Search

Web usage mining (WUM) is an important and fast developing area of web mining. Recently, some enterprises have been aware of its potentials, especially for applications in Business Intelligence (BI) and Customer Relationship Management (CRM). Therefore, it is crucial to analyze the behaviors and characteristics of web user so as to use this knowledge for advertising, targeted marketing, increasing competition ability, etc. This paper provides an analytic method, algorithm and procedure based on suggestions from literature and the authors' experiences from some practical web mining projects. Its application shows combined use of frequent sequence patterns (FSP) discovery and the characteristic analysis of user clusters can contribute to improve and optimize marketing and CRM. [ABSTRACT FROM AUTHOR]

The abstract should summarize the contents of the paper and should contain at least 70 and at most 150 words. It should be set in 9-point font size and should be inset 1.0 cm from the right and left margins. There should be two blank (10-point) lines before and after the abstract. This document is in the required format. Roadmapping, as a strategic planning tool, is attracting increasing applications in industries. By applying the principles of Interactive Planning (IP), this paper puts forward a new solution for making personal academic research roadmaps. Then this paper introduces an ongoing project that is to develop a roadmapping support system based on the solution, and gives some considerations about applying the i-System Methodology for enhancing the knowledge creation in a roadmapping process. [ABSTRACT FROM AUTHOR]

SymGrid-Par is a new framework for executing large computer algebra problems on computational Grids. We present the design of SymGrid-Par, which supports multiple computer algebra packages, and hence provides the novel possibility of composing a system using components from different packages. Orchestration of the components on the Grid is provided by a Grid-enabled parallel Haskell (GpH). We present a prototype implementation of a core component of SymGrid-Par, together with promising measurements of two programs on a modest Grid to demonstrate the feasibility of our approach. [ABSTRACT FROM AUTHOR]

In this paper, we develop an efficient implementation of two tree accumulations. In this implementation, we divide a binary tree based on the idea of m-bridges to obtain high locality, and represent local segments as serialized arrays to obtain high sequential performance. We furthermore develop a cost model for our implementation. The experiment results show good performance. [ABSTRACT FROM AUTHOR]

A new approach for fast parallel reductions on trees over distributed memory environments is proposed. By employing serialized trees as the data representation, our algorithm has a communication-efficient BSP implementation regardless of the shapes of inputs. The prototype implementation supports its real efficacy. [ABSTRACT FROM AUTHOR]

Efficient execution of multithreaded iterative numerical computations requires to carefully take into account data dependencies. This paper presents an original way to express and schedule general dataflow multithreaded computations. We propose a distributed dataflow stack implementation which efficiently supports work stealing and achieves provable performances on heterogeneous grids. It exhibits properties such as non-blocking local stack accesses and generation at runtime of optimized one-sided data communications. [ABSTRACT FROM AUTHOR]

Unlike fixed networks, mobile ad-hoc networks have diverse characteristics, such as dynamic topologies, bandwidth-constrained links, energy-constrained operation, limited physical security, etc. Due to these characteristics their security requirements are different from those of fixed networks. This paper presents a method of authenticating the nodes in the ad-hoc network and securely transmitting information in consideration of the characteristics of the mobile ad hoc network. To this end, this paper proposes a method of combining asymmetric and symmetric cryptography to enhance the efficiency of secret key management for authentication and secure channel establishment. In particular, this paper proposes a method that introduces the concept of Hamming Distance to share the secret keys between the nodes. According to this proposal, secure communication between the nodes is possible when one node manages only log2N secret keys. [ABSTRACT FROM AUTHOR]

We propose a new classification of complex networks in association with a function of networks. Networks are considered to be input-output system where the initial condition is input and the evolving dynamics is output. We study a functional relationship between the input and the output which depend on network structures. A function of network are modeled as a spin interaction system driven by Glauber dynamics with arbitrary initial conditions. Through numerical studies, we show a novel classification of networks. The results are applied to examples of real-world networks, which proved the classification to be useful for analysis of the inherent characteristics and model assumption for the real-world networks. [ABSTRACT FROM AUTHOR]

The study of networked systems has experienced a particular surge of interest in the last decade. One issue that has received a considerable amount of attention is the detection and characterization of community structure in networks, meaning the appearance of densely connected groups of vertices, with only sparser connections between groups. In this paper, we present an approach for the problem of community detection using genetic algorithm (GA) in conjunction with the method of clustering. We demonstrate that our algorithms are highly effective at discovering community structure in both computer-generated and real-world network data, and show how they can be used to shed light on the sometimes daunting complex real-world systems of scale-free network structure. [ABSTRACT FROM AUTHOR]

