SUMMARYClonal Selection Algorithm (CSA), based on the clonal selection theory proposed by Burnet, has gained much attention and wide applications during the last decade. However, the proliferation process in the case of immune cells is asexual. That is, there is no information exchange during different immune cells. As a result the traditional CSA is often not satisfactory and is easy to be trapped in local optima so as to be premature convergence. To solve such a problem, inspired by the quantum interference mechanics, an improved quantum crossover operator is introduced and embedded in the traditional CSA. Simulation results based on the traveling salesman problems (TSP) have demonstrated the effectiveness of the quantum crossover-based Clonal Selection Algorithm. key words: clonal selection algorithm, quantum interference crossover, traveling salesman problem, hybrid model

No abstract

Recommender systems improve access to relevant products and information by making suggestions based on page ranking technology. Existing approaches to learning to rank, however, did not consider the pages in the deep web which have valuable information. In this paper, we present a novel product recommendation algorithm based on the content of web pages including the product information and customer reviews. Our algorithm uses the customer reviews to calculate the score of dynamic web pages. The paper further focus on classifying the semantic orientation of the customer reviews through a progressed Bayesian Classifier and calculating the support value of each review. In addition, we also analyze the change tendency of customer reviews based on the temporal dimension. Experimental results shows that this approach can produce accurate recommendations.keyword--Review, Bayesian Classifier, Recommendation PageRank

Many studies reported that spontaneous fluctuation of the blood oxygen level-dependent signal exists in multiple frequency components and changes over time. By assuming a reliable energy contrast between low- and high-frequency bands for each voxel, we developed a novel spectrum contrast mapping (SCM) method to decode brain activity at the voxel-wise level and further validated it in designed experiments. SCM consists of the following steps: first, the time course of each given voxel is subjected to fast Fourier transformation; the corresponding spectrum is divided into low- and high-frequency bands by given reference frequency points; then, the spectral energy ratio of the low- to high-frequency bands is calculated for each given voxel. Finally, the activity decoding map is formed by the aforementioned energy contrast values of each voxel. Our experimental results demonstrate that the SCM (1) was able to characterize the energy contrast of task-related brain regions; (2) could decode brain activity at rest, as validated by the eyes-closed and eyes-open resting-state experiments; (3) was verified with test-retest validation, indicating excellent reliability with most coefficients > 0.9 across the test sessions; and (4) could locate the aberrant energy contrast regions which might reveal the brain pathology of brain diseases, such as Parkinson’s disease. In summary, we demonstrated that the reliable energy contrast feature was a useful biomarker in characterizing brain states, and the corresponding SCM showed excellent brain activity-decoding performance at the individual and group levels, implying its potentially broad application in neuroscience, neuroimaging, and brain diseases.

With the characteristics of simple structure and low cost, the dendritic neuron model (DNM) is used as a neuron model to solve complex problems such as nonlinear problems for achieving high-precision models. Although the DNM obtains higher accuracy and effectiveness than the middle layer of the multilayer perceptron in small-scale classification problems, there are no examples that apply it to large-scale classification problems. To achieve better performance for solving practical problems, an approximate Newton-type method-neural network with random weights for the comparison; and three learning algorithms including back-propagation (BP), biogeography-based optimization (BBO), and a competitive swarm optimizer (CSO) are used in the DNM in this experiment. Moreover, three classification problems are solved by using the above learning algorithms to verify their precision and effectiveness in large-scale classification problems. As a consequence, in the case of execution time, DNM + BP is the optimum; DNM + CSO is the best in terms of both accuracy stability and execution time; and considering the stability of comprehensive performance and the convergence rate, DNM + BBO is a wise choice.