Skip to main content
main-content

Über dieses Buch

This book and its companion volumes, LNCS volumes 9140, 9141 and 9142, constitute the proceedings of the 6th International Conference on Swarm Intelligence, ICSI 2015 held in conjunction with the Second BRICS Congress on Computational Intelligence, CCI 2015, held in Beijing, China in June 2015. The 161 revised full papers presented were carefully reviewed and selected from 294 submissions. The papers are organized in 28 cohesive sections covering all major topics of swarm intelligence and computational intelligence research and development, such as novel swarm-based optimization algorithms and applications; particle swarm opt8imization; ant colony optimization; artificial bee colony algorithms; evolutionary and genetic algorithms; differential evolution; brain storm optimization algorithm; biogeography based optimization; cuckoo search; hybrid methods; multi-objective optimization; multi-agent systems and swarm robotics; Neural networks and fuzzy methods; data mining approaches; information security; automation control; combinatorial optimization algorithms; scheduling and path planning; machine learning; blind sources separation; swarm interaction behavior; parameters and system optimization; neural networks; evolutionary and genetic algorithms; fuzzy systems; forecasting algorithms; classification; tracking analysis; simulation; image and texture analysis; dimension reduction; system optimization; segmentation and detection system; machine translation; virtual management and disaster analysis.

Inhaltsverzeichnis

Frontmatter

Neural Networks

Frontmatter

The Effective Neural Network Implementation of the Secret Sharing Scheme with the Use of Matrix Projections on FPGA

In this paper neural network implementation of the modified secret sharing scheme based on a matrix projection is offered. Transition from a finite simple Galois field to a complex field allows to reduce by 16 times memory size, necessary for storage of the precalculated constants. Implementation of the modified secret sharing scheme based on a matrix projection with use of the neural network of a finite ring for execution of modular arithmetical addition and multiplication operations in a finite field allows to save on average 30% of the device area and increases the speed of scheme’s work on average by 17%.

Nikolay Ivanovich Chervyakov, Mikhail Grigorevich Babenko, Nikolay Nikolaevich Kucherov, Anastasiia Igorevna Garianina

A ROP Optimization Approach Based on Improved BP Neural Network PSO

Effective optimization of ROP (Rate of Penetration) is a crucial part of successful well drilling process. Due to the penetration complexities and the formation heterogeneity, traditional way such as ROP equations and regression analysis are confined by their limitations in the drilling prediction. Intelligent methods like ANN and PSO become powerful tools to obtain the optimized parameters with the accumulation of the geology data and drilling logs. This paper presents a ROP optimization approach based on improved BP neural network and PSO algorithm. The main idea is, first, to build prediction model of the target well from well logs using BP neural network, and then obtain the optimized well operating parameters by applying PSO algorithm. During the modelling process, the traditional BP training algorithm is improved by introducing momentum factor. Penalty function is also introduced for the constraints fulfillment. We collect and analyze the well log of the No.104 well in Yuanba, China. The experiment results show that the proposed approach is able to effectively utilize the engineering data to provide effective ROP prediction and optimize well drilling parameters.

Jinan Duan, Jinhai Zhao, Li Xiao, Chuanshu Yang, Changsheng Li

Structure Determination of a Generalized ADALINE Neural Network for Application in System Identification of Linear Systems

This paper presents a structure determination method of a GADALINE based neural network used for linear system identification and parameter estimation. In GADALINE linear system identification, the past input data are used as its input and the past output data are also used as its input in the form of feedback because in such a linear system, the current system output is dependent on past outputs and on both the current and past inputs. The structure determination is then to determine how many past inputs should be included as its input and how many past output should be fed-back as its input also. The measured data set can then be used to train the GADALINE and during training, the performance error can be used to determine the network structure in our method just as the Final Prediction Error used in Akaike’s criterion. One advantage of the method is its simplicity. Simulation results show that the proposed method provides satisfactory performance.

Wenle Zhang

Evolutionary and Genetic Algorithms

Frontmatter

A Self-configuring Metaheuristic for Control of Multi-Strategy Evolutionary Search

There exists a great variety of evolutionary algorithms (EAs) that represent different search strategies for many classes of optimization problems. Real-world problems may combine several optimization features that are not known beforehand, thus there is no information about what EA to choose and which EA settings to apply. This study presents a novel metaheuristic for designing a multi-strategy genetic algorithm (GA) based on a hybrid of the island model, cooperative and competitive coevolution schemes. The approach controls interactions of GAs and leads to the self-configuring solving of problems with a priori unknown structure. Two examples of implementations of the approach for multi-objective and non-stationary optimization are discussed. The results of numerical experiments for benchmark problems from CEC competitions are presented. The proposed approach has demonstrated efficiency comparable with other well-studied techniques. And it does not require the participation of the human-expert, because it operates in an automated, self-configuring way.

Evgenii Sopov

Fast Genetic Algorithm for Pick-up Path Optimization in the Large Warehouse System

The order-picking processes of fixed shelves in a warehouse system require high speed and efficiency. In this study, a novel fast genetic algorithm was proposed for constrained multi-objective optimization problems. The handling of constraint conditions were distributed to the initial population generation and each genetic process. Combine the constraint conditions and objectives, a new partial-order relation was introduced for comparison of individuals. This novel algorithm was used to optimize the stacker picking path in an Automated Storage/Retrieve System (AS/RS) of a large airport. The simulation results indicates that the proposed algorithm reduces the computational complexity of time and space greatly, and meets the needs of practical engineering of AS/RS optimization control.

