Copyright © 2018 Authors. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted
use, distribution, and reproduction in any medium, provided the original work is properly cited.
International Journal of Engineering & Technology, 7 (3.20) (2018) 381-384
International Journal of Engineering & Technology
Website: www.sciencepubco.com/index.php/IJET
Research paper
Learning Vector Quantization Implementation to Predict the
Provision of Assistance for Indonesian Telematics Services
SMES
Eneng Tita Tosida, Fajar Delli Wihartiko and Indra Lumesa
Computer Science Department, Pakuan University, Jl. Pakuan PO Box 452, Bogor, Indonesia
*Corresponding Author Email: enengtitatosida@unpak.ac.id
Abstract
Implementation of Learning Vector Quantization (LVQ) Algorithm for classification of Indonesia telematics service is designed and
created as a classification system to support the decision of grant aid for Small Medium Enterprises (SMEs). Based on the test results, the
LVQ algorithm has the best accuracy (93.11%) when compared with ID3 algorithm (64%) and C45 (62%) for telematics data of National
Census of Economic (Susenas 2006). The data is still valid and relevant for use in this research because in Indonesia census data is done
every 10 years and there is no update of data until now. LVQ implementation results are applied to a web-based decision support system
to predict the provision of assistance for Indonesian telematics services SMEs. Unlike the C45 and ID3 algorithms, the LVQ algorithm
generates the weight of a neural network where it difficult to know which attributes are most influential for decision making. But in this
study LVQ able to show good performance through the analysis of the relevance of existing conditions by comparing it with the weight
value produced by the model that are implemented in a web-based decision support system
Keyword: Decision Support System, Learning Vector Quantization, Small Medium Enterprises
1. Introduction
Indonesia has a tremendous opportunity to win the competition in
the Asian Economic Community (AEC). This is supported by the
number of Small and Medium Enterprises (SMEs) in the field of
telematics that can survive even in the global economic crisis.
SME data on telematics is available at the Central Bureau of
Statistics (BPS), in the format of the National Economic Census
(susenas 2006). But this data has not been used optimally by the
government to support the decision to empower SMEs. In fact,
susenas data related to SME telematics has complete variables and
has conformity with the process of providing assistance conducted
by the Ministry of Cooperatives and SMEs Indonesia [11]. In
addition, the 2006 susenas data can be used in this research
because the data is still in accordance with the current condition
and there is no update data from BPS until 2017.The telematics
industry (Information and Communication Technology - ICT)
itself is one of the priority industries that will and will be
developed by the Government through the National Industrial
Development Policy. The telematics industry itself is currently a
rapidly growing industry in the world with 6.9% growth per year.
In 2004 the world's ICT market reached US $ 533 billion, while
Asian ICT market was US $ 42 billion with 23% growth per year.
In Indonesia alone, the market was recorded at only US $ 1.3
billion with growth in 2004 and 2005 of 9.8% and 22.1%,
respectively. Of that amount, it is estimated that US $ 0.5 billion
to US $ 0.75 billion is absorbed by the banking sector. The
telematics industry consists of a group of goods and services,
including the Device Industry, Infrastructure / Networks (access,
nodes, transport & support) and software including applications
(content). For developing countries software and services
generally have greater opportunities because they do not require
large investments in research and production support equipment.
This is mainly due to more software based on knowledgeable
workforce [12].
Based on previous research by [11] the system has not built a
system of determining the provision of telematics services
assistance limited to SMEs data visualization of each region. On
the feasibility of assistance for Indonesian telematics services
Micro Small Medium Enterprises (MSMEs) involve complex
criteria consisting of 21 criteria [12]. The relationship between the
criteria for the feasibility of the aid is non-linear, Indonesian
telematics services SMEs therefore it can be approximated by
artificial neural network method. The assistance scheme for
telematics SMEs has a character that is almost the same as the
credit scheme for others SMEs in bank. Therefore research that
can be used as a reference among others related to credit scoring
mechanism for business owners including SMEs ([5]; [9]).
Research on feasibility of assistance for SMEs through credit risk
in Supply Chain Finance (SCF) has been done by [14], which
predict SMEs credit by six machine learning methods. There are
one individual machine learning (i.e. decision tree), three
ensemble machine learning methods (i.e. bagging, boosting and
multi boosting) and two integrated ensemble machine learning
machines methods (i.e. RS-boosting and multi boosting). [13]
applies credit scoring using non parametric method for Peruvian
microfinance industry. The model used is neural network multi
layer perceptron (MLP) which compared its performance with
linear discriminant analysis (LDA) approach, quadratic
discriminant analysis (QDA) and logistic regression (LR). MLP
performance shows a higher degree of accuracy than the other
three approaches. Therefore in this research will be done