World Academy of Science, Engineering and Technology

Authors: Laila Y. Fannas, Ahmed Y. Ben Sasi

Abstract:

This research presents a handwritten signature recognition based on angle feature vector using Artificial Neural Network (ANN). Each signature image will be represented by an Angle vector. The feature vector will constitute the input to the ANN. The collection of signature images will be divided into two sets. One set will be used for training the ANN in a supervised fashion. The other set which is never seen by the ANN will be used for testing. After training, the ANN will be tested for recognition of the signature. When the signature is classified correctly, it is considered correct recognition otherwise it is a failure.

Keywords: Signature Recognition, Artificial Neural Network, Angle Features.

Authors: Loai AbdAllah, Ilan Shimshoni

Abstract:

Missing values in data are common in real world applications. Since the performance of many data mining algorithms depend critically on it being given a good metric over the input space, we decided in this paper to define a distance function for unlabeled datasets with missing values. We use the Bhattacharyya distance, which measures the similarity of two probability distributions, to define our new distance function. According to this distance, the distance between two points without missing attributes values is simply the Mahalanobis distance. When on the other hand there is a missing value of one of the coordinates, the distance is computed according to the distribution of the missing coordinate. Our distance is general and can be used as part of any algorithm that computes the distance between data points. Because its performance depends strongly on the chosen distance measure, we opted for the k nearest neighbor classifier to evaluate its ability to accurately reflect object similarity. We experimented on standard numerical datasets from the UCI repository from different fields. On these datasets we simulated missing values and compared the performance of the kNN classifier using our distance to other three basic methods. Our  experiments show that kNN using our distance function outperforms the kNN using other methods. Moreover, the runtime performance of our method is only slightly higher than the other methods.

Keywords: Missing values, Distance metric, Bhattacharyya distance.

Authors: R. Saraçoğlu, N. Allahverdi

Abstract:

Searching similar documents and document management subjects have important place in text mining. One of the most important parts of similar document research studies is the process of classifying or clustering the documents. In this study, a similar document search approach that includes discussion of out the case of belonging to multiple categories (multiple categories problem) has been carried. The proposed method that based on Fuzzy Similarity Classification (FSC) has been compared with Rocchio algorithm and naive Bayes method which are widely used in text mining. Empirical results show that the proposed method is quite successful and can be applied effectively. For the second stage, multiple categories vector method based on information of categories regarding to frequency of being seen together has been used. Empirical results show that achievement is increased almost two times, when proposed method is compared with classical approach.

Keywords: Document similarity, Fuzzy classification, Multiple categories, Text mining.

Authors: Daniel F. García, Adrián Fernández

Abstract:

Today, computer systems are more and more complex and support growing security risks. The security managers need to find effective security risk assessment methodologies that allow modeling well the increasing complexity of current computer systems but also maintaining low the complexity of the assessment procedure. This paper provides a brief analysis of common security risk assessment methodologies leading to the selection of a proper methodology to fulfill these requirements. Then, a detailed analysis of the most effective methodology is accomplished, presenting numerical examples to demonstrate how easy it is to use.

Keywords: Computer security, qualitative and quantitative methods, risk assessment methodologies, security risk assessment.

Authors: Cecilia Nugraheni, Luciana Abednego

Abstract:

Sudoku is a kind of logic puzzles. Each puzzle consists of a board, which is a 9×9 cells, divided into nine 3×3 subblocks and a set of numbers from 1 to 9. The aim of this puzzle is to fill in every cell of the board with a number from 1 to 9 such that in every row, every column, and every subblock contains each number exactly one. Sudoku puzzles belong to combinatorial problem (NP complete). Sudoku puzzles can be solved by using a variety of techniques/algorithms such as genetic algorithms, heuristics, integer programming, and so on. In this paper, we propose a new approach for solving Sudoku which is by modelling them as block-world problems. In block-world problems, there are a number of boxes on the table with a particular order or arrangement. The objective of this problem is to change this arrangement into the targeted arrangement with the help of two types of robots. In this paper, we present three models for Sudoku. We modellized Sudoku as parameterized multi-agent systems. A parameterized multi-agent system is a multi-agent system which consists of several uniform/similar agents and the number of the agents in the system is stated as the parameter of this system. We use Temporal Logic of Actions (TLA) for formalizing our models.