Yongjie Ma, Zhi Li, Wenxia Yun

A Randomized Multi-start Genetic Algorithm for the One-Commodity Pickup-and-Delivery Traveling Salesman Problem

The One-Commodity Pickup-and-Delivery Traveling Salesman Problem (1-PDTSP) is a generalization of the standard travelling salesman problem. 1-PDTSP is to design an optimal tour that minimizes the overall travelled distance through the depot and a set of customers. Each customer requires either a pickup service or a delivery service. We propose a Randomized Multi-start Genetic Algorithm (RM-GA) to solve the 1-PDTSP. Experimental investigations show that the proposed algorithm is competitive against state-of-the-art methods.

Hiba Yahyaoui, Takwa Tlili, Saoussen Krichen

The Optimum Design of High-Rise Building Structure Based on the Strength and Stiffness of Genetic Algorithm

The high-rise building structure have characteristics that multi-operating mode, multivariate, multivariate, multiple constraints and multi-objective, and the characteristics of complex discrete have a great influence on structural strength and stiffness. So this paper built a comprehensive and practical objective function for optimization design of high-rise building structure. This function made the design not only meet the requirement of structure safety and good performance, but also should be make structure is reasonable and material consumption as low as possible to achieve optimal allocation of resources. On this basis, it used genetic algorithm by strict derivation to code variables, and then evolved until the objective function converged to complete the optimization design of strength and stiffness for high-rise building strength. The work put forward a method of quantitative mathematical description for structure optimization design, and has brought considerable economic benefits.

Kaike Sun, Shoucai Li, Shihong Chu, Shuhai Zheng, Shuai Guo

Fuzzy Systems

Frontmatter

Fuzzy Concepts in Formal Context

Formal concept analysis (FCA) provides a theoretical framework for learning hierarchies of knowledge clusters. This paper is devoted to the study of the fuzzy concept in FCA. We propose a fuzzy relation on the universe to characterize the similarity of the objects. Based on fuzzy rough set model, we present a kind of approximation operators to characterize the fuzzy concept and its accuracy degree in FCA. The basic properties of these operators are investigated.

Luodan Meng, Keyun Qin

Fuzzy Clustering-Based Quantitative Association Rules Mining in Multidimensional Data Set

In order to solve the problem of mining quantitative association rules, an algorithm named Fuzzy Pattern Fusion based on Competitive Agglomeration (FPF-CA) is developed in this paper. The proposed algorithm is based on the superior functionalities of Fuzzy Pattern Fusion (FPF) for mining quantitative association rules and Competitive Agglomeration (CA) for finding the optimal number of clusters. The popular data set of UCI machine learning repository is used to demonstrate the feasibility of the FPF-CA algorithm. The simulation experiment results show that the proposed algorithm can efficiently mine quantitative association rules according to the actual data distribution.

Jining Jia, Yongzai Lu, Jian Chu, Hongye Su

The Application of Fuzzy Pattern Fusion Based on Competitive Agglomeration in Coal-Fired Boiler Operation Optimization

In order to solve the coal-fired boiler operation optimization problem with multiple main controllable parameters, a fuzzy pattern fusion based on competitive agglomeration (FPF-CA) algorithm is developed and applied in this paper. A simplified mathematical model for coal-fired boiler systems is applied in terms of both historical operational and thermal efficiency data under different load conditions. The FPF-CA algorithm can be applied to perform information fusion in terms of combining fuzzy clusters of some quantitative attributes with generated quantitative association rules. The historical data collected from a 130/2.82-M circulating fluidized bed (CFB) boiler being installed in a production scale thermal power plant is used to specifically analyze, and the simulation experiment results show the application of FPF-CA algorithm is a decision oriented intelligent technology and may provide an efficient results for coal-fired boiler operation optimization in thermal power plant.

Jining Jia, Yongzai Lu, Jian Chu, Hongye Su

Forecasting Algorithms

Frontmatter

Technical Indicators for Forex Forecasting: A Preliminary Study

Traders and economists are often at odds with regards to the approach taken towards Forex financial market forecasting. Methods originating from the Artificial Intelligence (AI) area of study have been used extensively throughout the years in predicting the trading pattern as it is deemed to be robust enough to handle the uncertainty associated with Forex trading time series data. Herein this paper, the effects of different input types, in particular: close price as well as various technical indicators derived from the close price are investigated to determine its effects on the Forex trend predicted by an intelligent machine learning module.

Yoke Leng Yong, David C.L. Ngo, Yunli Lee

Short Term Load Forecasting Based on Hybrid ANN and PSO

Short term load forecasting (STLF) is the prediction of electrical load for a period that ranges from one hour to a week. The main objectives of the (STLF) are to predict future load for the generation scheduling at power stations; assess the security of the power system as well as for timely dispatching of electrical power. The traditional load forecasting tools utilize time series models which extrapolate historical load data to predict the future loads. These tools assume a static load series and retain normal distribution characteristics. Due to their inability to adapt to changing environments and load characteristics, they often lead to large forecasting errors. In an effort to reduce the forecasting error, hybrid artificial neural network (ANN) and particle swarm optimization (PSO) is used in this paper.It is shown that the hybridization of ANN and PSO gives better resultscompared to the standard ANN with back propagation.

