Optimization of Topic Mining and Sentiment Analysis for Telematics Workforce Development Strategies

Abstract

The dynamics of the development of the telematics business is strongly influenced by
the implementation of the industrial revolution 4.0. This study aims to optimize the topic
mining techniques and analysis sentiment in developing a telematics employment strategy
in Indonesia related to the challenges of the industrial revolution 4.0. Optimization of
topic mining and analysis sentiment reaches an average accuracy rate of 73%. Topic
mining is carried out on the ten highest rated news sites according to Alexa rank, using
the Latent Dirichlet Allocation (LDA) technique. Sentiment analysis is conducted on
social media twiter, using Support Vector Machine (SVM) technique. The results of
optimization of topic mining show that Indonesia's telematics employment strengthening
strategy can refer to topics related to the fields of Big Data, Artificial Intelligence and
Digital. This is reinforced by the growth of startups that are most in demand by investors
covering all three areas. Optimization of sentiment analysis shows that the community
responds positively to the conditions of telematics employment. This can also be
interpreted as public optimism about the prospects of the telematics business. This
condition is in line with the number of telematic micro small and medium enterprises
(MSMEs) which experienced very significant growth in 2017 which was 78.24%. The
growth of the telematics business is expected to increase with the strategy of
strengthening telematics competencies both in terms of hard skills and soft skills, starting
from basic education by referring to fields that are trending into world topics. Another
strategy is the need for very strong synergy, especially for the Ministry of Communication
and Information, the Ministry of Manpower and the Ministry of Industry, in formulating
policies related to strengthening the telematics business.