Keywords: Sudoku puzzle, block world problem, parameterized multi agent systems modelling, Temporal Logic of Actions.

Authors: Dias de J. C. Lilian, Lobato M. F. Fábio, de Santana L. Ádamo

Abstract:

The occurrence of missing values in database is a serious problem for Data Mining tasks, responsible for degrading data quality and accuracy of analyses. In this context, the area has shown a lack of standardization for experiments to treat missing values, introducing difficulties to the evaluation process among different researches due to the absence in the use of common parameters. This paper proposes a testbed intended to facilitate the experiments implementation and provide unbiased parameters using available datasets and suited performance metrics in order to optimize the evaluation and comparison between the state of art missing values treatments.

Keywords: Data imputation, data mining, missing values treatment, testbed.

Authors: Helyane B. Borges, Julio Cesar Nievola

Abstract:

Hierarchical classification is a problem with applications in many areas as protein function prediction where the dates are hierarchically structured. Therefore, it is necessary the development of algorithms able to induce hierarchical classification models. This paper presents experimenters using the algorithm for hierarchical classification called Multi-label Hierarchical Classification using a Competitive Neural Network (MHC-CNN). It was tested in ten datasets the Gene Ontology (GO) Cellular Component Domain. The results are compared with the Clus-HMC and Clus-HSC using the hF-Measure.

Keywords: Hierarchical Classification, Competitive Neural Network, Global Classifier.

Authors: I. A. Farhat

Abstract:

The problem of optimal planning of multiple sources of distributed generation (DG) in distribution networks is treated in this paper using an improved Ant Colony Optimization algorithm (ACO). This objective of this problem is to determine the DG optimal size and location that in order to minimize the network real power losses. Considering the multiple sources of DG, both size and location are simultaneously optimized in a single run of the proposed ACO algorithm. The various practical constraints of the problem are taken into consideration by the problem formulation and the algorithm implementation. A radial power flow algorithm for distribution networks is adopted and applied to satisfy these constraints. To validate the proposed technique and demonstrate its effectiveness, the well-know 69-bus feeder standard test system is employed.cm.

Keywords: About Ant Colony Optimization (ACO), Distributed Generation (DG).

Authors: Liew Yee Ping, Pang Ying Han, Lau Siong Hoe, Ooi Shih Yin, Housam Khalifa Bashier Babiker

Abstract:

In face recognition, feature extraction techniques attempts to search for appropriate representation of the data. However, when the feature dimension is larger than the samples size, it brings performance degradation. Hence, we propose a method called Normalization Discriminant Independent Component Analysis (NDICA). The input data will be regularized to obtain the most reliable features from the data and processed using Independent Component Analysis (ICA). The proposed method is evaluated on three face databases, Olivetti Research Ltd (ORL), Face Recognition Technology (FERET) and Face Recognition Grand Challenge (FRGC). NDICA showed it effectiveness compared with other unsupervised and supervised techniques.

Keywords: Face recognition, small sample size, regularization, independent component analysis.

Authors: Junghwan Choi, Seongsoo Lee

Abstract:

Histogram equalization is often used in image enhancement, but it can be also used in auto exposure. However, conventional histogram equalization does not work well when many pixels are concentrated in a narrow luminance range.This paper proposes an auto exposure method based on 2-way histogram equalization. Two cumulative distribution functions are used, where one is from dark to bright and the other is from bright to dark. In this paper, the proposed auto exposure method is also designed and implemented for image signal processors with full-HD images.

Keywords: Histogram equalization, Auto exposure, Image signal processor, Low-cost, Full HD Video.