Ellen Banda, Komla A. Folly

Using Big Data Technology to Contain Current and Future Occurrence of Ebola Viral Disease and Other Epidemic Diseases in West Africa

West Africa is currently plagued with Ebola Viral Disease (EVD) and other minor epidemic diseases which has led to major economic meltdown and high mortality rate in countries like Guinea, Sierra Leone and Liberia as a result of immigration, emigration, foreign trade and investment, bilateral, poor health care issues amidst others. Harmonized EVD related data can help identify individuals who are at risk of contracting the terminal disease and at the same time controlling the outbreak which will in turn lower cost of health care across West Africa. This paper presents the significance, framework as well as an implementation plan and design for using Big Data Technologies (BDT) as an aid to prevent and control EVD in West Africa and the provision of how the principles of cloud computing could be applied to present and impending expectations of the West African Health sector.

Foluso Ayeni, Sanjay Misra, Nicholas Omoregbe

Real-Time Diagnosis and Forecasting Algorithms of the Tool Wear in the CNC Systems

The article proposes concept of solution development for diagnosis and control of real-time cutting tool in for edge cutting machining. The functional model of a diagnosis subsystem based on data reading from sensors of various types established in a cutting zone is developed. Algorithms of subsystem accepted signals processing and averaging, allowing define condition of cutting tool with defined preciseness and it’s condition forecast in future are offered. Architectural features of subsystem program realization are exposed and solutions for integration into CNC system are described. Testing results of the diagnosis subsystem and its main algorithms during manufacturing processes control on turning machine tools are presented.

Georgi M. Martinov, Anton S. Grigoryev, Petr A. Nikishechkin

Classification

Frontmatter

Wavelet Domain Digital Watermarking Algorithm Based on Threshold Classification

This paper is to analyze the complexity of the images for the robustness of the watermark, and to propose a digital watermarking algorithm based on threshold classification in the wavelet domain combined with iterative threshold method. In this algorithm, the original image is divided into different parts of blocks. Then parts of the blocks are selected to be embedded with watermark respectively according to the image entropy. The watermark embedding strength for each block is set according to the image entropy. Optimal threshold of low-frequency sub-band from the two-stage decomposition is to be gained one by one with the iterative threshold method. With the optimal threshold derived, the DWT coefficient of the low-frequency sub-bands is to be classified. According to the result of the classification, different methods are used to overlay the watermark signal respectively. The experimental results have shown that, the algorithm has good imperceptibility and robustness to some common attacks.

Zhiyun Chen, Ya Chen, Wenxin Hu, Dongming Qian

A New Disagreement Measure for Characterization of Classification Problems

Robert P.W. Duin, Elzbieta Pekalska and David M.J. Tax proposed the characterization of classification problems by classifier disagreement. They showed that it is possible to use a standard set of supervised classification problems for constructing a rule that allows deciding about the similarity of new problems to the existing ones. The classifier disagreement could be used to group classification problems in a way which could help to select the appropriate tools for solving new problems. Duin et al proposed a dissimilarity measure between two problems taking into account only the full disagreement matrices. They used a measure of the disagreement based on the coincidence of the classifier output however the correctness was not considered. In this work, we propose a new measure of disagreement which takes into account the correctness of classification result. To calculate the disagreement each object is analyzed to verify if it was classified correctly or incorrectly by the classifiers. We use this new disagreement measure to calculate the dissimilarity between two problems. Some experiments were done and the results were compared against Duin’s et al results.

Yulia Ledeneva, René Arnulfo García-Hernández, Alexander Gelbukh

Tracking Analysis

Frontmatter

User Intention Mining: A Non-intrusive Approach to Track User Activities for Web Application

Monitoring user interaction with web applications is of vital importance as it helps in finding the user’s cognitive behavior towards applications and also helps in analyzing various web metrics. Understanding activity data and finding user insights is next challenge to the web world. User intention can be understood if the user activity logs are associated appropriately with application context. In current scenario, web applications are using crawlers, scrappers, bugs, bots etc. to track user activities and extract out relevant information. Thus, by modifying the source code, it is easy to track and maintain the logs of user activities. But, many organization use third party web applications, where changing the source code is out of reach, making it trickier to maintain the logs of user activities. In this paper, we present an innovative way of logging, processing and extracting out meaningful information by tracking user’s activities on web application.

Piyush Yadav, Kejul Kalyani, Ravi Mahamuni

A Multiple Objects Tracking Method Based on a Combination of Camshift and Object Trajectory Tracking

Multiple objects tracking in dynamic background is one of the key techniques in computer vision. An improved method of multiple objects tracking based on a combination of Camshift and object trajectory tracking is presented in this paper. The algorithm uses Harris corner matching to estimate background movement parameters, adopts two-frame difference to detect moving objects, combines object trajectory tracking with Camshift track moving objects. Our improved algorithm can achieve satisfactory effect not only in tracking multiple objects, but also in tracking continuously the objects which are static, re-enter the current scene or recover motion. The experiments show that the improved algorithm can achieve better result in the accuracy and robustness of detecting and tracking moving objects for dynamic background.

