Riau

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24014/ijaidm.v6i2.22644

Abstract

Comparative opinion mining has lately gained traction among individuals and businesses due to its growing range of applications in brand reputation monitoring and consumer decision making among others. Past research in sub-field of opinion mining have mostly explored single-entity opinion mining models and the mining of comparative sentences suing single classifiers. Most of these studies relied on a limited number of comparative opinion labels and datasets while applying the techniques in limited domains. Consequently, the reported performances of the techniques might not be optimal in some cases like working with big data. In this study, however, we developed four hybrid machine learning techniques, with which we performed multi-class based comparative opinion mining using three datasets from different domains.  From our results, the best-performing hybrid machine learning technique for comparative opinion mining using a multi-layer perceptron as the base estimator was the Multilayer Perceptron + Random Forest (MLP + RF). This technique had an average accuracy of 93.0% and an F1-score of 93.0%. These results show that our hybrid machine learning techniques could reliably be used for comparative opinion mining to support business needs like brand reputation monitoring.
Co-Authors Benson Kituku Bernard Omoi Ondara John Kandiri Stephen Waithaka Wanjiku Nganga