Authors: Joo-Hyung Oh, Dongwan Kang, JunHyung Cho, Chaetae Im

Abstract:

As the mobile Internet has become widespread in recent years, communication based on mobile networks is increasing. As a result, security threats have been posed with regard to the abnormal traffic of mobile networks, but mobile security has been handled with focus on threats posed by mobile malicious codes, and researches on security threats to the mobile network itself have not attracted much attention. In mobile networks, the IP address of the data packet is a very important factor for billing purposes. If one mobile terminal use an incorrect IP address that either does not exist or could be assigned to another mobile terminal, billing policy will cause problems. We monitor and analyze 3G mobile data networks traffics for a period of time and finds some abnormal IP packets. In this paper, we analyze the reason for abnormal IP packets on 3G Mobile Data Networks. And we also propose an algorithm based on IP address table that contains addresses currently in use within the mobile data network to detect abnormal IP packets.

Keywords: WCDMA, 3G, Abnormal IP address, Mobile Data Network Attack

Authors: Phanu Waraporn

Abstract:

This paper discusses the designing of knowledge integration of clinical information extracted from distributed medical ontologies in order to ameliorate a machine learning-based multilabel coding assignment system. The proposed approach is implemented using a decision tree technique of the machine learning on the university hospital data for patients with Coronary Heart Disease (CHD). The preliminary results obtained show a satisfactory finding that the use of medical ontologies improves the overall system performance.

Keywords: Medical Ontology, Knowledge Integration, Machine Learning, Medical Coding, Text Assignment.

Authors: Madihah Mohd Saudi, Zul Hilmi Abdullah

Abstract:

This research paper presents a framework on how to build up malware dataset.Many researchers took longer time to clean the dataset from any noise or to transform the dataset into a format that can be used straight away for testing. Therefore, this research is proposing a framework to help researchers to speed up the malware dataset cleaningprocesses which later can be used for testing. It is believed, an efficient malware dataset cleaning processes, can improved the quality of the data, thus help to improve the accuracy and the efficiency of the subsequent analysis. Apart from that, an in-depth understanding of the malware taxonomy is also important prior and during the dataset cleaning processes. A new Trojan classification has been proposed to complement this framework.This experiment has been conducted in a controlled lab environment and using the dataset from VxHeavens dataset. This framework is built based on the integration of static and dynamic analyses, incident response method and knowledge database discovery (KDD) processes.This framework can be used as the basis guideline for malware researchers in building malware dataset.

Keywords: Dataset, knowledge database discovery (KDD), malware, static and dynamic analyses.

Authors: C. E. Nugraheni, L. Abednego

Abstract:

This paper introduces a framework based on the collaboration of multi agent and hyper-heuristics to find a solution of the real single machine production problem. There are many techniques used to solve this problem. Each of it has its own advantages and disadvantages. By the collaboration of multi agent system and hyper-heuristics, we can get more optimal solution. The hyper-heuristics approach operates on a search space of heuristics rather than directly on a search space of solutions. The proposed framework consists of some agents, i.e. problem agent, trainer agent, algorithm agent (GPHH, GAHH, and SAHH), optimizer agent, and solver agent. Some low level heuristics used in this paper are MRT, SPT, LPT, EDD, LDD, and MON

Keywords: Hyper-heuristics, multi-agent systems, scheduling problem.

Authors: Nchimbi Edward Pius, Liu Qin, Fion Yang, Zhu Hong Ming

Abstract:

The current Hadoop block placement policy do not fairly and evenly distributes replicas of blocks written to datanodes in a Hadoop cluster.

This paper presents a new solution that helps to keep the cluster in a balanced state while an HDFS client is writing data to a file in Hadoop cluster. The solution had been implemented, and test had been conducted to evaluate its contribution to Hadoop distributed file system.

It has been found that, the solution has lowered global execution time taken by Hadoop balancer to 22 percent. It also has been found that, Hadoop balancer respectively over replicate 1.75 and 3.3 percent of all re-distributed blocks in the modified and original Hadoop clusters.

The feature that keeps the cluster in a balanced state works as a core part to Hadoop system and not just as a utility like traditional balancer. This is one of the significant achievements and uniqueness of the solution developed during the course of this research work.

Keywords: Balancer, Datanode, Distributed file system, Hadoop, Replicas.