Guo-wu Yuan, Ji-xian Zhang, Yun-hua Han, Hao Zhou, Dan Xu

An Access Point Trajectory Tracking Method by the Weight Update

In recent years, wireless access technology is quite popular for being convenient, fast and flexible. However, due to the openness of wireless network, this technology is also faced with a number of security challenges, one of which is how to deal with the unauthorized access point effectively. As we all know, the unauthorized access point leads to not only increasing interference between signals induced by the fierce competition of wireless channel resources, but also data leakage resulting in ‘‘wireless phishing”. In response to these security threats, much importance has been put on the research of unauthorized access point location and trajectory tracking. This paper firstly proposes an optimization model of wireless signal propagation. Then an access point location and tracking method called APL_T is put forward, which supports three-dimensional location based on the weight update improving the location accuracy effectively and raises the trajectory tracking of the access point in the light of the three-dimensional location. Finally, the experimental results show that APL_T has high accuracy and can meet practical requirements.

Lin Chen, Guilin Cai, Yun Lan

Urban Principal Traffic Flow Analysis Based on Taxi Trajectories Mining

The understanding of urban traffic pattern can benefit the urban operation a lot, including the traffic forecasting, traffic jam resolution, emergency response and future infrastructure planning. In modern cities, thousands of taxicabs equipped with GPS can be considered as a large number of ubiquitous mobile probes traversing and sensing in the urban area, whose trajectories will bring great insight into the urban traffic management. Thus, in this paper we investigate the urban traffic pattern based on the taxi trajectories, especially the principal Origin-Destination traffic flow (OD flow) extraction. Focusing on the picking-up and dropping-off events, the issue is solved by a spatiotemporal density-based clustering method. The OD flow analysis is formulated as a 4-D node clustering problem and the relative distance function between two OD flows is defined, including a clustering preference factor which is adjustable according to the observation scale favor. Finally, we conduct the method on the taxi trajectory dataset generated by 28,000 taxicabs in Beijing from May 1st to May 30th, 2009 to evaluate its performance and interpret some underlying insights of the time-resolved results.

Bing Zhu, Xin Xu

Simulation

Frontmatter

Building Ontologies for Agent-Based Simulation

Using ontologies for simulation models construction has some advantages that cannot be underestimated. Building the ontology, a modeler has to choose conceptualization method, which significantly affects the structure and usability of resulting models. A tendency of using standard ontologies without critically estimating their applicability for particular tasks may even lead to the loss of the model’s efficiency and reliability. In this work, we are considering a simple criterion which may be used to pragmatically assess applicability of particular modeling techniques for building ontologies for simulation models. We will specially focus on the temporal aspect, states and events representation methods in the model. A fragment of ontology for the city social infrastructure optimization modeling will be considered.

Sergey Gorshkov

Simulation with Input Uncertainties Using Stochastic Petri nets

Simulation with input uncertainties using stochastic Petri nets (SPN) takes into account the effects of uncertainties in exponential transition rates when calculating performance indices. The proposed simulation method, using interval arithmetic, allows the computing with simultaneous influence of uncertainties and variabilities of parameters. A case study is presented for processing uncertainty from stochastic Petri nets models. Interval SPN simulation do not compute the reachability graph, but follows the proposed Monte Carlo interval style simulation.

Sérgio Galdino, Francisco Monte

Image and Texture Analysis

Frontmatter

Single Image Dehazing Based on Improved Dark Channel Prior

The sky region of restored images often appears serious noise and color distortion using classical dark channel prior algorithm. To address this issue, we propose an improved dark channel prior algorithm which recognizes the sky regions in hazy image by gradient threshold combined with the absolute value of the difference of atmospheric light and dark channel. And then we estimate the transmission in sky and non-sky regions separately. At last, we enhance the brightness and contrast of results. Experimental results show that our restored images are more natural and smooth in sky regions.

Taimei Zhang, Youguang Chen

Rotation Invariant Texture Analysis Based on Co-occurrence Matrix and Tsallis Distribution

This article addressed some extensions of a texture classifier invariant to rotations. Originally, that classifier is an improvement of the seminal Haralick’s paper in a sense that the former is rotation invariant due to a circular kernel, which encompasses two concentric circles with different radii and then the co-occurrence matrix is formed. It is not considered only pixels falling exactly on the circle, but also others in its vicinity according to a Gaussian scattering. Firstly, 6 attributes are computed from each of the 18 texture patterns, after that texture patterns are rotated and a correct classification, considering Euclidian distance, is sought. The present paper assesses the performance of the afore-mentioned approach with some alterations: Tsallis rather than Gaussian distribution; addition of noise to rotated images before classification; and Principal Components Analysis during the extraction of features.

Mateus Habermann, Felipe Berla Campos, Elcio Hideiti Shiguemori

Dimension Reduction

Frontmatter

A Fast Isomap Algorithm Based on Fibonacci Heap

For the slow operational speed problem of Isomap algorithm in which the Floyd-Warshall algorithm is applied to finding shortest paths, an improved Isomap algorithm is proposed based on the sparseness of the adjacency graph. In the improved algorithm, the runtime for shortest paths is reduced by using Dijkstra’s algorithm based on Fibonacci heap, and thus the Isomap operation is speeded up. The experimental results on several data sets show that the improved version of Isomap is faster than the original one.

