Yubo Yuan

A study on effectiveness of extreme learning machine

Abstract Extreme learning machine (ELM), proposed by Huang et al., has been shown a promising learning algorithm for single-hidden layer feedforward neural networks (SLFNs). Nevertheless, because of the random choice of input weights and biases, the ELM algorithm sometimes makes the hidden layer output matrix H of SLFN not full column rank, which lowers the effectiveness of ELM. This paper discusses the effectiveness of ELM and proposes an improved algorithm called EELM that makes a proper selection of the input weights and bias before calculating the output weights, which ensures the full column rank of H in theory. This improves to some extend the learning rate (testing accuracy, prediction accuracy, learning time) and the robustness property of the networks. The experimental results based on both the benchmark function approximation and real-world problems including classification and regression applications show the good performances of EELM.

A matrix algorithm for mining association rules

Finding association rules is an important data mining problem and can be derived based on mining large frequent candidate sets. In this paper, a new algorithm for efficient generating large frequent candidate sets is proposed, which is called Matrix Algorithm. The algorithm generates a matrix which entries 1 or 0 by passing over the cruel database only once, and then the frequent candidate sets are obtained from the resulting matrix. Finally association rules are mined from the frequent candidate sets. Numerical experiments and comparison with the Apriori Algorithm are made on 4 randomly generated test problems with small, middle and large sizes. Experiments results confirm that the proposed algorithm is more effective than Apriori Algorithm.

A polynomial smooth support vector machine for classification

A new polynomial smooth method for solving the support vector machine (SVM) is presented in this paper. It is called the polynomial smooth support vector machine (PSSVM). BFGS method and Newton-Armijo method are applied to solve the PSSVM. Numerical experiments confirm that PSSVM is more effective than SVM.

Real-time hand gesture recognition using finger segmentation.

Hand gesture recognition is very significant for human-computer interaction. In this work, we present a novel real-time method for hand gesture recognition. In our framework, the hand region is extracted from the background with the background subtraction method. Then, the palm and fingers are segmented so as to detect and recognize the fingers. Finally, a rule classifier is applied to predict the labels of hand gestures. The experiments on the data set of 1300 images show that our method performs well and is highly efficient. Moreover, our method shows better performance than a state-of-art method on another data set of hand gestures.

Forecasting the movement direction of exchange rate with polynomial smooth support vector machine

Abstract It is a very interesting topic to forecast the movement direction of financial time series by machine learning methods. Among these machine learning methods, support vector machine (SVM) is the most effective and intelligent one. A new learning model is presented in this paper, called the polynomial smooth support vector machine (PSSVM). After being solved by Broyden–Fletcher–Goldfarb–Shanno (BFGS) method, optimal forecasting parameters are obtained. The exchange rate movement direction of RMB (Chinese renminbi) vs USD (United States Dollars) is investigated. Six indexes of Dow Jones China Index Series are used as the input. 4 sections with 180 time experiments have been completed. Many results show that the proposed learning model is effective and powerful.

Multilabel image annotation based on double-layer PLSA model.

Due to the semantic gap between visual features and semantic concepts, automatic image annotation has become a difficult issue in computer vision recently. We propose a new image multilabel annotation method based on double-layer probabilistic latent semantic analysis (PLSA) in this paper. The new double-layer PLSA model is constructed to bridge the low-level visual features and high-level semantic concepts of images for effective image understanding. The low-level features of images are represented as visual words by Bag-of-Words model; latent semantic topics are obtained by the first layer PLSA from two aspects of visual and texture, respectively. Furthermore, we adopt the second layer PLSA to fuse the visual and texture latent semantic topics and achieve a top-layer latent semantic topic. By the double-layer PLSA, the relationships between visual features and semantic concepts of images are established, and we can predict the labels of new images by their low-level features. Experimental results demonstrate that our automatic image annotation model based on double-layer PLSA can achieve promising performance for labeling and outperform previous methods on standard Corel dataset.

Social Network Analysis in Multiple Social Networks Data for Criminal Group Discovery

A criminal network is a kind of social networks with both secrecy and efficiency. The hidden knowledge in criminal networks can be regarded as an important indicators for criminal investigations which can help finding the criminals relationship and identifying suspects. However such criminal network analysis has not been studied well in an applied way and remains primarily a manual process. To assist investigators to find criminals relationship, criminal leader, and identify suspicious guys of a conspiracy, we built a comprehensive indicator model using methods developed in the field of Social Network Analysis (SNA). A simulation is done on a large office from open source reports and the ranked list with respect to comprehensive indicator indicates that we provide a reasonable ranking based on the proposed model.

Structure-aware image inpainting using patch scale optimization

Local structure multiplier in the priority function ensures the structure continuity.Color and space distance take into consideration in path searching process.Find the patch with optimized scale between under each candidate scale. Image inpainting is widely used in many image processing applications such as image stitching, image editing and object removal. The main challenge stems from producing visually plausible results after reconstruction. Most of the image inpainting algorithms cannot maintain structure continuity and texture consistency precisely. To address this problem, we propose a robust exemplar-based inpainting algorithm. Firstly, we present local structure multiplier to contain sufficient structure information in the priority function which ensures the structure continuity. Secondly, we combine color feature and space distance between two patches to search for the optimized patch to avoid texture inconsistency. At last, we calculate the average pixel difference between two patches under each candidate scale, we select the scale which the minimal average pixel difference is to be the optimized scale. We copy the target patch with the optimized patch. Extensive experiments show the effectiveness of the proposed method.

Representation of image content based on RoI-BoW

A model named as RoI-BoW is proposed, which is effective in image retrieval.Influence of different scale segmentation on image content representation is studied.A filtering operator is suggested to find the most important key points out. Representation of image content is an important part of image annotation and retrieval, and it has become a hot issue in computer vision. As an efficient and accurate image content representation model, bag-of-words (BoW) has attracted more attention in recent years. After segmentation, BoW treats all of the image regions equally. In fact, some regions of image are more important than others in image retrieval, such as salient object or region of interest. In this paper, a novel region of interest based bag-of-words model (RoI-BoW) for image representation is proposed. At first, the difference of Gaussian (DoG) is adopted to find key points in an image and generates different size grid as RoI to construct visual words by the BoW model. Furthermore, we analyze the influence of different size segmentation on image content representation by content based image retrieval. Experiments on Corel 5K verify the effectiveness of RoI-BoW on image content representation, and prove that RoI-BoW outperforms the BoW model significantly. Moreover, amounts of experiments illustrate the influence of different size segmentation on image representation based on the Bow model and RoI-BoW model respectively. This work is helpful to choose appropriate grid size in different situations when representing image content, and meaningful to image classification and retrieval.

A three-dimensional display for big data sets

Facing with high dimensional information in fields of Science, Technology and Commerce, users need effective visualization tools to find more useful information. For big data sets, it is very difficult to get useful information because the dimension is too large for a practical solution. This paper proposes a 3-D visualization method for big data sets. First of all, we employed the K-means clustering method to get the basic vectors. Then, we use these vectors to construct the reduction mapping. Finally, we get the three dimensional display for a sample point. To verify the feasibility of this method, we perform experiment on some well-known databases such as iris, wine and a large data set: Pendigits. The results are favorable. According to the 3-D display results, we can also get messages like classification, outliers, and classification level when given the level standards.