Authors: T. S. Myers, J. Trevathan

Abstract:

Improvements in the data fusion and data analysis phase of research are imperative due to the exponential growth of sensed data. Currently, there are developments in the Semantic Sensor Web community to explore efficient methods for reuse, correlation and integration of web-based data sets and live data streams. This paper describes the integration of remotely sensed data with web-available static data for use in observational hypothesis testing and the analysis phase of research. The Semantic Reef system combines semantic technologies (e.g., well-defined ontologies and logic systems) with scientific workflows to enable hypothesis-based research. A framework is presented for how the data fusion concepts from the Semantic Reef architecture map to the Smart Environment Monitoring and Analysis Technologies (SEMAT) intelligent sensor network initiative. The data collected via SEMAT and the inferred knowledge from the Semantic Reef system are ingested to the Tropical Data Hub for data discovery, reuse, curation and publication.

Keywords: Information architecture, Semantic technologies Sensor networks, Ontologies.

Authors: Miin-Shen Yang, Yee-Shan Pan

Abstract:

Although there have been many researches in cluster analysis to consider on feature weights, little effort is made on sample weights. Recently, Yu et al. (2011) considered a probability distribution over a data set to represent its sample weights and then proposed sample-weighted clustering algorithms. In this paper, we give a sample-weighted version of generalized fuzzy clustering regularization (GFCR), called the sample-weighted GFCR (SW-GFCR). Some experiments are considered. These experimental results and comparisons demonstrate that the proposed SW-GFCR is more effective than the most clustering algorithms.

Keywords: Clustering; fuzzy c-means, fuzzy clustering, sample weights, regularization.

Authors: Sara El Manar El Bouanani, Ismail Kassou

Abstract:

Our objective in this paper is to propose an approach capable of clustering web messages. The clustering is carried out by assigning, with a certain probability, texts written by the same web user to the same cluster based on Stylometric features and using fuzzy clustering algorithms. Focus in the present work is on comparing the most popular algorithms in fuzzy clustering theory namely, Fuzzy C-means, Possibilistic C-means and Fuzzy Possibilistic C-Means.

Keywords: Authorship detection, fuzzy clustering, profiling, stylometric features.

Authors: Tainchi Lu, Chochih Lin

Abstract:

With a rapid growth in 3D graphics technology over the last few years, people are desired to see more flexible reacting motions of a biped in animations. In particular, it is impossible to anticipate all reacting motions of a biped while facing a perturbation. In this paper, we propose a three-level tracking method for animating a 3D humanoid character. First, we take the laws of physics into account to attach physical attributes, such as mass, gravity, friction, collision, contact, and torque, to bones and joints of a character. The next step is to employ PD controller to follow a reference motion as closely as possible. Once the character cannot tolerate a strong perturbation to prevent itself from falling down, we are capable of tracking a desirable falling-down action to avoid any falling condition inaccuracy. From the experimental results, we demonstrate the effectiveness and flexibility of the proposed method in comparison with conventional data-driven approaches.

Keywords: Character Animation, Forward Dynamics, Motion Tracking, PD Control.

Authors: G. Nur

Abstract:

Monitored 3-Dimensional (3D) video experience can be utilized as “feedback information” to fine tune the service parameters for providing a better service to the demanding 3D service customers. The 3D video experience which includes both video quality and depth perception is influenced by several contextual and content related factors (e.g., ambient illumination condition, content characteristics, etc) due to the complex nature of the 3D video. Therefore, effective factors on this experience should be utilized while assessing it. In this paper, structural information of the depth map sequences of the 3D video is considered as content related factor effective on the depth perception assessment. Cartoon-like filter is utilized to abstract the significant depth levels in the depth map sequences to determine the structural information. Moreover, subjective experiments are conducted using 3D videos associated with cartoon-like depth map sequences to investigate the effectiveness of ambient illumination condition, which is a contextual factor, on depth perception. Using the knowledge gained through this study, 3D video experience metrics can be developed to deliver better service to the 3D video service users. 

Keywords: 3D Video, Ambient Illumination, Cartoon Effect, Depth Perception.

