Towards forecasting business prepaid mobile using neural network technology

• Main Authors: Abdul Rahman, Shuzlina, Wan Ishak, Wan Hussain, Tg. Osman @ Tg. Ramli, Tg. Halim
• Format: Conference or Workshop Item
• Language: English
• Published: 2004
• Subjects: QA76 Computer software
• Online Access: http://repo.uum.edu.my/3456/1/Shu1.pdf; http://repo.uum.edu.my/3456/
• Institution: Universiti Utara Malaysia

Prepaid mobile service has become necessity to the society and contributed success to many businesses. In order to be sustained in mobile telecommunication (Telco) industries, carriers need to plan their businesses as to ensure the increase use of their products and services. One of the strategles that can be applied in Telco industry is by forecasting the business trends using the appropriate technology that would assist the decision making process. This study explored the neural networks technology in analysing the historical customer requirements based on the teletraffic data that was produced from Telco peripherals. A Farecas Telco simulator was developed to experiment the capability of neural network technology in classifying the outputs. Several factors that contribute to the success of calls are determined. five types of output are categorized ranging from not very successful to very successful. The simulator managed to generalize with 98% of accuracy.