Browsing by Author "Abdel-Jaber, Hussein"
Now showing 1 - 2 of 2
Results Per Page
Sort Options
Item Modeling discrete-time analytical models based on random early detection : exponential and linear(World Scientific Publishing Co. Pte Ltd, 2015) Abdel-Jaber, Hussein; Thabtah, Fadi; Woodward, MikeCongestion control is among primary topics in computer network in which random early detection (RED) method is one of its common techniques. Nevertheless, RED suffers from drawbacks in particular when its "average queue length" is set below the buffer's "minimum threshold" position which makes the router buffer quickly overflow. To deal with this issue, this paper proposes two discrete-time queue analytical models that aim to utilize an instant queue length parameter as a congestion measure. This assigns mean queue length (mql) and average queueing delay smaller values than those for RED and eventually reduces buffers overflow. A comparison between RED and the proposed analytical models was conducted to identify the model that offers better performance. The proposed models outperform the classic RED in regards to mql and average queueing delay measures when congestion exists. This work also compares one of the proposed models (RED-Linear) with another analytical model named threshold-based linear reduction of arrival rate (TLRAR). The results of the mql, average queueing delay and the probability of packet loss for TLRAR are deteriorated when heavy congestion occurs, whereas, the results of our RED-Linear were not impacted and this shows superiority of our model. © 2015 World Scientific Publishing Company.Item Parallel associative classification data mining frameworks based mapreduce(World Scientific Publishing Co. Pte Ltd, 2015-06) Thabtah, Fadi; Hammoud, Suhel; Abdel-Jaber, HusseinAssociative classification (AC) is a research topic that integrates association rules with classification in data mining to build classifiers. After dissemination of the Classification-based Association Rule algorithm (CBA), the majority of its successors have been developed to improve either CBA's prediction accuracy or the search for frequent ruleitems in the rule discovery step. Both of these steps require high demands in processing time and memory especially in cases of large training data sets or a low minimum support threshold value. In this paper, we overcome the problem of mining large training data sets by proposing a new learning method that repeatedly transforms data between line and item spaces to quickly discover frequent ruleitems, generate rules, subsequently rank and prune rules. This new learning method has been implemented in a parallel Map-Reduce (MR) algorithm called MRMCAR which can be considered the first parallel AC algorithm in the literature. The new learning method can be utilised in the different steps within any AC or association rule mining algorithms which scales well if contrasted with current horizontal or vertical methods. Two versions of the learning method (Weka, Hadoop) have been implemented and a number of experiments against different data sets have been conducted. The ground bases of the comparisons are classification accuracy and time required by the algorithm for data initialization, frequent ruleitems discovery, rule generation and rule pruning. The results reveal that MRMCAR is superior to both current AC mining algorithms and rule based classification algorithms in improving the classification performance with respect to accuracy. © 2015 World Scientific Publishing Company.