Analysis of Mobile Service Providers Performance Using Naive Bayes Data Mining Technique

Abstract

Recently, the mobile service providers have been growing rapidly in Malaysia. In this paper, we propose analytical method to find best telecommunication provider by visualizing their performance among telecommunication service providers in Malaysia, i.e. TM Berhad, Celcom, Maxis, U-Mobile, etc. This paperuses data mining technique to evaluate the performanceof telecommunication service providers using their customers feedback from Twitter Inc. It demonstrates on how the system could process and then interpret the big data into a simple graph or visualization format. In addition, build a computerized tool and recommend data analytic model based on the collected result. From prepping the data for pre-processing until conducting analysis, this project is focusing on the process of data science itself where Cross Industry Standard Process for Data Mining (CRISP-DM) methodology will be used as a reference. The analysis was developed by using R language and R Studio packages. From the result, it shows that Telco 4 is the best as it received highest positive scores from the tweet data. In contrast, Telco 3 should improve their performance as having less positive feedback from their customers via tweet data. This project bring insights of how the telecommunication industries can analyze tweet data from their customers. Malaysia telecommunication industry will get the benefit by improving their customer satisfaction and business growth. Besides, it will give the awareness to the telecommunication user of updated review from other users.

A data mining approach for desire and intention to participate in virtual communities

“…A Bayesian network B over a set of variables U is a network structure Bs, which is directed acyclic graph (DAG) over U and set of probability tables Bp={p(u|pa(u))|u Є U} where pa(u) is the set of parents of u in Bs. A Bayesian network represents probability distributions [23,24].…”

Utilization of Machine Learning in Supporting Occupational Safety and Health Decisions in Hospital Workplace

“…The nodes for G are labeled with the elements of X and the arcs of G represent the probabilistically conditional dependence relationship between nodes. Each factor is associated with a conditional probabilistic table that defines the probability of each state for that factor [10,44] [39,45]. An MLP is a class of feedforward ANN and one of the most widely used.…”

Sentimental classification analysis of polarity multi-view textual data using data mining techniques

“…The J48 classification algorithm is based on decision tree. Naïve bayes classification algorithm is based on probability [28]. Successively, they extend right from the simple distance measure to other path of probabilistic distance measures to search the similarity criteria among a set of documents and then classify them directly based on their particular category.…”