Authors: S. Hma Salah, H. Du, N. Al-Jawad

Abstract:

Ethnicity identification of face images is of interest in many areas of application, but existing methods are few and limited. This paper presents a fusion scheme that uses block-based uniform local binary patterns and Haar wavelet transform to combine local and global features. In particular, the LL subband coefficients of the whole face are fused with the histograms of uniform local binary patterns from block partitions of the face. We applied the principal component analysis on the fused features and managed to reduce the dimensionality of the feature space from 536 down to around 15 without sacrificing too much accuracy. We have conducted a number of preliminary experiments using a collection of 746 subject face images. The test results show good accuracy and demonstrate the potential of fusing global and local features. The fusion approach is robust, making it easy to further improve the identification at both feature and score levels.

Keywords: Ethnicity identification, fusion, local binary patterns, wavelet.

Authors: Ming-Hui Cheng, Jyh-Horng Jeng

Abstract:

In this study, a classification-based video super-resolution method using artificial neural network (ANN) is proposed to enhance low-resolution (LR) to high-resolution (HR) frames. The proposed method consists of four main steps: classification, motion-trace volume collection, temporal adjustment, and ANN prediction. A classifier is designed based on the edge properties of a pixel in the LR frame to identify the spatial information. To exploit the spatio-temporal information, a motion-trace volume is collected using motion estimation, which can eliminate unfathomable object motion in the LR frames. In addition, temporal lateral process is employed for volume adjustment to reduce unnecessary temporal features. Finally, ANN is applied to each class to learn the complicated spatio-temporal relationship between LR and HR frames. Simulation results show that the proposed method successfully improves both peak signal-to-noise ratio and perceptual quality.

Keywords: Super-resolution, classification, spatio-temporal information, artificial neural network.

Authors: Kai Qian, Ming Yang, Minzhe Guo, Prabir Bhattacharya, Lixin Tao

Abstract:

Computer network courses are essential parts of college computer science curriculum and hands-on networking experience is well recognized as an effective approach to help students understand better about the network concepts, the layered architecture of network protocols, and the dynamics of the networks. However, existing networking labs are usually server-based and relatively cumbersome, which require a certain level of specialty and resource to set up and maintain the lab environment. Many universities/colleges lack the resources and build-ups in this field and have difficulty to provide students with hands-on practice labs. A new affordable and easily-adoptable approach to networking labs is desirable to enhance network teaching and learning. In addition, current network labs are short on providing hands-on practice for modern wireless and mobile network learning. With the prevalence of smart mobile devices, wireless and mobile network are permeating into various aspects of our information society. The emerging and modern mobile technology provides computer science students with more authentic learning experience opportunities especially in network learning. A mobile device based hands-on labware can provide an excellent ‘real world’ authentic learning environment for computer network especially for wireless network study. In this paper, we present our mobile device-based hands-on labware (series of lab module) for computer network learning which is guided by authentic learning principles to immerse students in a real world relevant learning environment. We have been using this labware in teaching computer network, mobile security, and wireless network classes. The student feedback shows that students can learn more when they have hands-on authentic learning experience. 

Keywords: Mobile computing, android, network, labware.

Authors: Daniela I. Borissova, Ivan C. Mustakerov, Evgeni D. Bantutov

Abstract:

Nowadays the devices of night vision are widely used both for military and civil applications. The variety of night vision applications require a variety of the night vision devices designs. A web-based architecture of a software system for design assessment before producing of night vision devices is developed. The proposed architecture of the web-based system is based on the application of a mathematical model for designing of night vision devices. An algorithm with two components – for iterative design and for intelligent design is developed and integrated into system architecture. The iterative component suggests compatible modules combinations to choose from. The intelligent component provides compatible combinations of modules  satisfying given user requirements to device parameters. The proposed web-based architecture of a system for design assessment of night vision devices is tested via a prototype of the system. The testing showed the applicability of both iterative and intelligent components of algorithm. 

Keywords: Night vision devices, design modeling, software architecture, web-based system.

Authors: Ivan C. Mustakerov, Daniela I. Borissova

Abstract:

In recent years, new product development became more and more competitive and globalized, and the designing phase is critical for the product success. The concept of modularity can provide the necessary foundation for organizations to design products that can respond rapidly to market needs. The paper describes data structures and algorithms of intelligent Web-based system for modular design taking into account modules compatibility relationship and given design requirements. The system intelligence is realized by developed algorithms for choice of modules reflecting all system restrictions and requirements. The proposed data structure and algorithms are illustrated by case study of personal computer configuration. The applicability of the proposed approach is tested through a prototype of Web-based system.

