Predicting User Entries by Using Data Mining Algorithms

The information systems are widely spread in most official institutions, and become certified in all areas of our life such as education, health and entertainment. Usability is one of the most important factors, which encourages users to deal with these systems or refuse it. Data mining is the process of finding correlations or patterns among dozens of fields in large relational databases. In this paper we analyze the stored data in database of Palestinian Government decisions system in order to study the relationship between some attributes. Accordingly, we can find patterns that help us to make the system more user-friendly by offering suggestions to the users during data entry process. Naive Bayes, Rule Induction, K-NN, and Decision Tree methods are applied to the stored data in order to produce a prediction model that predicts entries to the user during the entry process, which can make the entry system more user-friendly. The experiment result shows the Naïve Bayes is the best model among the other techniques by achieving the highest accuracy of 68.41%. Future efforts can apply this model in the Government decisions system of Palestinian Ministers Council in Gaza.