Recently, general-purpose computation on graphics processing units (GPGPU) has become an increasingly popular field of study as graphics processing units (GPUs) continue to be proposed as high performance and relatively low cost implementation platforms for scientific computing applications. Among these applications figure astrophysical N-body simulations, which form one of the most challenging problems in computational science. However, in most reported studies, a simple O(N 2 ) algorithm was used for GPGPUs, and the resulting performances were not observed to be better than those of conventional CPUs that were based on more optimized O(N log N) algorithms such as the tree algorithm or the particle-particle particle-mesh algorithm. Because of the difficulty in getting efficient implementations of such algorithms on GPUs, a GPU cluster had no practical advantage over general-purpose PC clusters for N-body simulations. In this paper, we report a new method for efficient parallel implementation of the tree algorithm on GPUs. Our novel tree code allows the realization of an N-body simulation on a GPU cluster at a much higher performance than that on general PC clusters. We practically performed a cosmological simulation with 562 million particles on a GPU cluster using 128 NVIDIA GeForce 8800GTS GPUs at an overall cost of 168 172 $. We obtained a sustained performance of 20.1 Tflops, which when normalized against a general-purpose CPU implementation leads to a performance of 8.50 Tflops. The achieved cost/performance was hence a mere $19.8 /Gflops which shows the high competitiveness of GPGPUs.

Successful research collaborations may facilitate major outcomes in science and their applications. Thus, identifying effective collaborators may be a key factor that affects success. However, it is very difficult to identify potential collaborators and it is particularly difficult for young researchers who have less knowledge about other researchers and experts in their research domain. This study introduces and defines the problem of collaborator recommendation for 'isolated' researchers who have no links with others in coauthor networks. Existing approaches such as link-based and contentbased methods may not be suitable for isolated researchers because of their lack of links and content information. Thus, we propose a new approach that uses additional information as new features to make recommendations, i.e., the strength of the relationship between organizations, the importance rating, and the activity scores of researchers. We also propose a new method for evaluating the quality of collaborator recommendations. We performed experiments by crawling publications from the Microsoft Academic Search website. The metadata were extracted from these publications, including the year, authors, organizational affiliations of authors, citations, and references. The metadata from publications between 2001 and 2005 were used as the training data while those from 2006 to 2011 were used for validation. The experimental results demonstrated the effectiveness and efficiency of our proposed approach.

Abstract. This paper introduces a novel approach for large-scale unsupervised segmentation of bibliographic elements. Our problem is to segment a word token sequence representing a citation into subsequences each corresponding to a different bibliographic element, e.g. authors, paper title, journal name, publication year, etc. Obviously, each bibliographic element should be represented by contiguous word tokens. We call this constraint contiguity constraint. Therefore, we should infer a sequence of assignments of word tokens to bibliographic elements so that this constraint is satisfied. Many HMM-based methods solve this problem by prescribing fixed transition patterns among bibliographic elements. In this paper, we use generalized Mallows models (GMM) in a Bayesian multi-topic model, effectively applied to document structure learning by Chen et al. [4], and infer a permutation of latent topics each of which can be interpreted as one among the bibliographic elements. According to the inferred permutation, we arrange the order of the draws from a multinomial distribution defined over topics. In this manner, we can obtain an ordered sequence of topic assignments satisfying contiguity constraint. We do not need to prescribe any transition patterns among bibliographic elements. We only need to specify the number of bibliographic elements. However, the method proposed by Chen et al. works for our problem only after introducing modification. The main contribution of this paper is to propose strategies to make their method work also for our problem.

Abstract. In this paper, we propose an acceleration of collapsed variational Bayesian (CVB) inference for latent Dirichlet allocation (LDA) by using Nvidia CUDA compatible devices. While LDA is an efficient Bayesian multi-topic document model, it requires complicated computations for parameter estimation in comparison with other simpler document models, e.g. probabilistic latent semantic indexing, etc. Therefore, we accelerate CVB inference, an efficient deterministic inference method for LDA, with Nvidia CUDA. In the evaluation experiments, we used a set of 50,000 documents and a set of 10,000 images. We could obtain inference results comparable to sequential CVB inference.

Results of queries by personal names often contain documents related to several people because of the namesake problem. In order to differentiate documents related to different people, an effective method is needed to measure document similarities and to find documents related to the same person. Some previous researchers have used the vector space model or have tried to extract common named entities for measuring similarities. We propose a new method that uses Web directories as a knowledge base to find shared contexts in document pairs and uses the measurement of shared contexts to determine similarities between document pairs. Experimental results show that our proposed method outperforms the vector space model method and the named entity recognition method.