Thin films of nano-particles deposited on substrates are important for new technological applications. Their physical properties depend crucially on particle size and structural inhomogeneity of the deposition. To systematically investigate these dependencies, we apply graph theory and simulations of voltage-driven tunneling conduction. We implement a network model of nano-particle films, based on data of particle positions on a two-dimensional plane. Assuming that connected particles on the network are within the tunneling distance, we then implement a continuous-time electron tunneling on the network and show how the current-voltage characteristics depend on the graph structure. We visualize the conduction paths through the network which correspond to the measured currents both in random and spatially inhomogeneous films. [ABSTRACT FROM AUTHOR]

In this paper we discuss how extensible meta-data featured by virtual machines, such as JVM and CLR, can be used to specify the parallelization aspect of annotated programs. Our study focuses on annotated CLR programs written using a variant of C#; we developed a meta-program that processes these sequential programs in their binary form and generates optimized parallel code. We illustrate the techniques used in the implementation of our tool and provide some experimental results that validate the approach. [ABSTRACT FROM AUTHOR]

New insights have been gained recently into the interplay between complex network architectures and the collective dynamics of simple elements coupled through them. The usual paradigm for studying systems of this kind is a network of one-dimensional interacting logistic maps, which provides a plausible model for a variety of complex systems. In this work we consider a network of two-dimensional standard maps coupled through a scalefree tree topology. The coupling is solely in the angle coordinate and involves a time-delay: this approach is motivated by the node-to-node information flow view of the collective dynamics. We observe a rich variety of dynamical behavior including self-organized patterns: nodes synchronize in clusters each having specific motion quantities whose values are belonging to a discrete set of possible values. We conclude by studying the relationship between dynamically induced cluster-organization of the nodes and their network-structurization. [ABSTRACT FROM AUTHOR]

We suggest a network representation of dynamical maps (such as the logistic map) on the unit interval. The unit interval is partitioned into N subintervals, which are associated with 'nodes' of the network. A link from node i to j is defined as a possible transition of the dynamical map from one interval i, to another j. In this way directed networks more generally allow phasespace representations (i.e. representations of transitions between different phasespace cells), of dynamical maps defined on finite intervals. We compute the diameter of these networks as well as the average path length between any two nodes. We numerically demonstrate that these network characteristics seem to diverge at the (first) zeros of the Lyapunov exponent and could thus provide an alternative measure to detect the 'edge of chaos' in dynamical systems. [ABSTRACT FROM AUTHOR]

In this study, we propose an intelligent customer relationship management (CRM) system that uses support vector machine (SVM) ensembles to help enterprise managers effectively manage customer relationship from a risk avoidance perspective. Different from the classical CRM for retaining and targeting profitable customers, the main focus of our proposed CRM system is to identify high-risk customers for avoiding potential loss. Through experiment analysis, we find that the Bayesian-based SVM ensemble data mining model with diverse components and "choose from space" selection strategy show the best performance over the testing samples. [ABSTRACT FROM AUTHOR]

Today, clusters are very important computing resources and many computing centers manage their clusters in private networks. But parallel programs may not work in private clusters. Because hosts in private clusters are not globally reachable, hosts behind different private clusters cannot be reached directly in order to communicate. It will certainly be a huge loss of resources if private clusters are excluded from the computing due to this reason. There has been much research on this issue, but most of them concentrate on user-level relaying because it is a general and easily-implementable solution. However, even well-implemented, user-level solutions result in much longer communication latency than kernel-level solutions. This paper adopted a novel kernel-level solution and applied it to MPICH-G2. Our scheme is generally applicable, simple and efficient. The experimental results show that our scheme incurs very little overhead except when small messages are transmitted. That is, it supports a more universal computing environment by including private clusters with remarkably little overhead. [ABSTRACT FROM AUTHOR]

