S. Nickolas

A hybrid discrete firefly algorithm for solving multi-objective flexible job shop scheduling problems

Firefly algorithm FA is a nature-inspired optimisation algorithm that can be successfully applied to continuous optimisation problems. However, lot of practical problems are formulated as discrete optimisation problems. In this paper a hybrid discrete firefly algorithm HDFA is proposed to solve the multi-objective flexible job shop scheduling problem FJSP. FJSP is an extension of the classical job shop scheduling problem that allows an operation to be processed by any machine from a given set along different routes. Three minimisation objectives - the maximum completion time, the workload of the critical machine and the total workload of all machines are considered simultaneously. This paper also proposes firefly algorithms discretisation which consists of constructing a suitable conversion of the continuous functions as attractiveness, distance and movement, into new discrete functions. In the proposed algorithm discrete firefly algorithm DFA is combined with local search LS method to enhance the searching accuracy and information sharing among fireflies. The experimental results on the well-known benchmark instances and comparison with other recently published algorithms shows that the proposed algorithm is feasible and an effective approach for the multi-objective flexible job shop scheduling problems.

Performance Analysis of Datamining Algorithms for Software Quality Prediction

Data mining techniques are applied in building software fault prediction models for improving the software quality. Early identification of high-risk modules can assist in quality enhancement efforts to modules that are likely to have a high number of faults. Classification tree models are simple and effective as software quality prediction models, and timely predictions of defects from such models can be used to achieve high software reliability. In this paper, the performance of five data mining classifier algorithms named J48, CART, Random Forest, BFTree and Naïve Bayesian classifier(NBC) are evaluated based on 10 fold cross validation test. Experimental results using KC2 NASA software metrics dataset demonstrates that decision trees are much useful for fault predictions and based on rules generated only some measurement attributes in the given set of the metrics play an important role in establishing final rules and for improving the software quality by giving correct predictions. Thus we can suggest that these attributes are sufficient for future classification process. To evaluate the performance of the above algorithms Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), Receiver Operating Characteristic (ROC) and Accuracy measures are applied.

Enhancing the Performance of LibSVM Classifier by Kernel F-Score Feature Selection

Medical Data mining is the search for relationships and patterns within the medical datasets that could provide useful knowledge for effective clinical decisions. The inclusion of irrelevant, redundant and noisy features in the process model results in poor predictive accuracy. Much research work in data mining has gone into improving the predictive accuracy of the classifiers by applying the techniques of feature selection. Feature selection in medical data mining is appreciable as the diagnosis of the disease could be done in this patient-care activity with minimum number of significant features. The objective of this work is to show that selecting the more significant features would improve the performance of the classifier. We empirically evaluate the classification effectiveness of LibSVM classifier on the reduced feature subset of diabetes dataset. The evaluations suggest that the feature subset selected improves the predictive accuracy of the classifier and reduce false negatives and false positives.

A Level-wise Priority Based Task Scheduling for Heterogeneous Systems

—Achieving high performance without proper scheduling of application tasks is impossible in a heterogeneous environment. To solve such an issue, in this paper, a new static scheduling algorithm is proposed called expected completion time based scheduling (ECTS) algorithm, which is used to effectively schedule application tasks on to the heterogeneous processors. The ECTS algorithm finds the task sequence for execution by assigning priority and then maps the selected task sequence on to the processors. In order to give the comparison of proposed algorithm over the existing algorithms, a real Fast Fourier application graphs are considered as experimental test case. The results show the effectiveness of the proposed algorithm to those presented previously. The algorithm is mainly focused on producing minimum makespan.

Modified multi-objective firefly algorithm for task scheduling problem on heterogeneous systems

Scheduling an application in a heterogeneous environment to find an optimal schedule is a challenging optimisation problem. Maximising the reliability of the application even when processors fails, adds more complexity to the problem. Both the objectives are conflict in nature, where maximising reliability of the application may increase applications completion time. Meta-heuristic algorithms are playing important role in solving the optimisation problem. In this paper, the applicability and efficiency of the new meta-heuristic algorithm called firefly algorithm to solve the workflow multi-objective task scheduling problem is studied. A modified version of the firefly algorithm MFA using weighted sum method and a modified version of multi-objective firefly algorithm MMOFA using Pareto-dominance method are proposed to solve the multi-objective task scheduling problem. The simulation results show that the proposed algorithms can be used for producing task assignments and also give significant improvements in terms of generating schedule with minimum makespan and maximum reliability compared with existing algorithms.

