Fuzzy ARTMAP neural network for seafloor classification from multibeam sonar data

This paper presents a seafloor classification method of multibeam sonar data, based on the use of Adaptive Resonance Theory （ART） neural networks. A general ART-based neural network, Fuzzy ARTMAP, has been proposed for seafloor classification of multibeam sonar data. An evolutionary strategy was used to generate new training samples near the cluster boundaries of the neural network, therefore the weights can be revised and refined by supervised learning. The proposed method resolves the training problem for Fuzzy ARTMAP neural networks, which are applied to seafloor classification of multibeam sonar data when there are less than adequate ground-troth samples. The results were synthetically analyzed in comparison with the standard Fuzzy ARTMAP network and a conventional Bayesian classifier. The conclusion can be drawn that Fuzzy ARTMAP neural networks combining with GA algorithms can be alternative powerful tools for seafloor classification of multibeam sonar data.     

Similarity identification method on formational interfaces and application in general granite

An instrumented drilling system can be applied for the acquisition of drilling process parameters. The system can provide continuous and huge data for geotechnical engineering. However, due to the complexity of ground strata, the variation in the drilling parameters with stratigraphical characteristics is great and the correlation between likely comparable parameters is not high, which limits the use of conventional correlation approaches in this field. How to use the data for engineering and how to get a reasonable interpretation for the relationships among the drilling parameters, as well as between a drilling parameter and formational characteristics, become a technical choke point for the development and application of the instrumented drilling system. Based on similarity criteria, the extraction of sample data and characteristics, the pretreatment of data and feature matching algorithms have been analyzed and an approach of slope coefficient searching identification has been established. A case study was carried out for the similarity between the rotational speed of the drill bit, flushing pressure, and effective thrust force graphics in general weathered granite. The result shows that the similarity coefficients between the rotational speed of the drill bit, flushing pressure, and effective thrust force are 0.72 and 0.83, respectively. Although there are differences between the distances of the graphics, the curves of both rotational speed and flushing pressure agree with the effective thrust curve in shape, which provides a possible method for the identification of various formations by use of the similarity between feature drilling parameters.     

A local geopotential model for implementation of underwater passive navigation

A main aspect of underwater passive navigation is how to identify the vehicle location on an existing gravity map, and several matching algorithms as ICCP and SITAN are the most prevalent methods that many scholars are using. In this paper, a novel algorithm that is different from matching algorithms for passive navigation is developed. The algorithm implements underwater passive navigation by directly estimating the inertial errors through Kalman filter algorithm, and the key part of this implementation is a Fourier series-based local geopotential model. Firstly, the principle of local geopotential model based on Fourier series is introduced in this paper, thus the discrete gravity anomalies data can be expressed analytically with respect to geographic coordinates to establish the observation equation required in the application of Kalman filter. Whereafter, the indicated gravity anomalies can be gotten by substituting the inertial positions to existing gravity anomalies map. Finally, the classical extended Kalman filter is introduced with the differences between measured gravity and indicated gravity used as observations to optimally estimate the errors of the inertial navigation system （INS）. This navigation algorithm is tested on simulated data with encouraging results. Although this algorithm is developed for underwater navigation using gravity data, it is equally applicable to other domains, for example vehicle navigation on magnetic or terrain data.     

A survey on routing in wireless sensor networks

One of the most important issues in wireless sensor networks is data delivery service between sensors and the data collection unit (called sink). Although sensor networks and mobile ad hoc networks are similar to some extent, they are radically distinct in many aspects. Sensor networks have many unique features, making them more challenging and need further research efforts. The existing routing protocols for sensor networks can be classified as indicator-based and indicator-free. In this survey, we make a comparative study of these protocols. Open issues and research directions are pointed out as guidelines for our future work.     

A survey on routing in wireless sensor networks

One of the most important issues in wireless sensor networks is data delivery service between sensors and the data collection unit (called sink). Although sensor networks and mobile ad hoc networks are similar to some extent, they are radically distinct in many aspects. Sensor networks have many unique features, making them more challenging and need further research efforts. The existing routing protocols for sensor networks can be classified as indicator-based and indicator-free. In this survey, we make a comparative study of these protocols. Open issues and research directions are pointed out as guidelines for our future work.     

Packet matching algorithm based on improving differential evolution

The performance of network equipments, such as firewall, router, etc., is decided by the efficiency of patch matching. It is difficult to adapt the speed of packet matching with packets linear forwarding by traditional algorithms. The purpose of this paper is to develop a novel algorithm of packet matching based on improving differential evolutionary algorithm, which also combines with classic packets matching algorithms to improve the performance of algorithm. For the sake of objectivity, the statistics method was used to compute the fitting value. Experiments showed that this new algorithm effectively improved the performance in the speed and storage space, as compared with the traditional one. For the first time, evolutionary algorithm is used to solve the network data packet forwarding, and packets can be forwarded at the linear speed. In addition, this new algorithm is universal, so it can be adapted for many equipment.     

