Semi-supervised classification based on p-norm multiple kernel learning with manifold regularization

Consider the efficiency of p-norm multiple kernel learning (MKL), which is extended to a semi-supervised learning (SSL) scenario by applying the manifold regularization technique. A manifold regularized p-norm multiple kernels model is constructed and applied to a semi-supervised classification task. Solutions are proposed for the case of p = 1, p > 1 and p = ∞, with an analysis of theorems and their proofs. In addition, experiments are conducted on several datasets using state-of-the-art methods to verify the efficiency of the proposed manifold regularized p-norm multiple kernels model in semi-supervised classification. ? 2016 Beijing Institute of Aerospace Information.

Data mining to effect of key alloying elements on corrosion resistance of low alloy steels in Sanya seawater environmentAlloying Elements

In this paper,the relationship model between seawater environment,chemical composition and corrosion potential of low alloy steel is established and the distribution of corrosion potential of low alloy steel with changes in key alloying elements is excavated.The research was carried out with the following steps:Firstly,the relationship model between corrosion potential of low alloy steel and its influencing factors was established by data dimension reduction and artificial neural network(ANN).Secondly,key alloying elements of experimental steels were selected out by Pearson correlation analysis,then the corrosion resistance element model was visualized to show the effect of key alloying elements on corrosion potential of low alloy steel.Finally,corrosion potential of low alloy steel with the change of key alloying elements was classified and visualized by classification method.The mining results can reflect the validity of the proposed mining methods to a certain extent and provide an intuitive data basis for the development of high-quality and low-cost low alloy steels.

An improved deep forest model for forecast the outdoor atmospheric corrosion rate of low-alloy steels

The paper proposes a new deep structure model,called Densely Connected Cascade Forest-Weighted K Nearest Neighbors(DCCF-WKNNs),to implement the corrosion data modelling and corrosion knowledgemining.Firstly,we collect 409 outdoor atmospheric corrosion samples of low-alloy steels as experiment datasets.Then,we give the proposed methods process,including random forests-K nearest neighbors(RF-WKNNs)and DCCF-WKNNs.Finally,we use the collected datasets to verify the performance of the proposed method.The results show that compared with commonly used and advanced machine-learning algorithms such as artificial neural network(ANN),support vector regression(SVR),random forests(RF),and cascade forests(cForest),the proposed method can obtain the best prediction results.In addition,the method can predict the corrosion rates with variations of any one single environmental variable,like pH,temperature,relative humidity,SO2,rainfall or Cl-.By this way,the threshold of each variable,upon which the corrosion rate may have a large change,can be further obtained.