Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing 2011

Social informatics, as a sub-field of the general field of informatics, deals with processing and analysis of data for social studies. One of the social data repositories is the International Social Survey Program (ISSP) the provides crossnational surveys on various social topic. Previous studies of this data often used a subset of available variables and sometimes a reduced number of records, and most of these analyses have focused on predictive techniques such as regression. In this paper, we analyze the Environment II module of this data set using variable clustering to produce meaningful clusters related to questionnaire sections and provide information to reduce the number of demographic variables considered in further analysis. Case level clustering was attempted, but did not produce adequate results.

Offline handwritten recognition is an important automated process in pattern recognition and computer vision field. This paper presents an approach of polar coordinate-based handwritten recognition system involving Support Vector Machines (SVM) classification methodology to achieve high recognition performance. We provide comparison and evaluation for zoning feature extraction methods applied in Polar system. The recognition results we proposed were trained and tested by using SVM with a set of 650 handwritten character images. All the input images are segmented (isolated) handwritten characters. Compared with Cartesian based handwritten recognition system, the recognition rate is more stable and improved up to 86.63%.

A wide range of emerging real-time services (e.g. VoIP, video conferencing, video-on-demand) require different levels of Quality of Services (QoS) guarantees over wireless networks. Scheduling algorithms play a key role in meeting these QoS requirements. A distinction of QoS guarantees is made between deterministic and statistical guarantees. Most of research in this area have been focused on deterministic delay bounds and the statistical bounds of differentiated real-time services are not well known. This paper provides the mathematical analysis of the statistical delay bounds of different levels of Constant Bit Rate (CBR) traffic under First Come First Served with static priority (P-FCFS) scheduling. The mathematical results are supported by the simulation studies. The statistical delay bounds are also compared with the deterministic delay bounds of several popular rate-based scheduling algorithms. It is observed that the deterministic bounds of the scheduling algorithms are much larger than the statistical bounds and are overly conservative in the design and analysis of efficient QoS support in wireless access systems.

New open box and nonlinear model of Sin and Sigmoid Higher Order Neural Network (SS-HONN) is presented in this paper. A new learning algorithm for SS-HONN is also developed from this study. A time series data simulation and analysis system, SS-HONN Simulator, is built based on the SS-HONN models too. Test results show that every error of SS-HONN models are from 2.1767% to 4.3114%, and the average error of Polynomial Higher Order Neural Network (PHONN), Trigonometric Higher Order Neural Network (THONN), and Sigmoid polynomial Higher Order Neural Network (SPHONN) models are from 2.8128 to 4.9076%. It means that SS-HONN models are 0.1131% to 0.6586% better than PHONN, THONN, and SPHONN models.

In this paper, we examine a method in which distinctive mouth shapes are processed using a computer.When lip-reading skill holders do lip-reading, they stare at the changes in mouth shape of a speaker. In recent years, some researches into lip-reading using information technology has been pursued. There are some researches based on the changes in mouth shape. The researchers analyze all data of the mouth shapes during an utterance, whereas lip-reading skill holders look at distinctive mouth shapes. We found that there was a high possibility for lip-reading by using the distinctive mouth shapes. To build the technique into a lip-reading system, we propose an expression method of the distinctive mouth shapes which can be processed using a computer. In this way, we acquire knowledge about the relation between Japanese phones and mouth shapes. We also propose a method to express order of the distinctive mouth shapes which are formed by a speaker.

Resource constraint sensors of a Wireless Sensor Network (WSN) cannot afford the use of costly encryption techniques like public key while dealing with sensitive data. So symmetric key encryption techniques are preferred where it is essential to have the same cryptographic key between communicating parties. To this end, keys are preloaded into the nodes before deployment and are to be established once they get deployed in the target area. This entire process is called key predistribution. In this paper we propose one such scheme using unique factorization of polynomials over Finite Fields. To the best of our knowledge such an elegant use of Algebra is being done for the first time in WSN literature. The best part of the scheme is large number of node support with very small and uniform key ring per node. However the resiliency is not good. For this reason we use a special technique based on Reed Muller codes proposed recently by Sarkar, Saha and Chowdhury in 2010. The combined scheme has good resiliency with huge node support using very less keys per node.

With the increasing applications of computing systems, recognizing accurate and application oriented human expressions, is becoming a challenging topic. The face is a highly attractive biometric trait for expression recognition because of its physiological structure and location. In this paper we proposed two different subspace projection methods that are the extensions of basis subspace projection methods and applied them successfully for facial expression recognition. Our first proposal is an improved principal component analysis for facial expression recognition in frontal images by using an extension of eigenspaces and we term this as WR-PCA (region based and weighted principal component analysis). Secondly we proposed locally salient Independent component analysis(LS-ICA) method for facial expression analysis. These two methods are extensively discussed in the rest of the paper. Experiments with Cohn-kanade database show that these techniques achieves an accuracy rate of 93% when using LS-ICA and 91.81% when WR-PCA and 83.05% when using normal PCA. Our main contribution here is that by performing WR-PCA, which is an extension of typical PCA and first investigated by us, we achieve a nearly similar result as LS-ICA which is a very well established technique to identify partial distortion.