High-Performance Broadcasting Algorithms on Cluster

In many clusters connected by high-speed communication networks, the exact structure of the underlying communication network and the latency difference between different sending and receiving pairs may be ignored when they broadcast, such as in the approach adopted by the broadcasting method in MPICH，a widely used MPI implementation. However, the underlying network cluster topologies are becoming more and more complicated and the performance of traditional broadcasting algorithms, such as MPICH‘s MPI_Bcast, is far from good. This paper analyzed the impact of communication latencies and the underlying topologies on the performance of broadcasting algorithms for multilevel clusters. A multilevel model was developed for broadcasting in clusters with complicated topologies, which divides the cluster topology into many levels based on the underlying topology. The multilevel model was used to develop a new broadcast algorithm，MLM broadcast-2 (MLMB-2), that adapts to a wide range of clusters. Comparison of the performance of the counterpart MPI operation MPI_Bcast and MLMB-2 shows that MLMB-2 outperforms MPI_Bcast by decreasing the broadcast running time by 60%-90%.     

Full-angle optical imaging of near-infrared fluorescent probes implanted in small animals

To provide a valuable experimental platform for in vivo biomedical research of small animal model with fluorescence mediated approach, we developed a whole-body near-infrared fluorescence molecular imaging system as described in this paper. This system is based on a sensitive CCD camera and has the ability to achieve 360~ full-angle source illuminations and projections capture of the targets to obtain the dense sampling by performing rotational scan. The measurement accuracy is validated from cylinder phantom experiments by the comparison between the experimental data and theoretical predictions. Finally, we also present typical in vivo images of fluorescent tube implanted into the mouse body. The results are promising and have proved the system imaging performance for macroscopic optical biomedical research.     

A Graph Drawing Algorithm for Visualizing Multivariate Categorical Data

In this paper, a new approach for visualizing multivariate categorical data is presented. The approach uses a graph to represent multivariate categorical data and draws the graph in such a way that we can identify patterns, trends and relationship within the data. A mathematical model for the graph layout problem is deduced and a spectral graph drawing algorithm for visualizing multivariate categorical data is proposed. The experiments show that the drawings by the algorithm well capture the structures of multivariate categorical data and the computing speed is fast.     

Distances Between Phylogenetic Trees： A Survey

Phylogenetic trees have been widely used in the study of evolutionary biology for representing the tree-like evolution of a collection of species. However, different data sets and different methods often lead to the construction of different phylogenetic trees for the same set of species. Therefore, comparing these trees to determine similarities or, equivalently, dissimilarities, becomes the fundamental issue. Typically, Tree Bisection and Reconnection(TBR)and Subtree Prune and Regraft(SPR) distances have been proposed to facilitate the comparison between different phylogenetic trees. In this paper, we give a survey on the aspects of computational complexity, fixed-parameter algorithms, and approximation algorithms for computing the TBR and SPR distances of phylogenetic trees.     

Inversion of oceanic chlorophyll concentrations by neural networks

Neural networks (NNs) for the inversion of chlorophyll concentrations from remote sensing reflectance measurements were designed and trained on a subset of the SeaBAM data set. The remaining SeaBAM data set was then applied to evaluating the performance of NNs and compared with those of the SeaBAM empirical algorithms. NNs achieved better inversion accuracy than the empirical algorithms in most of chlorophyll concentration range, especially in the intermediate and high chlorophyll regions and CaseⅡ waters. Systematic overestimation existed in the very low chlorophyll (<0.031 mg/m³) region, and little improvement was obtained by changing the size of the training data set.     

Recent developments in totally asymmetric simple exclusion processes with local inhomogeneity

A totally asymmetric simple exclusion process (TASEP) has become an essential tool in modeling and analyzing non-equilibrium systems.A wide variety of TASEP models have been developed that are motivated by real-world traffic,biological transport and by the dynamics of the process itself.This paper provides an overview of recent developments in TASEP with inhomogeneity.Some important generalizations and extensions of inhomogeneous TASEP models are reviewed,and several popular mean-field techniques used to analyze the inhomogeneous TASEP models are summarized.A comparison between similar TASEP models under different updating procedures is given.Phase separations in such disordered systems have been identified.The present status of the inhomogeneous TASEP models and areas for future investigations are also described.     