Taiguo Qu, Zixing Cai

A Novel Wavelet Transform – Empirical Mode Decomposition Based Sample Entropy and SVD Approach for Acoustic Signal Fault Diagnosis

An advanced and accurate intelligent fault diagnosis system plays an important role in reducing the maintenance cost of modern industry. However, a robust and efficient approach which can serve the purpose of detecting incipient faults still remains unachievable due to weak signals’ small amplitudes, and also low signal-to-noise ratios (SNR). One way to overcome the problem is to adopt acoustic signal because of its inherent characteristic in terms of high sensitive to early stage faults. Nonetheless, it also suffers from low SNR and results in high computational cost. Aiming to solve the aforesaid problems, a novel wavelet transform - empirical mode decomposition (WT-EMD) based Sample Entropy (SampEn) and singular value decomposition (SVD) approach is proposed. By exerting wavelet analysis on the intrinsic mode functions (IMFs), the end effects, which decreases the accuracy of EMD, is significantly alleviated and the SNR is greatly improved. Furthermore, SampEn and SVD, which function as health indicators, not only help to reduce the computational cost and enhance the SNR but also indicate both irregular and periodic faults adequately.

Jiejunyi Liang, Zhixin Yang

Unsatisfiable Formulae of Gödel Logic with Truth Constants and , $$\prec $$ , $$\Delta $$ Are Recursively Enumerable

This paper brings a solution to the open problem of recursive enumerability of unsatisfiable formulae in the first-order Gödel logic. The answer is affirmative even for a useful expansion by intermediate truth constants and the equality,

, strict order,

$$\prec $$

, projection

$$\Delta $$

operators. The affirmative result for unsatisfiable prenex formulae of

$$G_\infty ^\Delta $$

has been stated in [

1

]. In [

7

], we have generalised the well-known hyperresolution principle to the first-order Gödel logic for the general case. We now propose a modification of the hyperresolution calculus suitable for automated deduction with explicit partial truth.

Dušan Guller

System Optimization

Frontmatter

An Intelligent Media Delivery Prototype System with Low Response Time

Streaming media has been increasing in the Internet as a popular form of content. However, the streaming media delivery between server and client browser still has problems to be solved, such as the poor processing efficiency in dealing with large concurrent access and the high usage of bandwidth. In particular, large scale video site usually has a large number of distributed server nodes. It is of great importance for the system to respond to users’ request rapidly by choosing the proper video source for users and handling resource cache problem properly. To solve the above problems in streaming media delivery, this paper proposes a content delivery solution. The system consists of several nodes that differ in role function, and streaming media content is stored in these nodes. Users will get an optimized response, and the system selects the nearest node that has the requested video according to logical distance intellectually. The selected node will provide video stream for users. In addition, the system is equipped with a high performance content indexing and searching mechanism. The index is able to retrieve users’ requested resource rapidly and therefore guarantees a good performance in selecting nodes.

Jinzhong Hou, Tiejian Luo, Zhu Wang, Xiaoqi Li

Incomplete Distributed Information Systems Optimization Based on Queries

In this paper we assume there is a group of connected distributed information systems (DIS). They work under the same ontology. Each information system has its own knowledgebase. Values of attributes in information system form atomic expressions of a language used for communication with others. Collaboration among systems is initiated when one of them (called a client) is asked to resolve a query containing nonlocal attributes for. In such case, the client has to ask for help other information systems to have that query answered. As the result of its request, knowledge is extracted locally in each information system and sent back to the client. The outcome of this step is a knowledgebase created at the client site, which can be used to answer given query. In this paper we present a method of identifying which information system is semantically the closest to client.

Agnieszka Dardzinska, Anna Romaniuk

Strategies for Improving the Profitability of a Korean Unit Train Operator: A System Dynamics Approach

A unit train (UT) has been developed primarily in countries that have wide or long territories, to move freight quickly over long distances. In South Korea, UTs have contributed to the facilitation of the overland export/import logistics for the last decade. However, UT operators in South Korea, which is a small country surrounded by North Korea and bodies of water, suffer from low profitability when competing with trucking companies because of diverse reasons that they cannot control. On this account, this research aims to develop business strategies for improving the profitability of a Korean UT operator. We analyzed both the revenues and expenses of a representative operator in Korea, and found simple but meaningful financial circular causality, using the system dynamics methodology. Thus, we presented and scientifically reinterpreted two strategies that might be acceptable alternatives: the internalization of shuttle carriers and the securing of more freight.

Jae Un Jung, Hyun Soo Kim

Other Applications

Frontmatter

High Performance NetFPGA and Multi-fractal Wavelet Model Based Network Traffic Generator

Internet traffic generator plays very important roles in network measurement and management field.A distributed digital filter is designed in this paper,which united with Multi-fractal Wavelet model(MWM) to generate network traffic data, then a network traffic generator was researched and implemented on the NetFPGA platform. Experiment result shows that the designed network traffic generator has good ability to generate traffic in accordance with the real network condition. More importantly, the generated traffic shows the characters of self similar and multi-fractal, which are two of the most important features of real Internet traffic. With high speed network packets process ability of the NetFPGA Platform, the designed network traffic generator can generate traffic on a highest speed of 10Giabit/s. The designed network traffic generator displays the distinct advantages in performance and real condition simulation ability contract to the invented works.

Zhenxiang Chen, Keke Wang, Lizhi Peng, Tao Sun, Lei Zhang

An Approach to Integrating Emotion in Dialogue Management

Presented in this paper is a method for the construction of emotion-enabled embodied (conversational) agents. By using a modified POMDP model, this method allows dialogue management not only to include emotion as part of the observation of user’s actions, but also to take system’s response time into consideration when updating belief states. Consequently, a novel algorithm is created to direct conversation in different contextual control modes, whose dynamic changes further provide hints for emotion animation with facial expressions and voice tunes. Experiment results demonstrate that the integration of emotion in dialogue management makes embodied agents more appealing and yields much better performance in human/computer interaction.

