Robust Dimension Reduction for Clustering With Local Adaptive Learning.

In pattern recognition and data mining, clustering is a classical technique to group matters of interest and has been widely employed to numerous applications. Among various clustering algorithms, K-means (KM) clustering is most popular for its simplicity and efficiency. However, with the rapid development of the social network, high-dimensional data are frequently generated, which poses a considerable challenge to the traditional KM clustering as the curse of dimensionality. In such scenarios, it is difficult to directly cluster such high-dimensional data that always contain redundant features and noises. Although the existing approaches try to solve this problem using joint subspace learning and KM clustering, there are still the following limitations: 1) the discriminative information in low-dimensional subspace is not well captured; 2) the intrinsic geometric information is seldom considered; and 3) the optimizing procedure of a discrete cluster indicator matrix is vulnerable to noises. In this paper, we propose a novel clustering model to cope with the above-mentioned challenges. Within the proposed model, discriminative information is adaptively explored by unifying local adaptive subspace learning and KM clustering. We extend the proposed model using a robust l2,1 -norm loss function, where the robust cluster centroid can be calculated in a weighted iterative procedure. We also explore and discuss the relationships between the proposed algorithm and several related studies. Extensive experiments on kinds of benchmark data sets demonstrate the advantage of the proposed model compared with the state-of-the-art clustering approaches.

[Studies of F1 of transgenic allotetraploid hybrids of Carassius auratus red var. (Female) x Cyprinus carpio (Male)].

The tetraploid fish has been developed by assortative breeding the hybrids of Carassius auratus red var. (Female) x cyprinus carpio (Male), which has the stable genetic characters and can reproduce themselves. An "all-fish" recombinant DNA construct (pCAgcGHc) containing common carp beta-actin gene promoter and cDNA for grass carp (Ctenopharyngodon idella) growth hormone gene was introduced into fertilized eggs of the allotetroploid fish through microinjection as soon as artificial insemination was done. Artificial insemination was carried out between the female and the male transgenic allotetraploid fish which contain the "all-fish" recombinant DNA construct (pCAgcGHc) and are the biggest in the size. Fifty F1 samples of transgenic allotetraploid fish of 150 days and 50 allotetraploid fish (regarded as the control) were chosen, and the weight and the body length of each were measured, the results showed that F1 of transgenic allotetraploid fish of 150 days had obvious growth dominance compared with the control. Genomic DNA of tail fin was extracted from 20 F1 of transgenic allotetraploid fish of 150 days and the control. Proper primers were introduced to check whether the sample had the transgene. Pa, the upstream primer, is located in beta-actin promoter, and Pg, the downstream primer, is located in growth hormone cDNA for grass carp (gcGHc). The transgene was detected in 90% F1 of transgenic allotetraploid fish in tail fin DNA by polymerase chain reaction (PCR) amplification. Sperm could be squeezed out from a few F1 of transgenic allotetraploid fish of 150 days, however, this phenomenon did not exist in the controls. The importance of forming the pure line of transgenic allotetraploid was elucidate in the paper.