When we want to measure parameters of the Modulation Transfer Function (MTF) from remote sensing image directly, the sharp edges are usually used as targets. But for noise, blur and the complexity of the images, fully automatic locating the expected edge is still an unsolved problem. This paper improves the semi-auto edge extraction algorithm of live-wire [1] and introduces it into the knife-edge method [8] of MTF measuring in remote sensing image. Live-wire segmentation is a novel interactive algorithm for efficient, accurate, and reproducible boundary extraction that requires minimal user input with a mouse. The image is transformed into a weighted graph with variety restrictions. Edge searching is based on dynamic programming of Dijkstra's algorithm [5]. Optimal boundaries are computed and selected at interactive rates as the user moves the mouse starting from a manually specified seed point. In this paper, a promoted model of live-wire is proposed to measuring the on orbit Modulation Transfer Function for high spatial resolution imaging satellites. We add the no-linear diffusion filter in the local cost function to ensure the accurateness of the extraction of edges. It can both de-noise and do not affect the shape of the edges when we extracting the edges, so that the calculation of the MTF is more reasonable and precise. [ABSTRACT FROM AUTHOR]

In the GIS(Geospatial Information System) Web Services, SOAP/MTOM shows best performance when transferring large size of data between services. SOAP/MTOM uses XOP for message optimizing method, so data serialization and deserialization time is reduced. Additionally, integrating Ajax(Asynchronous JavaScript and XML) approach into GIS visualization Web Services have performance enhancement, because it provides more interactive user experience. Ajax, one of the technologies of "Web 2.0", is appeared after Dot Com bubble is occurred. The intent is to make web pages feel more responsive by exchanging small amounts of data with the server behind the scenes, so that the entire web page does not have to be reloaded each time the user requests a change. [ABSTRACT FROM AUTHOR]

Local spatial interaction (i.e. neighborhood interaction) between land-use types is an important component in Cellular Automata -based urban geosimulation models. Herein a new method based on the integration of Tobler's First Law of Geography with Reilly's gravity model and coupled with logistical regression approach is proposed to model and calibrate the neighborhood interaction. This method is embedded into a constrained CA model to simulate the spatial process of urban growth in the Tokyo metropolitan area. The results indicate that this method captures the main characteristics of neighborhood interactions in the spatial process of urban growth. Further, this method provides an alternative and extensive approach to present local spatial interactions for "bottom-up" urban models. [ABSTRACT FROM AUTHOR]

The signal at the top of the atmosphere will certainly contain information about both the surface and the atmosphere. To derive the geophysical parameters from satellites remote sensing images, the atmospheric effects must be decoupled. Aerosol Optical Thickness (AOT), an important aerosol optical property, should be correctly determined to remove the atmospheric effect. The retrieval process is great time-consuming and the EMS memory required is too large for a personal computing to run efficiently. Therefore, to facilitate the process smoothly, SYNTAM model is used to retrieve AOT over a wide range of land including China and one European area from MODIS data on the Remote Sensing Information Service Grid Node (RSIN, http://www.tgp.ac.cn) deployed at Institute of Remote Sensing Applications, Chinese Academy of Sciences. AOT retrieval results show that the RSIN Grid service is high efficient and has the potential to be applied to the remote sensing parameter inversion. [ABSTRACT FROM AUTHOR]

The problem of spatial configuration information retrieval is a Constraint Satisfaction Problem (CSP), which can be solved using traditional CSP algorithms. But the spatial data can be reorganized using index techniques like R-tree and the spatial data are approximated by their Minimum Bounding Rectangles (MBRs), so the spatial configuration information retrieval is actually based on the MBRs and some special techniques can be studied. This paper studies the mapping relationships among the spatial relations for real spatial objects, the corresponding spatial relations for their MBRs and the corresponding spatial relations between the intermediate nodes and the MBRs in R-tree. [ABSTRACT FROM AUTHOR]

In this paper a possibility of neural network application to data processing in magnetotelluric method is studied. The modular neural system, consisting of three multi-layer neural networks, is used for obtaining geoelectric model of lithosphere basing on amplitude and phase MT curves. Cases of two and three flat lithosphere layers are considered. [ABSTRACT FROM AUTHOR]

Geocomputation has a long tradition in dealing with fuzzyness in different contexts, most notably in the challenges created by the representation of geographic space in digital form. Geocomputation tools should be able to address the imminent continuous nature of geo phenomena, and its accompanying fuzzyness. Fuzzy Set Theory allows partial memberships of entities to concepts with non-crisp boundaries. In general, the application of fuzzy methods is distance-based and for that reason is insensible to changes in density. In this paper a new method for defining density-based fuzzy membership functions is proposed. The method automatically determines fuzzy membership coefficients based on the distribution density of data. The density estimation is done using a Self-Organizing Map (SOM). The proposed method can be used to accurately describe clusters of data which are not well characterized using distance methods. We show the advantage of the proposed method over traditional distance-based membership functions. [ABSTRACT FROM AUTHOR]