Xiaobu Yuan

Mathematics Wall: Enriching Mathematics Education Through AI

We present the progress of our ongoing research titled

Mathematics Wall

which aims to create an interactive problem solving environment where the system intelligently interacts with users. In its full glory, the wall shall provide answers and useful explanations to the problem-solving process using artificial intelligence (AI) techniques. In this report, we discuss the following components: the digital ink segmentation task, the symbol recognition task, the structural analysis task, the mathematics expression recognition, the evaluation of the mathematics expressions and finally present the results. We then present and discuss the design decisions of the whole framework and subsequently, the implementation of the prototypes. Finally, future work on the explanation facility is discussed.

Somnuk Phon-Amnuaisuk, Saiful Omar, Thien-Wan Au, Rudy Ramlie

Segmentation and Detection System

Frontmatter

Spatial Subdivision of Gabriel Graph

Gabriel graph is one of the well-studied proximity graphs which has a wide range of applications in various research areas such as wireless sensor network, gene flow analysis, geographical variation analysis, facility location, cluster analysis, and so on. In numerous applications, an important query is to find a specific location in a Gabriel graph at where a given set of adjacent vertices can be obtained if a new point is inserted. To efficiently compute the answer of this query, our proposed solution is to subdivide the plane associated with the Gabriel graph into smaller subregions with the property that if a new point is inserted anywhere into a specific subregion then the set of adjacent vertices in the Gabriel graph remains constant for that point, regardless of the exact location inside the subregion. In this paper, we examine these planar subregions, named redundant cells, including some essential properties and sketch an algorithm of running time

$${\mathcal {O}}(n^2)$$

to construct the arrangement that yields these redundant cells.

M. Z. Hossain, M. A. Wahid, Mahady Hasan, M. Ashraful Amin

Efficient Construction of UV-Diagram

Construction of the Voronoi diagram for exploring various proximity relations of a given dataset is one of the fundamental problems in numerous application domains. Recent developments in this area allow constructing Voronoi diagram based on the dataset of uncertain objects which is known as Uncertain-Voronoi diagram (UV-diagram). In compare to the conventional Voronoi diagram of point set, the most efficient algorithm known to date for the UV-diagram construction requires extremely long running time because of its sophisticated geometric structure. This text introduces several efficient algorithms and techniques to construct the UV-diagram and compares the advantages and disadvantages with previously known algorithms and techniques in literature.

M. Z. Hossain, Mahady Hasan, M. Ashraful Amin

Portrait Image Segmentation Based on Improved Grabcut Algorithm

Traditional computer portrait caricature system mainly take the method that exaggerate and deform real images directly, that lead the facial image background also been deformed when exaggerate facial image. If in pretreatment stage, we segmented the characters and background of the input image, and then do the subsequent processing, the problem may be solved. But for better portrait caricature effects, we need an excellent segmentation algorithm. So, we propose an improved Grabcut image segmentation algorithm and use it to extract the prospect character image for exaggeration and deformation. In practical application, we separate deform and exaggerate the foreground characters image with TPS method, then fuse it with the original or new background picture, get the final image. Application proves, the method solves the background deformation problem well, and improves the quality and rate of image segmentation, caricature synthesis effect reality and natural.

Shuai Li, Xiaohui Zheng, Xianjun Chen, Yongsong Zhan

A Multiple Moving Targets Edge Detection Algorithm Based on Sparse Matrix Block Operation

A multiple moving targets edge detection algorithm based on sparse matrix block operation is proposed in this paper. The algorithm uses background subtraction algorithm to obtain the foreground image contains multiple moving targets. After getting the ideal foreground image, active contour model is used for edge detection. Here, we improved the active contour model by introducing the sparse matrix block operation. Through the quad-tree decomposition of the foreground image, the proposed algorithm uses the sparse matrix block operation to calculate the corresponding regional seed position of multiple moving targets. Finally, it executes the active contour model in parallel to complete the edge detection. Experimental results show that edge detection of the algorithm similar to the human visual judgment, and the introduction of sparse matrix block operation to calculate regional seed for active contour model reduces the time, improves the convergence of the profile curve and edge detection accuracy.

Kun Zhang, Cuirong Wang, Yanxia Lv

Robust Corner Detection Based on Bilateral Filter in Direct Curvature Scale Space

In traditional Curvature Scale Space (CSS) corner detection algorithms, Gaussian filter is used to remove noise existing in canny edge detection results. Unfortunately, Gaussian filter will reduce the precision of corner detection. In this paper, a new method of robust corner detection based on bilateral filter in direct curvature scale space is proposed. In this method, bilateral filter is adopted to reduce image noise and keep image details. Instead of curvature scale space, direct curvature scale space is applied to reduce the computational complexity of the algorithm. Meanwhile, multi-scale curvature product with certain threshold is used to strengthen the corner detection. Experimental results show that our proposed method can improve the performance of corner detection in both accuracy and efficiency, and which can also gain more stable corners at the same time.

Bin Liao, Jungang Xu, Huiying Sun, Hong Chen

An Infrared Thermal Detection System

Anybody whose temperature differs from absolute zero (0K) emits and absorbs electromagnetic radiation coming from, one hand, the physico-chemical nature and on the other hand, the action of intrinsic mechanisms of vibrational energies of the molecules. In this paper, we propose a robot that exploits this characteristic to detect and to track a thermal source in its environment. The system consists of a mobile platform having mounted a thermal detection device using thermopiles that is controlled by a computer via a graphical user interface.

