Worku Abebe Degife

Work place: Faculty of Informatics, University of Gondar, Gondar, Ethiopia

E-mail: workuabebeis@gmail.com

Website:

Research Interests: Computational Learning Theory, Network Security, Data Mining, Database Management System, Data Structures and Algorithms

Biography

Worku A. Degife (MSc) lecturer at University of Gondar and also currently he studying his PhD in Taiwan at National Taiwan University of science and Technology, in major of Information Management. Mr. Worku A. received his MSc. in Information Technology from University of Gondar, Gondar, Ethiopia in 2017. His main research fields are data mining, Neural network, machine learning, data mining and knowledge based system, big data management and security, reinforcement learning and AI.

Author Articles
Efficient Predictive Model for Determining Critical Factors Affecting Commodity Price: The Case of Coffee in Ethiopian Commodity Exchange (ECX)

By Worku Abebe Degife Dr.ing. Abiot Sinamo

DOI: https://doi.org/10.5815/ijieeb.2019.06.05, Pub. Date: 8 Nov. 2019

In this paper, we have focused on the data mining technique on market data to establish meaningful relationships or patterns to determine the determinate critical factors of commodity price. The data is taken from Ethiopia commodity exchange and 18141 data sets were used. The dataset contains all main information. The hybrid methodology is followed to explore the application of data mining on the market dataset. Data cleaning and data transformation were used for preprocessing the data. WEKA 3.8.1 data mining tool, classification algorithms are applied as a means to address the research problem. The classification task was made using J48 decision tree classification algorithms, and different experimentations were conducted. The experiments have been done using pruning and unpruning for all attributes. The developed models were evaluated using the standard metrics of accuracy, ROC area. The most effective model to determine the determinate critical factors for the commodity has an accuracy of 88.35% and this result is a good experiment result.
The output of this study is helpful to support decision-making activities in the area of the Ethiopia Commodity Exchange. The study support commodity suppliers to take care of the determinant factors work towards maintaining quality. Ethiopia Commodity Exchange (ECX), as the main facilitator of commodity exchanges, can also use the model for setting price ranges and regulations.

[...] Read more.
Other Articles