An algorithm for calculation of degree distribution and detection of network type: with application in food webs

In present study a Java algorithm to calculate degree distribution and detect network type was presented. Some indices, e.g., aggregation index, coefficient of variation, skewness, etc., were first suggested for detecting network type. Network types of some food webs reported in Interaction Web Database were determined using the algorithm. The results showed that the degree of most food webs was power law or exponentially distributed and they were complex networks. Different from classical distribution patterns (bionomial distribution, Poisson distribution, and power law distribution, etc.), both network type and network complexity can be calculated and compared using the indices above. We suggest that they should be used in the network analysis. In addition, we defined E, E=s2-u, where u and s2 is mean and variance of degree respectively, as the entropy of network. A more complex network has the larger entropy. If E is not greater than 0, the network is a random network and, it is a complex network if E is greater than 0.