Zahira Ousaadi, Nadia Saadia

FDM-MC: A Fault Diagnosis Model of the Minimum Diagnosis Cost in the System Area Network

There are many problems in the fault diagnosis of system area network which facing enormous challenges. These problems are mainly due to the fact that network fault is often conditionally dependent on many factors, which are usually dependent on complex association relationship. Non-linear mapping may exists between symptoms and causes of network fault, and the same network fault often has different symptoms at different time, while one symptom can be the result of several network faults. Because there is a lot of correlative information in the network, how to construct the model of fault diagnosis is a challenging topic. In this paper, we firstly provided the description of the diagnosis costs, and then we proposed the model based on the condition of dependent diagnosis actions and the model based on the condition of dependent faults. Through a series of theoretical support, we have seen that our diagnostic model produces expected cost of diagnosis that are close to the optimal result and the lower than the simple planners for a domain of troubleshooting.

Jie Huang, Lin Chen

A Robust Point Sets Matching Method

Point sets matching method is very important in computer vision, feature extraction, fingerprint matching, motion estimation and so on. This paper proposes a robust point sets matching method. We present an iterative algorithm that is robust to noise case. Firstly, we calculate all transformations between two points. Then similarity matrix are computed to measure the possibility that two transformation are both true. We iteratively update the matching score matrix by using the similarity matrix. By using matching algorithm on graph, we obtain the matching result. Experimental results obtained by our approach show robustness to outlier and jitter.

Xiao Liu, Congying Han, Tiande Guo

Machine Translation

Frontmatter

An Enhanced Rule Based Arabic Morphological Analyzer Based on Proposed Assessment Criteria

Morphological analysis is a vital part of natural language processing applications, there are no definitive standards for evaluating and benchmarking Arabic morphological systems. This paper proposes assessment criteria for evaluating Arabic morphological systems by scrutinizing the input, output and architectural design to enables researchers to evaluate and fairly compare Arabic morphology systems. By scoring some state of the art Arabic morphological analyzers based on the proposed criteria; the accuracy scores showed that the best algorithm failed to achieve a reliable rate. Hence, this paper introduced an enhanced algorithm for resolving the inflected Arabic word, identifies its root, finds its pattern and POS tagging that will reduce the search time considerably and to free up the deficiencies identified by this assessment criteria. The proposed model uses semantic rules of the Arabic language on top of a hybrid sub-model based on two existing algorithms (Al-Khalil and An Improved Arabic morphology analyzer IAMA rules).

Abdelmawgoud Mohamed Maabid, Tarek Elghazaly, Mervat Ghaith

Morphological Rules of Bangla Repetitive Words for UNL Based Machine Translation

This paper develops new morphological rules suitable for Bangla repetition words to be incorporated into an interlingua representation called Universal Networking Language (UNL). The proposed rules are to be used to combine verb roots and their inflexions to produce words which are then combined with other similar types of words to generate repetition words. This paper outlines the format of morphological rules for different types of repetition words that come from verb roots based on the framework of UNL provided by the UNL centre of the Universal Networking Digital Language (UNDL) Foundation.

Md. Nawab Yousuf Ali, Golam Sorwar, Ashok Toru, Md. Anik Islam, Md. Shamsujjoha

River Network Optimization Using Machine Learning

Lack of potable water is a perennial problem in the day-to-day life of mankind around the world. The demand-supply variations have been on an increasing trend for so many years in different developing countries. To address this prevailing issue is the need of the hour for the society and the relevant government agencies. In this paper, as an explorative approach, we address this predominant issue in the form of an alternate solution which re-routes the course of the natural water sources, like rivers, through those areas, where the water supply is minimal in comparison with the demand, in a cost-effective and highly beneficial manner. Our analysis and discussions are more prone to Indian scenario where India is one of the worst affected fast developing countries for the water crisis. This involves the consideration of the physical, ecological and social characteristics of the lands on the route that fall under the course of the river and also the regions dependent on its flow. In order to understand and predict the optimized new flow paths to divert the water sources, we have employed Machine Learning algorithms like Multiple Regression and Multi-Swarm Optimization techniques. For selecting the most needed re-route, we have also considered the areas that are prone to droughts, and unite the re-routed water with the original course of the river, finally, draining into the sea, for the sustainable development. The proposed methodology is experimented by analyzing the flow areas (river basins) of river Mahanadi in India, one of the considerably important projects cited many times without any real implementation. The results are validated with the help of a study conducted earlier by the National Water Development Agency (NWDA), Government of India, in 2012.

M. Saravanan, Aarthi Sridhar, K. Nikhil Bharadwaj, S. Mohanavalli, V. Srividhya

Virtual Management and Disaster Analysis

Frontmatter

Gauging the Politeness in Virtual Commercial Contexts Based on Patrons’ Collective Perceptions

Politeness constantly plays a significant role in commercial contexts. In contrast to its importance, politeness-related issues in fast-growing virtual commercial contexts received rare attention. This article reports a work developing an instrument for gauging degree of politeness in online storefronts. The instrument’s reliability and validity were confirmed through analyzing empirical data, which distilled collective perceptions of 282 sampled patrons. A second-order confirmatory factor analysis revealed that online consumers’ tendency in paying relative more attention to their rights being respected and gaining useful information while they are assessing online retailers’ politeness. Using the instrument, people can measure the degree of politeness in online retailers.

