Software reliability is estimated and predicted based on software reliability model and software failure data. As a new optimization method, swarm intelligence algorithm has been widely used in solving the parameter optimization of the model. WPA (Wolf Pack Algorithm) and PSO (Particle Swarm Optimization) are two typical swarm intelligence algorithms. WPA has a strong global optimization ability, fast convergence speed and various optimization strategies, but the algorithm is relatively complex. PSO algorithm has a simple structure and fast convergence speed, but it is easy to fall into premature, which leads to low accuracy of solution. Considering the advantages and disadvantages of the two algorithms, a hybrid method of WPA and PSO is proposed, and a fitness function is constructed on maximum likelihood estimation, then the parameters of software reliability model are estimated and predicted based on the hybrid algorithm (WPA-PSO). Five sets of data from industry are used to estimate the parameters of GO model and make predictions. The simulation results show that the hybrid algorithm has higher accuracy of parameter estimation, better optimization performance, better accuracy of prediction and algorithm stability than single algorithm, and show obvious advantages than the single algorithm in the case of limited data.

Software defects reflect software quality, and software failures can be predicted through software reliability models. Aiming at the problem that the parameters of software reliability model are difficult to estimate, this paper used the hybrid algorithm for model parameter estimation to software defect prediction. As a typical swarm intelligence algorithm, PSO (Particle Swarm Optimization) has fast convergence but low solution accuracy. SSA (Sparrow Search Algorithm) not only has high search accuracy and fast convergence speed, but also has the advantages of good stability and strong robustness. Based on the characteristic that the fitness function proposed in this paper, this paper hybrid PSO and SSA to accelerate the convergence before the individual update of the SSA. At the same time, this paper also constructed a new fitness function based on the maximum likelihood estimation of the parameters, and used it for parameter initialization. Through the analysis of the experimental results of five sets of actual data sets, the optimization performance of the hybrid algorithm (SSA-PSO) was better than that of a single algorithm with higher convergence speed and more stable, accurate results. Moreover, with the support of the new fitness function, it effectively solved the problems of slow convergence speed and low accuracy of solution. The experimental results showed that the hybrid SSA-PSO could obtain the better solution, convergence speed and stability than single SSA and PSO in software defections estimation and prediction.

No abstract

This study manipulated the graphical representation of options by framing the physical characters in figures and found that preferences could be affected even when the words and numbers of the problem were constant. Based on attribute substitution theory and an equate-todifferentiate approach, we proposed a two-process model of graph-framing effects. In the first mental process, the graph-editing process, the physical features (e.g., distance, size) represented in the graph are visually edited, and the perceived numerical difference between the options is judged based on its physical features. The second mental process, the preferential choice process, occurs by an equate-to-differentiate approach in which people seek to equate the difference between options on the dimension on which the difference is smaller, thus leaving the greater other-dimensional difference to be the determinant of the final choice. Four experiments were tested for graph-framing effects. Experiment 1 found a graph-framing effect in coordinate graphs resting on the (de)compression of the scales employed in the figures. Experiment 2 revealed additional graph-framing effects in other question scenarios and showed that preference changes were mediated by perceived numerical distances. Experiment 3 further confirmed the presence of graph-framing effects in sector graphs similar to those found in coordinate ones. Experiment 4 suggested that such graph-framing effects could be eliminated when logical processing (e.g., introducing a mathematical operation before a choice task) was encouraged. This paper discusses related research and a possible substrate basis for graphframing effects.

AdaBoost has been proved a successful statistical learning method for concept detection with high performance of discrimination and generalization. However, it is computationally expensive to train a concept detector using boosting, especially on large scale datasets. The bottleneck of training phase is to select the best learner among massive learners. Traditional approaches for selecting a weak classifier usually run in , with N examples and T learners. In this paper, we treat the best learner selection as a Nearest Neighbor Search problem in the function space instead of feature space. With the help of Locality Sensitive Hashing (LSH) algorithm, the best learner searching procedure can be speeded up in the time of , where L is the number of buckets in LSH. Compared with the T (~500,000), the L (~600) is much smaller in our experiments. In addition, through studying the distribution of weak learners and candidate query points, we present an efficient method to try to partition the weak learner points and the feasible region of query points uniformly as much as possible, which can achieve significant improvement in both recall and precision compared with the random projection in traditional LSH algorithm. Experimental results reveal our method can significantly reduce the training time. And still the performance of our method is comparable with the state-of-art methods.