International Journal of Computational Systems Engineering (14 papers in press)
Towards Recent Developments in the Methods, Metrics and Datasets of Software Fault Prediction
by Deepak Sharma, Pravin Chandra
Abstract: The world of software systems is amplified with the changing environment magnifying the demand for quality software. Software fault prediction is a requisite activity ensuring the development of economic, efficient and quality software. It is the procedure for the development of models which help to identify faults in modules during early phases of software development lifecycle. Software fault prediction is one of the most prevalent research disciplines. The existing study in this domain includes numerous modeling techniques and software metrics for the early predictions of software faults. This paper aims to explore some of the prominent studies for software fault prediction in the existing literature. In this paper, software fault prediction papers since 1990 to 2017 are investigated. The paper includes the analysis of the studies having empirical validation and a good source of publication. The paper reflects the methods, metrics, and datasets available in the literature for software fault prediction. In addition, the modeling techniques based on traditional and computational intelligence based methods are also reviewed. This paper is an endeavor to assemble the existing techniques and metrics of software fault prediction with a motive to assist researchers for easy evaluation of suitable metrics for their own research scenarios.
Keywords: Software Fault Prediction; Fault Tolerance; Computational Intelligence; Software Metrics; Evaluation Metrics.
A computerised framework for characterisation of breast tissue using mammographic images
by Indrajeet Kumar, Jitendra Virmani, H.S. Bhadauria
Abstract: In this study various experiments have been conducted for four-class and two-class breast tissue pattern characterisation using various texture feature models in transform domain. These experiments have been conducted on 480 mammographic images taken from the DDSM dataset. From each image, ROI of predefined size 128 × 128 pixels has been extracted from the core location of the breast tissue where remarkable amount of glandular tissues are found. Various statistical parameters have been computed from each ROI using transform domain texture feature models. The computed texture feature vectors are passed to the classification phase. The classification phase is consisted of SVM classifier. From the exhaustive experiments conducted in this study, it can be concluded that the maximum accuracy of 86.2% has been obtained for four-class breast tissue characterisation and the maximum accuracy of 91.2% has been obtained for two-class breast tissue characterisation.
Keywords: mammography; breast tissue characterisation; BIRADS classification; texture feature models; support vector machine classifier.
Contest of strength game-based algorithm for decline of active power loss
by K. Lenin
Abstract: In this paper, contest of strength (COS) algorithm is proposed to solve the reactive power problem. It is inspired by the game of contest of strength and the algorithm imitates the features of the game. Each candidate solution is alike a team participating in a series of rope pulling competitions and based on exert pulling forces on each other the superiority of the solutions they exemplify. In this proposed algorithm the challenging teams are moving to the novel positions. Contest of strength (COS) algorithm maintains equal level of exploration and exploitation in search of near to global optimal solution in reactive power problem. Proposed contest of strength (COS) algorithm has been tested in standard IEEE 30, 57, 118 and practical 191 bus test systems and simulation results show clearly the better-quality performance of the projected algorithm in tumbling the real power loss.
Keywords: optimal reactive power; transmission loss; contest of strength; COS.
A hybrid approach to metamorphic cryptography using KIMLA and DNA concept
by Vikash Yadav, Indresh Kumar Gupta
Abstract: Increasing dependency on digital data in today's word required security in term of digital language. Some researcher advocates the use of cryptography to cope the required demand. On the other hand side some researcher advocate the use of steganography. Beauty of cryptography lies in the fact that o/p i.e., encrypted message is difficult to decode if key is unknown. On the other hand beauty of steganography lies in the fact that, secret message is hidden in such a way that it does not attract the attention of unauthorised user. In the proposed work, an effort has been made to fertilise the strength of both concept i.e., steganography and cryptography. DNA concept is used for encrypting the secured text. For hiding the message KIMLA algorithm is used. Implementation results shows that proposed algorithm is efficient in terms of providing security (robustness, payload capacity, etc.).
Keywords: KIMLA; DNA; secret text; least significant bit; LSB; cover image.
Special Issue on: ICMMCMSE-2017 Computational Problems in Science and Engineering
Edge fixed edge Steiner number of a graph
by M. Perumalsamy, P. Arul Paul Sudhahar, J. John, R. Vasanthi
Abstract: For a non-empty set W of vertices in a connected graph G, the Steiner distance d(W) of W is the minimum size of a connected sub-graph of G containing W. S(W) denotes the set of vertices that lies in Steiner W-trees. Steiner set of a graph was introduced by Chartrand and Zhang (2002). An edge Steiner set of a graph was introduced by Santhakumaran and John (2007). In this paper, we introduce edge fixed edge Steiner set and study some of the characteristics of edge fixed edge Steiner number and provide the bounds for edge fixed edge Steiner number. Also we prove the existence theorem.