In this article we describe a modelbase system used in Remote Sensing Information Analysis and Service Grid Node (RISN) at Institute of Remote Sensing Applications (IRSA), Chinese Academy of Sciences (CAS). The goal of the Node is to make good use of physically distributed resources in the field of remote sensing science such as data, models and algorithms, and computing resource left unused on Internet. With the modelbase system, we can organize and manage models better and make full use of models. With this system, we update both local and remote models easily, and we can also add remote modelbase into our modelbase management system. At the same time, we provide interfaces to access and run models from our node. Implementing it, we use Oracle to organize and manage models, use java language to connect with Oracle database and run models on Condor platform. [ABSTRACT FROM AUTHOR]

Interpolating scattered data points is a problem of wide ranging interest. One of the most popular interpolation methods in geostatistics is ordinary kriging. The price for its statistical optimality is that the estimator is computationally very expensive. We demonstrate the space and time efficiency and accuracy of approximating ordinary kriging through the use of covariance tapering combined with iterative methods. [ABSTRACT FROM AUTHOR]

In this study, we use least square support vector machines (LSSVM) to construct a credit scoring model and introduce conjoint analysis technique to analyze the relative importance of each input feature for making the decision in the model. A test based on a real-world credit dataset shows that the proposed model has good classification accuracy and can help explain the decision. Hence, it is an alternative model for credit scoring tasks. [ABSTRACT FROM AUTHOR]

Previous spatial characterization methods does not analyze well spatial regions for a given query since it only focus on characterization for user's pre-selected area and without consideration of spatial density. Consequently, the effectiveness of characterization knowledge is decreased in these methods. In this paper, we propose a new hybrid spatial characterization system combining the density-based clustering module which consists of the attribute removal generalization and the concept hierarchy generalization. The proposed method can generate characteristic rule and apply density-based clustering to enhance the effectiveness of generated rules. [ABSTRACT FROM AUTHOR]

In this paper, we designed and implemented a Conversion System for GML documents, and evaluated the performance of the system. The conversion system that uses the BXML-based binary encoding specification can directly display geographic information from BXML documents and can convert GML documents to BXML documents without loss of information, and vice versa. BXML is generally more effective in scanning cost and space requirement than GML and coordinate values in the form of BXML can be used directly without conversion. [ABSTRACT FROM AUTHOR]

In the increasingly competitive credit industry, one of the most interesting and challenging problems is how to manage existing customers. Behavior scoring models have been widely used by financial institutions to forecast customer's future credit performance. In this paper, a hybrid GA+SVM model, which uses genetic algorithm (GA) to search the promising subsets of features and multi-class support vector machines (SVM) to make behavior scoring prediction, is presented. A real life credit data set in a major Chinese commercial bank is selected as the experimental data to compare the classification accuracy rate with other traditional behavior scoring models. The experimental results show that GA+SVM can obtain better performance than other models. [ABSTRACT FROM AUTHOR]

This article presents the way how creditor can predict the trends of debtors financial standing. We propose the model for forecasting changes of financial standings. Model is based on the Self-organizing maps as a tool for prediction, grouping and visualization of large amount of data. Inputs for training of SOM are financial ratios calculated according any discriminate bankruptcy model. Supervised neural network lets automatically increase accuracy of performance via changing of weights of ratios. [ABSTRACT FROM AUTHOR]

Component-wise Iterative Optimization (CIO) is a method of dealing with a large data in the OLAP applications, which can be treated as the enhancement of the traditional batch version methods such as least squares. The salient feature of the method is to process transactions one by one, optimizes estimates iteratively for each parameter over the given objective function, and update models on the fly. A new learning algorithm can be proposed when applying CIO to feed-forward neural networks with a single hidden layer. It incorporates the internal structure of feed-forward neural networks with a single hidden layer by applying the algorithm CIO in closed-form expressions to update weights between the output layer and the hidden layer. Its optimally computational property is a natural consequence inherited from the property of the algorithm CIO and is also demonstrated in an illustrative example. [ABSTRACT FROM AUTHOR]