The use of a web Health Portal can be employed not only for reducing health costs but also to view patient’s latest medical information (e.g. clinical tests, pathology and radiology results, discharge summaries, prescription renewals, referrals, appointments) in real-time and carry out physician messaging to enhance the information exchanged, managed and shared in the Australian healthcare sector. The Health Portal connects all stakeholders (such as patients and their families, health professionals, care providers, and health regulators) to establish coordination, collaboration and a shared care approach between them to improve overall patient care safety. The paper outlines a Health Portal model for designing a real-time health prevention system. An application of the architecture is described in the area of web Health Portal.

The rising popularity of Web 2.0 applications has seen an increase in the volume of user-generated content. Web Applications allow users to define policies that specify how they wish their content to be accessed. In large Web 2.0 applications these policies can become quite complex, with users having to make decisions such as ‘who can access my image library?’, or ‘should my mobile number be made available to 3rd party agencies?’. As the policy size grows, the ability for everyday users to comprehend and manage their policy diminishes. This paper presents a model of policy configuration that harnesses the power of the Internet community by presenting average-sets of policy configuration. These policy “profiles” allow users to select a default set of policy values that line up with the average case, as presented by the application population. Policies can be promoted at an application level or at a group level. An XML approach is presented for representing the policy profiles. The approach allows for easy profile comparison and merging. A storage mechanism is also presented that describes how these policies should be made persistent in a distributed data storage system.

Spam or unwanted email is one of the potential issues of Internet security and classifying user emails correctly from penetration of spam is an important research issue for anti-spam researchers. In this paper we present an effective and efficient spam classification technique using clustering approach to categorize the features. In our clustering technique we use VAT (Visual Assessment and clustering Tendency) approach into our training model to categorize the extracted features and then pass the information into classification engine. We have used WEKA (www.cs.waikato.ac.nz/ml/weka/) interface to classify the data using different classification algorithms, including tree-based classifiers, nearest neighbor algorithms, statistical algorithms and AdaBoosts. Our empirical performance shows that we can achieve detection rate over 97%.

This paper describes the use of Elliptic Curve Cryptography (ECC) over Prime Field (F _p ) for encryption and digital signature of smart cards. The concepts of ECC over prime field (F _p ) are described, followed by the experimental design of the smart card simulation. Finally, the results are compared against RSA algorithms.

Wireless access technology has come a long way in its relatively short but remarkable lifetime, which has so far been led by WiFi technology. WiFi enjoys a high penetration in the market.Most of the electronic gadgets such as laptop, notepad, mobile set, etc., boast the provision ofWiFi. Currently most WiFi hotspots are connected to the Internet via wired connections (e.g., Ethernet), and the deployment cost of wired connection is high. On the other hand, since WiMAX can provide a high coverage area and transmission bandwidth, it is very suitable for the backbone networks of WiFi. WiMAX can also provide the better QoS needed for many 4G applications. WiMAX devices, however, are not as common as WiFi devices and it is also expensive to deploy aWiMAX-only infrastructure. An integrated WiMAX/WiFi architecture (using WiMAX as backhaul connection for WiFi) can support 4G applications with QoS assurance and mobility, and provide high-speed broadband services in rural, regional and urban areas while reducing the backhaul cost. WiMAX and WiFi have different MAC mechanisms to handle QoS. WiMAX MAC architecture is connection-oriented providing the platform for strong QoS control. In contrast,WiFi MAC is not connection-oriented, hence can provide only best effort services. Delivering improved QoS in an integrated WiMAX/WiFi architecture poses a serious technological challenge. The paper depicts a converged architecture of WiMAX and WiFi, and then proposes an adaptive resource distribution model for the access points. The resource distribution model ultimately allocates more time slots to those connections that need more instantaneous resources to meet QoS requirements. A dynamic splitting technique is also presented that divides the total transmission period into downlink and uplink transmission by taking the minimum data rate requirements of the connections into account. This ultimately improves the utilization of the available resources, and the QoS of the connections. Simulation results show that the proposed schemes significantly outperform the other existing resource sharing schemes, in terms of maintaining QoS of different traffic classes in an integratedWiMAX/WiFi architecture.

Robocup is a popular test bed for AI programs around the world. Robosoccer is one of the two major parts of Robocup, in which AIBO entertainment robots take part in the middle sized soccer event. The three key challenges that robots need to face in this event are manoeuvrability, image recognition and decision making skills. This paper focuses on the decision making problem in Robosoccer- The goal keeper problem. We investigate whether reinforcement learning (RL) as a form of semi-supervised learning can effectively contribute to the goal keeper’s decision making process when penalty shot and two attacker problem are considered. Currently, the decision making process in Robosoccer is carried out using rule-base system. RL also is used for quadruped locomotion and navigation purpose in Robosoccer using AIBO. Moreover the ball distance is being calculated using IR sensors available at the nose of the robot. In this paper, we propose a reinforcement learning based approach that uses a dynamic state-action mapping using back propagation of reward and Q-learning along with spline fit (QLSF) for the final choice of high level functions in order to save the goal. The novelty of our approach is that the agent learns while playing and can take independent decision which overcomes the limitations of rule-base system due to fixed and limited predefined decision rules. The spline fit method used with the nose camera was also able to find out the location and the ball distance more accurately compare to the IR sensors. The noise source and near and far sensor dilemma problem with IR sensor was neutralized using the proposed spline fit method. Performance of the proposed method has been verified against the bench mark data set made with Upenn’03 code logic and a base line experiment with IR sensors. It was found that the efficiency of our QLSF approach in goalkeeping was better than the rule based approach in conjunction with the IR sensors. The QLSF develops a semi-supervised learning process over the rule-base system’s input-output mapping process, given in the Upenn’03 code.