Skip navigation
st. Mary's University Institutional Repository St. Mary's University Institutional Repository

Please use this identifier to cite or link to this item:
Authors: Alibo, Yared
Keywords: Data Mining, Knowledge discovery
FMS, QoS, Classification, Hybrid, CRISP-DM
Issue Date: Nov-2018
Publisher: st.mary's University
Abstract: Access to telecommunication service is critical to the development of all aspects of a nation economy including manufacturing, banking, education, agriculture and government. However, this telecommunication services are not free from problems. Mobile call drop is the main problems of all telecom operators. In telecommunication, mobile call drop is a situation where calls have been cut off before the speaking parties had finished their conversation tone and before one of them had hung up. This research therefore aims to design a predictive model that can determine mobile call drops from ethio telecom mobile network data. To overcome the drawback of simple statistical method, we proposed data mining techniques, methods and methodologies and used in this research. We selected around 20,000 records of one year and six months collection of Fault Management data. After eliminating irrelevant and unnecessary data, a total of 16996 datasets with 8 attributes are used for the purpose of conducting this study. Data preprocessing was done to clean the datasets. After data preprocessing, the collected data has been prepared in a format suitable for the DM tasks. The study was conducted using WEKA software version 3.8.1 and four classification techniques; namely, J48 and Random forest algorithm from decision tree as well as PART and JRIP algorithm from rule induction are used. As a result, J48 decision tree algorithm with 10-fold cross validation registered better performance and processing speed of 95.43% and 0.06 sec respectively. The algorithm also used 8 attributes for this experimentation of the research. Unavailability of related works on telecommunication mobile call drop area was one of the major challenges during the study. Another challenge includes the FMS mobile network server can hold only two years data since the data is huge due to this, we can’t get the data before two years. Finally, we recommend ethio telecom to apply data mining techniques on mobile network data for identifying the reason for call drops.
URI: .
Appears in Collections:Master of computer science

Files in This Item:
File Description SizeFormat 
Yared Alibo Ayiza.pdf1.5 MBAdobe PDFView/Open
Show full item record

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.