Current Issue : January - March Volume : 2018 Issue Number : 1 Articles : 5 Articles
This paper describes the software requirements prioritization task and provides a systematic approach to\ndetermine what needs to be included in the next release of a software product. Minimizing the total cost\nof adding a new feature in the next release and maximizing overall customer satisfaction are contradictory\nobjectives. In this paper, first, an adaptive multi-objective prioritization model is discussed. Then we\ndescribe how discrete inverse problems ideas can in fact be formulated to obtain a smooth local ââ?¬Å?Added\nDegree of Importanceââ?¬Â (ADI) function of client requirements which could be used to classify and\nprioritize the software requirements for next release. The numerical implementation of the proposed\nmodel with a case study on software requirements selection shows the effectiveness of the multi-objective\ninverse model (IM) approach. The proposed model have been compared with some of the recent relevant\nmodels. Main future of the model is that it has been designed by the assignment of a real score for each of\nthe requirements unlike just classification provided in the literature....
The paper proposes a solution to the problem classification by calculating the\nsequence of matrices of feature indices that approximate invariants of the data\nmatrix. Here the feature index is the index of interval for feature values, and\nthe number of intervals is a parameter. Objects with the equal indices form\ngranules, including information granules, which correspond to the objects of\nthe training sample of a certain class. From the ratios of the information granules\nlengths, we obtain the frequency intervals of any feature that are the\nsame for the appropriate objects of the control sample. Then, for an arbitrary\nobject, we find object probability estimation in each class and then the class of\nobject that corresponds to the maximum probability. For a sequence of the\nparameter values, we find a converging sequence of error rates. An additional\neffect is created by the parameters aimed at increasing the data variety and\ncompressing rare data. The high accuracy and stability of the results obtained\nusing this method have been confirmed for nine data set from the UCI repository.\nThe proposed method has obvious advantages over existing ones due\nto the algorithm�s simplicity and universality, as well as the accuracy of the\nsolutions....
ABSTRACT\nIn today�s digital era, it becomes a challenge for netizens to find\nspecific information on the internet. Many web-based documents\nare retrieved and it is not easy to digest all the retrieved\ninformation. Automatic text summarization is a process that\nidentifies the important points from all the related documents to\nproduce a concise summary. In this paper, we propose a text\nsummarization model based on classification using neuro-fuzzy\napproach. The model can be trained to filter high-quality\nsummary sentences. We then compare the performance of our\nproposed model with the existing approaches, which are based\non fuzzy logic and neural network techniques. ANFIS showed\nimproved results compared to the previous techniques in terms of\naverage precision, recall and F-measure on the Document\nUnderstanding Conference (DUC) data corpus....
To deal with the problems of premature convergence and tending to jump into\nthe local optimum in the traditional particle swarm optimization, a novel\nimproved particle swarm optimization algorithm was proposed. The\nself-adaptive inertia weight factor was used to accelerate the converging speed,\nand chaotic sequences were used to tune the acceleration coefficients for the\nbalance between exploration and exploitation. The performance of the proposed\nalgorithm was tested on four classical multi-objective optimization\nfunctions by comparing with the non-dominated sorting genetic algorithm\nand multi-objective particle swarm optimization algorithm. The results verified\nthe effectiveness of the algorithm, which improved the premature convergence\nproblem with faster convergence rate and strong ability to jump out\nof local optimum....
This paper investigates the effectiveness of four different soft computing methods, namely\nradial basis neural network (RBNN), adaptive neuro fuzzy inference system (ANFIS) with subtractive\nclustering (ANFIS-SC), ANFIS with fuzzy c-means clustering (ANFIS-FCM) and M5 model tree\n(M5Tree), for predicting the ultimate strength and strain of concrete cylinders confined with\nfiber-reinforced polymer (FRP) sheets. The models were compared according to the root mean\nsquare error (RMSE), mean absolute relative error (MARE) and determination coefficient (R2) criteria.\nSimilar accuracy was obtained by RBNN and ANFIS-FCM, and they provided better estimates\nin modeling ultimate strength of confined concrete. The ANFIS-SC, however, performed slightly\nbetter than the RBNN and ANFIS-FCM in estimating ultimate strain of confined concrete, and\nM5Tree provided the worst strength and strain estimates. Finally, the effects of strain ratio and the\nconfinement stiffness ratio on strength and strain were investigated, and the confinement stiffness\nratio was shown to be more effective....
Loading....