This paper proposed a complex ontology evolution based method of extracting data, and also completely designed an extraction system, which consists of four important components: Resolver, Extractor, Consolidator and t...
详细信息
This paper proposed a complex ontology evolution based method of extracting data, and also completely designed an extraction system, which consists of four important components: Resolver, Extractor, Consolidator and the ontology construction components. The system gives priority to the construction of mini-ontology. When the user submits query keywords to the deep web query interface, the returned result will pass through the prior three components;after that, the final execution result will be returned to user in a unified form. This paper adopted an extraction method that is different from the general ontology extraction. More specifically, the ontology used in extraction here is dynamic evolution, which can adapt various data source better. Experimental results proved that this method could effectively extract the data in the query result pages.
Large scale terrain visualization with high- resolution has an increasing demand in many research fields. To realize the efficient rendering of terrain, this paper presents an out-of-core terrain visualization method ...
详细信息
Large scale terrain visualization with high- resolution has an increasing demand in many research fields. To realize the efficient rendering of terrain, this paper presents an out-of-core terrain visualization method based on multi-resolution storage techniques. In external memory, the terrain data set is subdivided from top to bottom to build a multi-resolution hierarchical structure based on a quad-tree. The hierarchical structure can decimate the elevation data that must be loaded into internal memory. Thus it can improve the efficiency of I/O access greatly. Moreover, in order to implement rapid data retrieval of the real time terrain flyover, an efficient indexing algorithm is proposed, in which those nodes in the hierarchical structure will be divided into several clusters in terms of the similarities of static error and the closed space constraint. In addition, a method for crack-free is also proposed here. The comprehensive experiment conducted on the GTOP30 data set shows that this approach outperforms the Block and the Hierarchy algorithms in the both ways of efficiency and simplification ratio.
Multi-agent reinforcement learning (MARL) algorithms have achieved great breakthroughs in many aspects. The MARL algorithms can learn effective policies in ideal simulation environments. But different from the ideal s...
详细信息
作者:
Wei DuZhongbo CaoYan WangEnrico BlanzieriChen ZhangYanchun LiangCollege of Mathematics
Jilin University Changchun 130012 China Department of Information and Communication Technology University of Trento Povo 38050 Italy College of Computer Science and Technology
Key Laboratory of Symbol Computation and Knowledge Engineering of the Ministry of Education Jilin University Changchun 130012 China College of Computer Science and Technology Key Laboratory of Symbol Computation and Knowledge Engineering of the Ministry of Education Jilin University Changchun 130012 China
Ontologies, seen as effective representations for sharing and reusing knowledge, have become increasingly important in biomedicine, usually focusing on taxonomic knowledge specific to a subject. Efforts have been made...
详细信息
Ontologies, seen as effective representations for sharing and reusing knowledge, have become increasingly important in biomedicine, usually focusing on taxonomic knowledge specific to a subject. Efforts have been made to uncover implicit knowledge within large biomedical ontologies by exploring semantic similarity and relatedness between concepts. However, much less attention has been paid to another potentially helpful approach: discovering implicit knowledge across multiple ontologies of different types, such as disease ontologies, symptom ontologies, and gene ontologies. In this paper, we propose a unified approach to the problem of ontology based implicit knowledge discovery - a Multi-Ontology Relatedness Model (MORM), which includes the formation of multiple related ontologies, a relatedness network and a formal inference mechanism based on set-theoretic operations. Experiments for biomedical applications have been carried out, and preliminary results show the potential value of the proposed approach for biomedical knowledge discovery.
Cancer classification and identification are major areas in medical research. DNA microarrays could provide useful information for cancer classification at the gene expression level. The number of genes in a microarra...
详细信息
Cancer classification and identification are major areas in medical research. DNA microarrays could provide useful information for cancer classification at the gene expression level. The number of genes in a microarray is always several thousands while the number of training samples always several dozens. In such case most of the machine learning models suffer from the overfitting and it is necessary to select a handful of most informative genes. An adaptive and iterative gene selection algorithm based on least squares support vector machines is proposed in this paper. The algorithm adopts sequential forward selection search scheme. The number of selected genes can be determined adaptively. The total number of genes processed by the proposed algorithm is smaller than that processed by other algorithms using support vector machines. Results of numerical experiments show that the proposed algorithm trains fast and achieves comparable performance on two well-known benchmark problems.
Generative Adversarial Networks (GANs) have achieved huge success in some unsupervised learning fields. There is no doubt that clustering takes a lot of weight in unsupervised algorithm. And in this paper, we raise th...
详细信息
Generative Adversarial Networks (GANs) have achieved huge success in some unsupervised learning fields. There is no doubt that clustering takes a lot of weight in unsupervised algorithm. And in this paper, we raise the Improved Information Maximizing Generative Adversarial Networks (IInfoGAN) algorithm for learning discriminative classifiers from unlabeled data. The basis of our method is an math function that contains the Mutual Information (MI) and Cross Entropy of the observed examples and their predicted classification category distribution, thus enhancing the robustness of the classifier to adversarial generative models. Experiments show that the interpretable representation learned by IInfoGAN is competitive with the representation learned by existing unsupervised methods.
Signed network embedding methods aim to learn vector representations of nodes in signed networks. However, existing algorithms only managed to embed networks into low-dimensional Euclidean spaces whereas many intrinsi...
详细信息
In this paper, we propose a new supervised compound learning algorithm for training our constructed approximated bivariate non-tensor product adaptive pre-wavelet neural network (APWNN). On the one hand, the linear we...
详细信息
In this paper, we propose a new supervised compound learning algorithm for training our constructed approximated bivariate non-tensor product adaptive pre-wavelet neural network (APWNN). On the one hand, the linear weights of APWNN are trained by the self-adaptive learning rate method. On the other hand an extended Kalman filter method is used to update the nonlinear parameters such as dilation parameters and translation parameters. Additionally we demonstrate the efficiency of our proposed method through a concrete example of function approximation.
There are lots of data with multidimensional attributes and spatial position information in precision agriculture applications. Based on the high dimensional spatial clustering algorithm, the agriculture breed partiti...
详细信息
There are lots of data with multidimensional attributes and spatial position information in precision agriculture applications. Based on the high dimensional spatial clustering algorithm, the agriculture breed partition method is proposed and applied in Chinese national 863 project. The data mining algorithm is performed in this way: cluster the multidimensional attributes first, then cluster the spatial position information using the former attributes clustering results. Experiments show that the two-phases clustering method is prior to the traditional single-phase clustering method. According to the clustering results, one will decide the region belongs to what kind of soil, select the breeds and the field managements.
暂无评论