Evolutionary soft co-clustering: formulations, algorithms, and applications

An incremental method to detect communities in dynamic evolving social networks

“…Some representative methods are found in [12,22,13], although they do not make use of the community information of previous epochs to infer hidden communities of the current epoch. The second type is the so called evolutionary clustering method, which produces local clusters for each time step by introducing temporal smoothness [18,19]. Wang et al [23] proposed a model named Dynamic Bayesian Nonnegative Matrix Factorization (DBNMF), for automatic detection of overlapping communities in temporal networks.…”

“…In light of the benefit of community detection from dynamic networks, a growing number of research work has been devoted to this topic, such as the methods based on two-steps strategy [12,13,14], incremental clustering [15,16,17], evolutionary clustering [18,19], multi-agent perspective [20], and stochastic block model [21]. Among them, one of the most promising research directions is to identify communities incrementally from dynamic evolving networks.…”

Collaborative Filtering Algorithm for Employment Based on Employment Behavior

“…One of the core steps of the collaborative filtering algorithm that is based on the mining of relation is to obtain the relation of the job seekers. At present, the ways of obtaining the relation of the job seekers are mainly divided into two categories, explicit and implicit category (Zhang, Li and Feng, et al, 2015;Bruyère, Cooper and Pelletier, et al,2014;Salehi and Kamalabadi, 2013). Literature (Lee, Kaoli and Huang,2014)proposes through the explicit social network relation to obtain the connection between the job seekers, and add the social relation matrix on the basis of the original score matrix increase, which has greatly improved the effect of the algorithm.…”

Curvilinear Triangular Discretization of Biomedical Images

“…Therefore, the second approach can be viewed as a reversal of the first approach, that still relies on a similar discretization technique. This second approach is a useful tool for bioinformatics applications, for example gene expression pattern analysis [10,11,4].…”

“…Furthermore, elements can have either straight or curved sides. In our previous work [10,11] we used triangular meshes with straight sides to discretize images of fruit fly embryos. However, the embryos, like most biological objects, have curved shapes, and their discretizations with straight-sided elements have limited accuracy.…”