Keywords: Data structures, algorithms, intelligent web-based system, modular design.

Authors: Nadia Hmad, Tony Allen

Abstract:

This paper presents an ESN-based Arabic phoneme recognition system trained with supervised, forced and combined supervised/forced supervised learning algorithms. Mel-Frequency Cepstrum Coefficients (MFCCs) and Linear Predictive Code (LPC) techniques are used and compared as the input feature extraction technique. The system is evaluated using 6 speakers from the King Abdulaziz Arabic Phonetics Database (KAPD) for Saudi Arabia dialectic and 34 speakers from the Center for Spoken Language Understanding (CSLU2002) database of speakers with different dialectics from 12 Arabic countries. Results for the KAPD and CSLU2002 Arabic databases show phoneme recognition performances of 72.31% and 38.20% respectively.

Keywords: Arabic phonemes recognition, echo state networks (ESNs), neural networks (NNs), supervised learning.

Authors: Yizhi Gai, Yukinori Kobayashi, Yohei Hoshino, Takanori Emaru

Abstract:

Motion control of flexible arms is more difficult than that of rigid arms, however utilizing its dynamics enables improved performance such as a fast motion in short operation time. This paper investigates a ball throwing robot with one rigid link and one flexible link. This robot throws a ball at a set speed with a proper control torque. A mathematical model of this ball throwing robot is derived through Hamilton’s principle. Several patterns of torque input are designed and tested through the proposed simulation models. The parameters of each torque input pattern is optimized and determined by chaos embedded vector evaluated particle swarm optimization (CEVEPSO). Then, the residual vibration of the manipulator after throwing is suppressed with input shaping technique. Finally, a real experiment is set up for the model checking.

Keywords: Motion control, flexible robotic arm, CEVEPSO, ball throwing robot.

Authors: Ahmad F. Subahi, Anthony J. H. Simons

Abstract:

This Paper presents an on-going research in the area of Model-Driven Engineering (MDE). The premise is that UML is too unwieldy to serve as the basis for model-driven engineering. We need a smaller, simpler notation with a cleaner semantics. We propose some ideas for a simpler notation with a clean semantics. The result is known as μML, or the Micro-Modelling Language.

Keywords: Model-driven engineering, model transformations, domain-specific languages, end-user development.

Authors: Lenka Krulikovská, Jaroslav Polec, Michal Martinovič

Abstract:

In this paper we propose a method which improves the efficiency of video coding. Our method combines an adaptive GOP (group of pictures) structure and the shot cut detection. We have analyzed different approaches for shot cut detection with aim to choose the most appropriate one. The next step is to situate N frames to the positions of detected cuts during the process of video encoding. Finally the efficiency of the proposed method is confirmed by simulations and the obtained results are compared with fixed GOP structures of sizes 4, 8, 12, 16, 32, 64, 128 and GOP structure with length of entire video. Proposed method achieved the gain in bit rate from 0.37% to 50.59%, while providing PSNR (Peak Signal-to-Noise Ratio) gain from 1.33% to 0.26% in comparison to simulated fixed GOP structures.

Keywords: Adaptive GOP structure, video coding, video content, shot cut detection.

Authors: Sushil Sarwa, Rajeev Kumar

Abstract:

The wireless mesh networks (WMNs) are emerging technology in wireless networking as they can serve large scale high speed internet access. Due to its wireless multi-hop feature, wireless mesh network is prone to suffer from many attacks, such as denial of service attack (DoS). We consider a special case of DoS attack which is selective forwarding attack (a.k.a. gray hole attack). In such attack, a misbehaving mesh router selectively drops the packets it receives rom its predecessor mesh router. It is very hard to detect that packet loss is due to medium access collision, bad channel quality or because of selective forwarding attack. In this paper, we present a review of detection algorithms of selective forwarding attack and discuss their advantage & disadvantage. Finally we conclude this paper with open research issues and challenges.

Keywords: CAD algorithm, CHEMAS, selective forwarding attack, watchdog & pathrater, wireless mesh network.