Search

Recently an increasing number of research is devoted to the question of how to find the most influential nodes effectively in complex networks Actually, the problem of complex network nodes attack is searching the way to find most important nodes in complex network. In this paper, the problem is transformed into traveling salesman problem. An evaluation method for measuring the importance of nodes based on the average network reliability is designed. The improved genetic algorithm is proposed to solve the optimization problem. We establish a mathematical optimization model with the minimization of network connectivity as a target function. We generate the node attack sequence using the improved genetic algorithm. Experiments showed that the new evaluation strategy is more effective than traditional methods.

The research in the high sensitivity of the biological sensorium has been the blank in the work of NN (neural network). According to the analyzing results of chaotic theory, the chaotic system has an extremely high sensitivity and resolving power. Based on the application of chaotic theory to measuring, this paper proposes a concept of the sensitive chaotic cell neuron (SCCN) and a mathematical model for SCCN. Like the biological sensorium, the SCCN can get a high sensitivity and resolving power to feeble signals. The simulation results show the high sensitivity of SCCN to the feeble signals.

An optimizing syntax-parsing algorithm is proposed on the basis of the best-first technique and the preference-based technique. By using SPR (structure preference relation) data as local preference information, we indirectly realize the best-first idea. Experimental results indicate that by determining a proper SPR threshold, we can increase both the whole efficiency and parsing accuracy of a Chinese probabilistic parser.