Link-Privacy Preserving Graph Embedding Data Publication with Adversarial Learning

The inefficient utilization of ubiquitous graph data with combinatorial structures necessitates graph embedding methods, aiming at learning a continuous vector space for the graph, which is amenable to be adopted in traditional machine learning algorithms in favor of vector representations. Graph embedding methods build an important bridge between social network analysis and data analytics, as social networks naturally generate an unprecedented volume of graph data continuously. Publishing socia...     

Automatic Data Extraction from Websites for Generating Aquatic Product Market Information

The massive web-based information resources have led to an increasing demand for effective automatic retrieval of target information for web applications. This paper introduces a web-based data extraction tool that deploys various algorithms to locate, extract and filter tabular data from HTML pages and to transform them into new web-based representations. The tool has been applied in an aquaculture web application platform for extracting and generating aquatic product market information. Results prove that this tool is very effective in extracting the required data from web pages.     

SDFA： a Uniform Model for String Matching Algorithms

String matching algorithms play an important role in computer science. However, there is no uniform mathematical model to describe these algorithms. Thus, read-head-Skippable DFA (SDFA) is put for-ward, which is an extension of two-way DFA. It is proved that SDFA is equivalent to DFA. Furthermore,SDFA is a more natural mathematical model for string matching algorithms. After that, four types of the movement of the read head of string matching are analyzed and modeled by SDFA. Finally, the SDFA model of BMA string matching algorithms is given.     

Identification model of multi-layered neural network parameters and its applications in the petroleum production

This paper creates a LM （Levenberg-Marquardt） algorithm model which is appropriate to solve the problem about weights value of feedforward neural network. On the base of this model, we provide two applications in the oilfield production. Firstly, we simulated the functional relationships between the petrophysical and electrical properties of the rock by neural networks model, and studied oil saturation. Under the precision of data is confirmed, this method can reduce the number of experiments. Secondly, we simulated the relationships between investment and income by the neural networks model, and studied invest saturation point and income growth rate. It is very significant to guide the investment decision. The research result shows that the model is suitable for the modeling and identification of nonlinear systems due to the great fit characteristic of neural network and very fast convergence speed of LM algorithm.     

Non-cooperative game for network access selection in heterogeneous integrated networks

The integration of different heterogeneous access networks is one of the remarkable characteristics of the next generation network,in which users with multi-network interface terminals can independently select access network to obtain the most desired service.A kind of unified quantification model of non-monotone quality of service （QoS） and a model of non-cooperative game between users and networks are proposed for heterogeneous network access selection.An optimal network pricing mechanism could be formulated by using a novel strategy which is used in this non-cooperative game model to balance the interests of both the users and the networks.This access network selection mechanism could select the most suitable network for users,and it also could provide the basis when formulating QoS standards in heterogeneous integrated networks.The simulation results show that this network selection decision-making algorithm can meet the users＇ demand for different levels service in different scenes and it can also avoid network congestion caused by unbalanced load.     

Finding finer functions for partially characterized proteins by protein-protein interaction networks

Based on high-throughput data, numerous algorithms have been designed to find functions of novel proteins. However, the effectiveness of such algorithms is currently limited by some fundamental factors, including (1) the low a-priori probability of novel proteins participating in a detailed function; (2) the huge false data present in high-throughput datasets; (3) the incomplete data coverage of functional classes; (4) the abundant but heterogeneous negative samples for training the algorithms; and (5) the lack of detailed functional knowledge for training algorithms. Here, for partially characterized proteins, we suggest an approach to finding their finer functions based on protein interaction sub-networks or gene expression patterns, defined in function-specific subspaces. The proposed approach can lessen the above-mentioned problems by properly defining the prediction range and functionally filtering the noisy data, and thus can efficiently find proteins’ novel functions. For thousands of yeast and human proteins partially characterized, it is able to reliably find their finer functions (e.g., the translational functions) with more than 90% precision. The predicted finer functions are highly valuable both for guiding the follow-up wet-lab validation and for providing the necessary data for training algorithms to learn other proteins.     

Indirect associations between multiple items and a mining algorithm

hldirect association is a high level relationship between items and frequent itemsets in data. Current research approaches on indirect association mining are limited to indirect association between itempairs,which will discover too many rules from dataset. A formal definition of indirect association between multiple items is presented, along with an algorithm, SET NIA, for mining this kind of indirect associations based on anti-monotonicity of indirect associations and frequent itempair support matrix. While the found rules contain same information as compared to the rules found by indirect association between itempairs algorithms, this notion brings space-saving in storage of the rules as well as superiority for human to understand and apply the ndes. Experiments conducted on two real-word datasets show that SET _ NIA can effectively find fewer rules than existing algorithms which mine indirect association between itempairs,the experimental results also prove that SET_NIA has better performance than existing algorithms.