Fuzzy modeling provides a very useful tool to deal with human vagueness in describing scales of value. This study examines the relative error in decision tree models applied to a real set of credit card data used in the literature, comparing crisp models with fuzzy decision trees as applied by See5, and as obtained by categorization of data. The impact of ordinal data is also tested. Modifying continuous data was expected to degrade model accuracy, but was expected to be more robust with respect to human understanding. The degree of accuracy lost by See5 fuzzification was minimal (in fact more accurate in terms of total error), although bad error was worse. Categorization of data yielded greater inaccuracy. However, both treatments are still useful if they better reflect human understanding. An additional conclusion is that when categorizing data, care should be taken in setting categorical limits. [ABSTRACT FROM AUTHOR]

Predictive models are generally fitted directly from the original noisy data. It is well known that noise can seriously limit the prediction performance on time series. In this study, we apply the nonlinear noise reduction methods to the problem of foreign exchange rates forecasting with neural networks (NNs). The experiment results show that the nonlinear noise reduction methods can improve the prediction performance of NNs. Based on the modified Diebold-Mariano test, the improvement is not statistically significant in most cases. We may need more effective nonlinear noise reduction methods to improve prediction performance further. On the other hand, it indicates that NNs are particularly well appropriate to find underlying relationship in the environment characterized by complex, noisy, irrelevant or partial information. We also find that the nonlinear noise reduction methods work more effectively when the foreign exchange rates are more volatile. [ABSTRACT FROM AUTHOR]

This paper aims to provide a scheme for effectively and efficiently finding an approximately optimal example size with respect to a given dataset when using Multiple-Criteria Linear Programming (MCLP) classification method. By integrating techniques of both progressive sampling and classification committee, it designs a dynamic classification committee scheme for MCLP. The experimental results have shown that our idea is feasible and the scheme is effective and efficient for exploring an approximately optimal sample size. The empirical results also help us to further investigate some general specialties of MCLP, such as the more general function expressions reflecting the relationship between accuracy and sample size, and between computing cost and sample size. [ABSTRACT FROM AUTHOR]

Operational risk has become increasingly important topics for Chinese Commercial Banks in recent years. Considering the huge operational losses, Extreme value theory (EVT) has been recognized as a useful tool in analyzing such data. In this paper, we presented an ERM-POT (Exponential Regression Model and the Peaks-Over-Threshold) method to measure the operational risk. The ERM-POT method can lead to bias-corrected estimators and techniques for optimal threshold selections. And the experiment results show that the method is reasonable. [ABSTRACT FROM AUTHOR]

There are a number of different attributes to describe ontology of proteins such as protein structure, biomolecular interaction, cellular location, and protein domains which represent the basic evolutionary units that form protein. In this paper, we propose a mathematical approach, formal concept analysis (FCA), which toward abstracting from attribute-based object descriptions. Based on this theory, we present extended version of algorithm, tripartite lattice, to compute a concept lattice. By analyzing tripartite lattice, we attempt to extract proteins, which are related to domains and gene ontology (GO) terms from bottom nodes to the top of lattice. In summary, using tripartite lattices, we classified proteins from protein domain composition with their describing gene ontology (GO) terms. [ABSTRACT FROM AUTHOR]

This study proposes a novel neural-network-based fuzzy group forecasting model for foreign exchange rates prediction. In the proposed model, some single neural network models are first used as predictors for foreign exchange rates prediction. Then these single prediction results produced by each single neural network models are fuzzified into some fuzzy prediction representations. Subsequently, these fuzzified prediction representations are aggregated into a fuzzy group consensus, i.e., aggregated fuzzy prediction representation. Finally, the aggregated prediction representation is defuzzified into a crisp value as the final prediction results. For illustration and testing purposes, a typical numerical example and three typical foreign exchange rates prediction experiments are presented. Experimental results reveal that the proposed model can significantly improve the prediction performance for foreign exchange rates. [ABSTRACT FROM AUTHOR]

A fast and accurate method for pricing early exercise options in computational finance is presented in this paper. The main idea is to reformulate the well-known risk-neutral valuation formula by recognizing that it is a convolution. This novel pricing method, which we name the ‘CONV' method for short, is applicable to a wide variety of payoffs and only requires the knowledge of the characteristic function of the model. As such the method is applicable within exponentially Lévy models, including the exponentially affine jump-diffusion models. For an M-times exercisable Bermudan option, the overall complexity is O(MNlog(N)) with N grid points used to discretize the price of the underlying asset. It is also shown that American options can be very efficiently computed by combining Richardson extrapolation to the CONV method. [ABSTRACT FROM AUTHOR]

