Print ISSN: 1681-6900

Online ISSN: 2412-0758

Keywords : association rules

Propose Data Mining System to Advance E-Learning Over Online Social Network (Facebook)

Soukaena Hassan Hashem; Sarraa Mowaffaq Abood

Engineering and Technology Journal, 2015, Volume 33, Issue 3, Pages 512-527

This research presents a proposal to advance e-learning over online social network, facebook, through analyzing the structure of this network and the behavior of their users. This proposalwill construct facebook group for Iraqi postgraduate higher education computer sciences students (IPHECSS), this group consist of 300 users.
The Proposal has four consequence steps to advance the e-learning over facebook, these steps are:
1. Constructing a proposed student’s facebooks dataset for Iraq students' society called Iraqi postgraduate higher education students (IPHES), which contains self-defined characteristics of a student’s facebooks.
2. Applying customized Frequent Pattern (FP-growth) Association Rule (AR) technique to IPHES dataset as a ranker (since it calculates the frequency of attributes) and mining technique (since it extracts knowledge to predict decision making to support e-learning over facebook through analyzing student’s behavior).
3. Applying Traditional k-mean and proposed Modified k-mean techniques to IPHES dataset to advance the traditional KM in clustering the students to introduce the structure of network’s users; this helps in supporting e-learning over facebok through analyzing students broadcasting and activities. Modification on k-mean is done by injecting a preprocessing substep in traditional KM called attributes weighting depending on ranking results obtained by applying AR as a ranker and modifying Euclidian distance similarity measure to result vectors instead of single value.
4. Analyzing the results of both association rules and clustering using excel2007 and UCINET software.

Proposed Enhancement algorithm for Company Employers Management using Genetic Algorithm in Data Mining

Dalia Nabeel Kamal

Engineering and Technology Journal, 2010, Volume 28, Issue 2, Pages 261-270

Data mining is a process of automatically discovering useful information in
large data repositories that uses a variety of data analysis tools to discover patterns
and relationships that can be hidden among vast amount of data. From these
patterns and relationships, businesses and organizations can make valid
predictions about future trends in all areas of business. Association rule mining is
a typical approach used in data mining domain for uncovering interesting trends,
patterns and rules in large datasets.
This research concentrates on one particular aspect to improve the efficiency of
the association rules technique in data mining and implement the proposed
algorithm on employers management system. The resulted association which
introduced by applying rule technique, will be treated by genetic algorithm to find
a new rules that might be more efficient and powerful for proposed data base by
propose cross point ,threshold for fitness to deal consistently with the formula of
the association rules, and gives good results.