PREDICTING THE EFFECTIVENESS OF WEB INFORMATION SYSTEMS USING NEURAL NETWORKS MODELING: FRAMEWORK & EMPIRICAL TESTING


PREDICTING THE EFFECTIVENESS OF WEB INFORMATION SYSTEMS USING NEURAL NETWORKS MODELING: FRAMEWORK & EMPIRICAL TESTING

Dr. Kamal Mohammed Alhendawi
Assistant Professor in Management, Al-Quds Open University,
Faculty of Management Sciences, Gaza, Palestine.
Email: This email address is being protected from spambots. You need JavaScript enabled to view it. or This email address is being protected from spambots. You need JavaScript enabled to view it.

ABSTRACT
The information systems (IS) assessment studies have still used the commonly traditional tools such as questionnaires in evaluating the dependent variables and specially effectiveness of systems. Neural networks have recently considered as an effective substitute tool for modelling the complicated systems and broadly used for prediction. A very few is known about the employment of Artificial Neural Network (ANN) in the prediction IS effectiveness. For this reason, this research is considered as one of the fewest research to investigate the efficiency and capability of using ANN for forecasting the user perceptions towards IS effectiveness where MATLAB is utilized for building and training the neural network model. A dataset of 175 subjects collected from international organization are utilized for ANN learning where each subject consists of 6 features (5 quality factors as inputs and one Boolean output). A percentage of 75% o subjects are used in the training phase. The results indicate an evidence on the ANN models has a reasonable accuracy in forecasting the IS effectiveness. For prediction, ANN with PURELIN (ANNP) and ANN with TANSIG (ANNTS) transfer functions are used. It is found that both two models have a reasonable prediction, however, the accuracy of ANNTS model is better than ANNP model (88.6% and 70.4% respectively). As the study proposes a new model for predicting IS dependent variables, it could save the considerably high cost that might be spent in sample data collection in the quantitative studies in the fields science, management, education, arts and others.

Keywords: Artificial Neural Network (ANN); Business Intelligence; Artificial Intelligence (AI); Forecasting; IS Effectiveness; Learning; MathLAB

pdf ico FULL PAPER
 
 
 
 
 

Contact Us

Managing Editor of IJSECS
Faculty of Computer Systems & Software Engineering (FSKKP)

Universiti Malaysia Pahang
Lebuhraya Tun Razak
26300 Gambang,
Kuantan, Pahang Darul Makmur.

Tel: +609 549 2133
Fax: +609 549 2144
Email: ijsecsfskkp@ump.edu.my

Visitor Counter

0052397
Today
Yesterday
This Week
Last Week
This Month
Last Month
All days
20
68
368
746
1484
2361
52397