In a risk management of derivative securities, Greeks, i.e. sensitivity coefficients, are important measures of market risk to evaluate the impact of misspecification of some stochastic model on the expected payoff function. We investigate a new computing method for Greeks based on Malliavin calculus without resort to a direct differentiation of the complex payoff functions. As a result, a new relation between Γ and Δ is obtained for the Asian option. [ABSTRACT FROM AUTHOR]

Recent studies have revealed that emerging modern machine learning techniques are advantageous to statistical models for credit risk evaluation, such as SVM. In this study, we discuss the applications of the support vector machine with mixture of kernel to design a credit evaluation system, which can discriminate good creditors from bad ones. Differing from the standard SVM, the SVM-MK uses the 1-norm based object function and adopts the convex combinations of single feature basic kernels. Only a linear programming problem needs to be resolved and it greatly reduces the computational costs. More important, it is a transparent model and the optimal feature subset can be obtained automatically. A real life credit dataset from a US commercial bank is used to demonstrate the good performance of the SVM-MK. [ABSTRACT FROM AUTHOR]

Various interactions of human immunodeficiency virus type 1 (HIV-1) proteins with those of the host cell are known, but a large-scale network encompassing all known interactions of HIV-1 proteins and host cell proteins have not been analyzed or visualized. This is partly because large-scale interaction data are not readily available from public databases and individual literatures report small-scale interactions. NCBI recently released a database of all known interactions of HIV-1 proteins and human proteins. Now, a challenging task is to analyze all the interactions in a systematic way and to identify biologically meaningful interaction groups or patterns. This paper presents the development of a web-based system (http://hiv1.hpid.org) for visualizing and analyzing the large-scale interactions between HIV-1 and human proteins and the comparative analysis of the interactions. The whole interaction network contains 1,768 interactions of 65 different types with 810 human proteins. The analysis identified several interesting interaction groups from the comparative analysis of the interaction networks. [ABSTRACT FROM AUTHOR]

This is a systematic gas phase study of the gauche and trans conformers of 1-bromo-2-chloroethane. The methods used are second order Moller-Plesset theory (MP2) and density functional theory (DFT). The basis set used is 6-311++G(d,p) for all atoms. The functional used for DFT method is B3LYP. G2/MP2 and CCSD(T) calculations have also been carried out using MP2 optimised structure. The results indicate that there is more preference for the trans conformer. The energy difference between the trans and gauche conformers (ΔEtg) and related rotational thermodynamics are reported. The MP2/6- 311++G(d,p) energy difference (ΔEtg) for 1-bromo-2-chloroethane is 7.08 kJ/mol. The conformers of 1-bromo-2-chloroethane have also been subjected to vibrational analysis. This study has also been extended to investigate solvent effects using the Self-Consistent Reaction Field methods. The structures of the conformers are not much affected by the solvents but the energy difference (ΔEtg) decreases with increasing polarity of the solvent. The results from the different theoretical methods are in good agreement. [ABSTRACT FROM AUTHOR]

Geometries of ground states of mer-tris(8-hydroxyquinoli- nato)metal (Mq3, M=Al3 + , Ga3 + , In3 + , Tl3 + ) are optimized by B3LYP/6-31G(d) methods. The bonding interactions between the metal fragment Mq2 and each single ligand q have been analyzed with the energy decomposition scheme. The calculated results suggest that the HOMO and LUMO distribution fashion can be simply traced back to the lowest electrostatic attractive and highest orbital interaction energy between fragments A-quinolate ligand and Mq2 and B-ligand and Mq2, respectively. [ABSTRACT FROM AUTHOR]

To regulate the switch effect induced by NH− anion substitution at the H8 site of the guanine-cytosine (GC) watson-crick pair and investigate the actual state in the solution, 1-9 water molecules attached GNH −C derivatives are probed in the present paper. Results show that the switch effect, i.e., deformation effect of hydrogen bonding between GNH − and C pair reduces step-wise as the number of discrete water molecules around the NH− group increases. Thus one can control the switch action of GNH −C by regulating the number of water molecules around the NH− group. [ABSTRACT FROM AUTHOR]