A Novel Algorithm for Mining Hybrid-Dimensional Association Rules

rule mining is a fundamental and vital functionality of data mining. M ost of the existing real time transactional databases are multidimensional in nature. In this paper, a novel algorithm is proposed for mining hybrid-dimensional association rules which are very useful in business decision making. The proposed algorithm uses multi index structures to store necessary details like item combination, support measure and transaction IDs, which stores all frequent 1-itemsets after scanning the entire database first time. Frequent k-itemsets are generated with previous level data, without scanning the database further. Compared to traditional algorithms, this algorithm efficiently finds association rules in multidimensional datasets, by scanning the database only once, thus enhancing the process of data mining.

ANOVA Discriminant Analysis for Features Selected through Decision Tree Induction Method

Feature selection plays an important role in recent years as the selected features improve the classification accuracy of the classifiers in software defect predictions. For improving the quality of the defect prediction model, effective features must be selected. Many feature selection algorithms have been developed for the improvement of defect prediction models. In this paper, we have used ANOVA Discriminant Analysis (ADA) to statistically prove that the features selected through Decision Tree Induction (DTI) approach are effective for defect predictions and only these features can be used for further use. ADA computes the means of variables and it selects the significant features from the groups iteratively for better results. By using ADA, it is found that the features selected through the DTI approach have higher discriminating power than others and the accuracy of the classifiers also increases when this feature set is used for many classifiers. Hence it is said that the selected feature set alone can be used for defect prediction instead of original feature set. It is also observed that the attributes selected through DTI are same as attributes used by ADA. Hence, these are proved to be significant. Wilk’s Lambda is taken as the significant measure. It shows the Discriminant power of the features i.e. discriminating power is high when Wilk’s lambda is low.

Effective task scheduling for heterogeneous distributed systems using firefly algorithm

Achieving minimum execution time for any application with better resource utilisation is a major challenge in heterogeneous distributed systems. But the performance can be exploited in these systems through proper scheduling of application tasks. An efficient meta-heuristic algorithm called firefly algorithm is applied in this paper to solve static task scheduling problem in heterogeneous systems. The social behaviour of fireflies is mimicked to generate optimal task schedule length. The efficiency of the firefly-based task scheduling algorithm is compared with the existing particle swarm optimisation-based scheduling algorithm. The experimental results show that the firefly algorithm-based approach gives better results when compared to PSO algorithm and performs well with minimum processors for effective scheduling of tasks.

HUPT-mine: an efficient algorithm for high utility pattern mining

In recent years, the problem of high utility pattern mining becomes one of the most important research areas in data mining. High utility pattern mining extracts patterns which have utility value higher than or equal to user specified minimum utility. The problem is challenging, because of the non-applicability of anti-monotone property of frequent pattern mining. The existing high utility pattern mining algorithm adopts level wise candidate generation and many recently proposed approaches also generate large number of candidate itemsets. In this paper, a novel high utility pattern tree (HUPT) is proposed by applying two pruning strategies to reduce number of candidate itemsets by scanning database twice. For each conditional pattern base, a local tree is constructed with required information to generate candidate itemsets, by employing pattern growth approach. The experimental results on different datasets show that it reduces the number of candidate itemsets and also outperforms two-phase algorithm for dense datasets with long transactions.

A path priority-based task scheduling algorithm for heterogeneous distributed systems

Finding an effective schedule of tasks of distributed applications on the heterogeneous distributed computing systems is a trade-off between minimising the makespan and maximising the processor utilisation. The performance can be exploited in these systems through proper scheduling of application tasks. In this paper, a new task scheduling algorithm, called path-based heuristic task scheduling PHTS algorithm is proposed. The performance of the proposed algorithm is experimentally compared with the three existing scheduling algorithms. The results show that the PHTS algorithm outperforms other algorithms in terms of schedule length ratio, speedup and efficiency for applications with more tasks.