Keywords: Steiner set; edge fixed Steiner set; Steiner number; edge fixed Steiner number; edge fixed edge Steiner set; edge fixed edge Steiner number.
Stochastic maximum principle for mean-field type singular optimal control problem with discounted cost
by Muthukumar Palanisamy, Deepa Ravi
Abstract: In this article, mean-field type stochastic singular optimal control problem with discounted cost is studied over an infinite time interval. The discounted cost makes the cost functional is bounded, which guarantees the existence of optimal control. The control variable has two components as classical and singular control. Moreover the singular control satisfies bounded variation, non-decreasing continuous on the left with right limits. The proposed system is investigated in two different cases, such as without and with delay. In addition, infinite horizon version of stochastic maximum principle is established by using the convex control domain in each case. The obtained theoretical results are applied to optimal harvesting problem and optimal consumption problem.
Keywords: infinite horizon; mean-field; optimal control; singular control; stochastic maximum principle.
Discrete MDP problem in service facility systems with inventory management
by C. Selvakumar, P. Maheswari, C. Elango
Abstract: In this article, we consider a discrete time (equally spaced) service facility system in which arrival of customers to the system is controlled by taking decision at the beginning of each period. Customers arrive for service are placed in a specified queue called potential queue. Controller observes the number of customers in the system (eligible queue + server), the number of items in inventory and decides the quantity of customers to be admitted from the potential queue, others are rejected. Demands arrive throughout the period, but they are satisfied (service completion) only at the end of the period. Inventory is maintained in the server to satisfy the customers at the service completion time. Inventory is replenished instantaneously when the level become zero. The system is formulated as a Markov decision process. A stationary cost structure is imposed at each decision epoch and optimal policy is obtained by using policy iteration algorithm. A numerical example is provided to illustrate the problem.
Keywords: discrete-time MDP; admission control; service facility with inventory; Markov decision process; MDP.
On the generalised degree distance of graphs
by K. Pattabiraman, Manzoor Ahmad Bhat
Abstract: The generalised degree distance of a connected graph studied by Hamzeh et al. (2012, 2013). In this paper, we investigate the mathematical properties of the generalised degree distance of a connected graphs. In addition, we present the relation between the generalised degree distance and other graph invariants.
Keywords: Wiener index; generalised degree distance; graph invariants.
Special Issue on: SCDA 2018 Advances and Challenges of Soft Computing in Data Mining
A Framework for Ensemble Classification and Sensitivity Analysis in Privacy Preserving Data Mining
by Chandrakanth Patrapati, Chandrakanth Patrapati
Abstract: The perturbation mechanism for data streams is a challenging task. In the emerging world, data is erupting from various sources. The core applications need care on the data streams for further analysis and experimentation. As the micro data available with the core applications shall not be revealed to the public without taking any chance of breach, the perturbation challenges the analysis to get through the like results as of on the original data sets. In this paper we have applied a concept of Perlin noise to distract the original data from the eyes of the analysts, however allowing them to perform their activities well. The data evolving in the trendy applications of the today world needs security, almost of the applications in the world are related directly or indirectly to the personally identifiable information (micro data) of the people. The security that assures the privacy on these data is hitherto static in the applications. Applying security dynamically on such data is a challenging task. The perturbation mechanism canonically applies the noise on the datasets, where some of them do not require. This paper deals about the concepts of generation of smooth noise and syntactic perturbation mechanism on the selective tuples as selective perturbation.
Keywords: privacy preserving data mining;data streams;ensemble classifier;sensitivity; smooth noise.
Feed Forwarded CT Image Registration for Tumor and Cyst Detection using rigid transformation with HSV color segmentation
by PRABHAT BARIK, R.A.M. BARIK
Abstract: This paper automates the Medical diagnosis process by Image Registration feed forwarded to object detection with prediction of tumour and cyst by using K-means clustering over HSV colour features. Diagnosis of life killer disease is a complex process which requires Bio-medical image such as MRI, CT, and Endoscopy etc. Many Biomedical Images is used for the same case to predict the disease. Because of different view point of different Photographic sensors at different time obtained medical images are not aligned. So the manual diagnosis makes harder because of the images are not registered or not aligned properly. The inherent cause is the distortion of the imaging signal where object may be miss-transformed due to different camera focus and projection. Image registration is an essence to bypass the non-alignment issue. Here we have proposed and analysed a combined solution towards the miss transformed object or region of interest by performing reverse geometric transformation with different angle to produce the better perspective image for diagnosis which is feed forwarded to HSV colour model based segmentation to predict the cyst, tumour presence.
Keywords: Image Registration; Color Segmentation; HSV (Hue Saturation Value); MRI (Machine Resonance Imaging); CT (Computer Tomography); Rigid Transformations; K-means Clustering.