We present the lowest energy conformations for several large benchmark problems in 3D HP model. We found these solutions with MC and genetic algorithm using new move set. The new move set including rotation and mirror reflection is suitable for use in protein folding simulation. Experiment results show that new move set can find these best solutions in less time on average and dramatically superior to that of the commonly used move set. [ABSTRACT FROM AUTHOR]

With the availability of a growing collection of biological pathway datasets, there is an increasing need for an efficient repository with the powerful search capability to facility large scaled analyses of biological pathway. A novel data store was designed based upon the BioPAX object models, and implemented with the emerging XML database technology. Our approach significantly reduces the system complexity while maintaining powerful search functionalities. [ABSTRACT FROM AUTHOR]

Long-standing theoretical concerns about mutation accumulation within the human population can now be addressed with numerical simulation. We apply a biologically realistic forward-time population genetics program to study human mutation accumulation under a wide-range of circumstances. Using realistic estimates for the relevant biological parameters, we investigate the rate of mutation accumulation, the distribution of the fitness effects of the accumulating mutations, and the overall effect on mean genotypic fitness. Our numerical simulations consistently show that deleterious mutations accumulate linearly across a large portion of the relevant parameter space. This appears to be primarily due to the predominance of nearly-neutral mutations. The problem of mutation accumulation becomes severe when mutation rates are high. Numerical simulations strongly support earlier theoretical and mathematical studies indicating that human mutation accumulation is a serious concern. Our simulations indicate that reduction of mutation rate is the most effective means for addressing this problem. [ABSTRACT FROM AUTHOR]

Since most knowledge about protein-protein interactions still hides in biological publications, there is an increasing focus on automatically extracting information from the vast amount of biological literature. Existing approaches can be broadly categorized as rule-based or statistically-based. Rule-based approaches require heavy manual effort. On the other hand, statistically-based approaches require large-scale, richly annotated corpora in order to reliably estimate model parameters. This is normally difficult to obtain in practical applications. We have proposed a hidden vector state (HVS) model for protein-protein interactions extraction. The HVS model is an extension of the basic discrete Markov model in which context is encoded as a stack-oriented state vector. State transitions are factored into a stack shift operation similar to those of a push-down automaton followed by the push of a new preterminal category label. In this paper, we propose a novel approach based on the k-nearest-neighbors classifier to automatically train the HVS model from un-annotated data. Experimental results show the improved performance over the baseline system with the HVS model trained from a small amount of the annotated data. [ABSTRACT FROM AUTHOR]

In this paper a Simulated Annealing algorithm (SA) for solving the Protein Folding Problem (PFP) is presented. This algorithm has two phases: quenching and annealing. The first phase is applied at very high temperatures and the annealing phase is applied at high and low temperatures. The temperature during the quenching phase is decreased by an exponential function. We run through an efficient analytical method to tune the algorithm parameters. This method allows the change of the temperature in accordance with solution quality, which can save large amounts of execution time for PFP. [ABSTRACT FROM AUTHOR]

An improved tool to explore the origin of human immunoglobulin heavy chains, named Ab-origin, has been developed. It can analyze in detail the V-D-J joints from the rearranged sequence by searching against germline databases. In addition to the known information about antibody recombination, appropriate score system and restriction of searching location are also incorporated to improve computing performance. When compared with a newly developed software SoDA, Ab-origin performed much better in both accuracy and stability, with 2, 7 and 1 percent higher for V, D and J respectively. Though only taking the human heavy chain for an example here, the algorithm also suits for the analysis of all immunoglobulin and TCR sequences. [ABSTRACT FROM AUTHOR]

Artificial immune recognition system (AIRS) has been convincingly proved a highly effective classifier, which has been successfully applied to pattern recognition and etc. However, there are two shortcomings that limit its further applications, one is the huge size of evolved memory cells pool, and the other is low classification accuracy. In order to overcome these limitations, a supervised artificial immune classifier, UCAIS, is presented. The implementation of UCAIS includes: the first is to create a pool of memory cells. Then, B-cell population is evolved and the memory cells pool is updated until the stopping criterion is met. Finally, classification is accomplished by majority vote of the k nearest memory cells. Compared with AIRS, UCAIS not only reduces the huge size of evolved memory cells pool, but also improves the classification accuracy on the four famous datasets, the Iris dataset, the Ionosphere dataset, the Diabetes dataset, and the Sonar dataset. [ABSTRACT FROM AUTHOR]