I-Ching Chen, Shueh-Cheng Hu

Implementation and Theoretical Analysis of Virtualized Resource Management System Based on Cloud Computing

With the continuous and rapid development of computational science and data engineering related techniques, the transmission and protection of data are crucial in the computer science community. Cloud computing is becoming increasingly important for provision of services and storage of data in the Internet. Cloud computing as newly emergent computing environment offers dynamic flexible infrastructures and QoS guaranteed services in pay-as-you-go manner to the public. System virtualization technology providing a flexible and extensible system service is the foundation of cloud computing. How to provide the infrastructure for a self – management and independent cloud computing through virtualization has become one of the most important challenges. In this paper, using feedback control theory, we present VM-based architecture for adaptive management of virtualized resources in cloud computing and model an adaptive controller that dynamically adjusts multiple virtualized resources utilization to achieve application Service Level Objective (SLO) in cloud computing. Through evaluating the proposed methodology, it is shown that the model could allocate resources reasonably in response to the dynamically changing resource requirements of different applications which execute on different VMs in the virtual resource pool to achieve applications SLOs. Further optimization and n-depth discussion are also taken into consideration in the end.

Yong Li, Qi Xu

Integration of a Mathematical Model Within Reference Task Model at the Procurement Process Using BPMN for Disasters Events

The presented approach on this article is related to the task streams activities according to the reference model from humanitarian logistics tasks proposed by Blecken (2009) using the Business Process Modeling Notation (BPMN). Steps were presented initially for the mathematical model insertion, emphasizing on the function of supplies acquisition during the phase of response in case of disasters. The outcome from this model consists on determining which suppliers have the capability to deliver emergency supply items on the requested date. Illustrating the usage of this model, we will show the development of a small example. The first approach contributions is the BPMN specification for assessment tasks and procurement, according to the reference model, including new tasks. The second contribution is represented by a mathematical model insertion, specifically on the task “select supplier”, from the reference model. Therefore, there was a graph model proposed for the Network Flow Model, which was adapted to the Humanitarian Logistics context.

Fabiana Santos Lima, Mirian Buss Gonçalves, Márcia Marcondes Altimari Samed, Bernd Hellingrath

Other Applications

Frontmatter

An Efficient Design of a Reversible Fault Tolerant $$n$$ -to-2 $$^n$$ Sequence Counter Using Nano Meter MOS Transistors

This paper proposes an efficient reversible synthesis for the

$$n$$

-to-2

$$^n$$

sequence counter, where

$$n\ge $$

2 and

$$n \epsilon $$

$$N$$

. The proposed circuits are designed using only reversible fault tolerant gates. Thus, the entire circuit inherently becomes fault tolerant. In addition, an algorithm to design the

$$n$$

-to-2

$$^n$$

reversible fault tolerant sequence counter based on fault tolerant J-K flip-flops has been presented. The functional verification of the proposed circuit is completed through the simulation results. Moreover, the comparative results show that the proposed method performs much better and is much more scalable than the existing approaches.

Md. Shamsujjoha, Sirin Nahar Sathi, Golam Sorwar, Fahmida Hossain, Md. Nawab Yousuf Ali, Hafiz Md. Hasan Babu

Transfer of Large Volume Data over Internet with Parallel Data Links and SDN

The transfer of large volume data over computer network is important and unavoidable operation in the past, now and in any feasible future. There are a number of methods/tools to transfer the data over computer global network (Internet). In this paper the transfer of data over Internet is discussed. Several free of charge utilities to transfer the data are analyzed here. The most important architecture features are emphasized and suggested idea to add SDN Openflow protocol technique for fine tuning the data transfer over several parallel data links.

S. E. Khoruzhnikov, V. A. Grudinin, O. L. Sadov, A. Y. Shevel, A. B. Kairkanov

Program of Educational Intervention for Deaf-Blind Students

The purpose of the study was to develop an interventional educational plan for a deaf-blind student with screened difficulties in body schema awareness. This was part of a more extensive research in developing the screening inventory for the deaf-blind students’ cognitive and communicative profile. The study used a qualitative research methodology and adopted an interpretative position. The aim of the inquiry was descriptive. We followed the case study methodology. The application of the interventional program aimed to help the deaf-blind student in promoting early concept development (body schema). The student was offered multisensory and concrete experiences in order to promote the body schema awareness.

After the intervention, it was observed that the student became aware of having a body with a center (midline) and two sides. The student succeeded in naming her body parts and matching them to others.

Maria Zeza, Pilios-D. Stavrou

Using Map-Based Interactive Interface for Understanding and Characterizing Crime Data in Cities

Crime Data Analysis is vital to all cities and has become a major challenge. It is important to understand crime data to help law enforcement and the public in finding solutions and in making decisions. Data mining algorithms in conjunction with information system technologies and detailed public data about crimes in a given area have allowed the government and the public to better understand and characterize crimes. Furthermore, using visualization tools, the data can be represented in forms that are easy to interpret and use. This paper describes the design and implementation of a map-based interactive crime data web application that can help identify spatial temporal association rules around various types of facilities at different time, and extract important information that will be visualized on a map.

Zhenxiang Chen, Qiben Yan, Lei Zhang, Lizhi Peng, Hongbo Han

Backmatter

Weitere Informationen

Premium Partner

    Bildnachweise