A NOVEL RELAYING APPROACH OF COMBINED DISCRETE WAVELET TRANSFORM AND ARTIFICIAL NEURAL NETWORK BASED RELAYING SCHEME IN AN UNIFIED POWER FLOW CONTROLLER INTEGRATED WIND FED TRANSMISSION LINE
by SANJAY KUMAR MISHRA, Loknath Tripathy
Abstract: This paper focuses a novel relaying approach of combined Discrete Wavelet Transform (DWT) and Artificial Neural Network (ANN) based relaying scheme in Unified Power Flow Controller (UPFC) integrated Wind fed transmission line. This novelty of the scheme illustrates the fault detection of transmission line and minimises the fault detection time through the combined algorithm of DWT and ANN. However the combined scheme (DWT and ANN) works better as compared to DWT alone and fault time reduces to quarter cycle time (5msec). The process starts with DWT processor to find Spectral energy (SE). The differential spectral energy (DSE) is computed using the difference of spectral energy obtained from both sending and receiving end of the transmission line. The computed DSE value of each phase from DWT is fed to the ANN fault detector to register the fault pattern. It is trained and tested with different parameter variation of the line to obtain fault detection output and fault time. The advantages of using the scheme is that its performance accuracy is higher and detects the fault within 5msec time and compared with the existing scheme. The proposed approach is very effective, accurate, reliable and simple to understand for fault detection, classification of transmission line in comparison to existing methodology.
Keywords: ANN; DSE; UPFC; WIND; DWT;.
Empirical Validation of Object-Oriented Metrics on Cross-Projects with Different Severity Levels
by Aarti Aarti, Geeta Sikka, Renu Dhir
Abstract: An object-oriented (OO) metrics has become crucial desideratum for software effort and fault predictions. To strengthen the adequacy of object-oriented metrics, it becomes important to know relationship between OO metrics and fault proneness at different levels of severity. It is inconceivable to build model of accurate estimate due to the inherent uncertainty in development projects. Empirical validations of software metrics are essential issue to determine applicability of prediction model. In this study, empirical validation is done on OO metrics given by Chidamber and Kemerer (CK suite) for predicting faults at different severity levels. This paper also instanced on defect prediction using cross-projects (CP) because of the unpredictability in selection of software attributes by analogy based approach that deliver imprecise and ambiguous solution. This paper depicts detection of fault-proneness by extracting the relevant OO metrics and such models helps to focus on fault prone modules of new projects by allocating more resources to them with use of regression and other machine learning methods. Combination of CP data with regression techniques improves effectiveness of prediction by extracting similar features impacted by all datasets. The performance is evaluated using the receiver and operator (ROC) parameters. The results concluded that proposed methodology has great potential for conducting prediction of faults and shows that analysis of result using machine learning techniques outperforms as compared to logistic regression.
Keywords: Fault; Object-oriented (OO) metrics; Classification; ROC; Level of severity; Empirical Validation.
Inverse kinematic Solution of 6 -DOF Industrial Robot using Nero-Fuzzy Technology
by Kshitish Dash, Bibhuti Choudhury
Abstract: The robot inverse kinematic controller does not give the shut frame arrangement. Henceforth Mechanical controller can accomplish end effectors position in more than one arrangement. To accomplish correct arrangement of the joint angle has been the fundamental worry in the research work. In this paper the analytical solution has been done using D-H method. The method gives the 6 DOF industrial robot with D-H Parameter value which will be the best uses for any inverse kinematics algorithm. Levenberg-Marquardt algorithm is used to solve inverse kinematic of 6-DOF industrial robot arm and the result has been simulated with different soft computing method like ANN and Fuzzy logic . A comparison is taken between both the result obtain from different sources.
Keywords: Inverse Kinematics; ANN ; Fuzzy logic; Industrial Robot; forward kinematics,D-H parameters.
A phase Entropy based novel Machine learning structure conditioned for classifying Ictal and Non-Ictal signal aimed at proper clinical diagnosis
by SANTOSH KUMAR SAHOO
Abstract: Proposed Scheme is based on the detection and classification of Ictal and pre-Ictal Electro -Encephalogram (EEG) signal of a partially affected epileptic patients. This work helps for the detection of ictal EEG signal as compared to pre-ictal EEG signal. In this work a conceptual method is used for the perfect detection of seizures. Here Phase Entropies (PEnS1 and PEnS2) are followed for feature extraction Ictal and pre-Ictal EEG signals. Again extracted features are classified through Multi-Layer Perception Neural Network (MLPNN) classifier tool. To train and test the classifier, Map Reduce quantum PSO (MRQPSO) is used. By considering various statistical parameters like Accuracy, Sensitivity, Specificity, positive predictive value, negative predictive value and Matthews correlation coefficient, the performance of the proposed scheme has evaluated.
Keywords: EEG; MLPNN; ICTAL; MRQPSO.