Nooritawati Md Tahir

Statistical analysis of parkinson disease gait classification using Artificial Neural Network

The aim of this study is to investigate the parameters that could be used to identify abnormal gait pattern in Parkinsons disease subjects during normal walking. Hence, three types of gait parameters namely basic, kinematic and kinetic are evaluated. Initial findings showed that the average mean of cadence, step length and walking speed for Parkinsons disease patients are lower than normal subjects, while the mean of stride time for Parkinsons disease patients are higher. Further, for kinematic parameter, overall joint angle of hip, knee and ankle mean values are lower for Parkinsons disease patients as compared to normal group. In addition, for kinetic parameter, all mean values of ground reaction force parameters are higher for normal subjects with walking speed contributed as the major determinant. To evaluate the significant features that could be used as identification between PD and normal subjects, statistical analysis is conducted. Hence, based on the statistical analysis results, it was found that step length, walking speed, knee angle as well as vertical parameter of ground reaction force are the four significant features as indicators for classification of subject with Parkinsons disease based on the accuracy attained with Artificial Neural Network as classifier.

Background modelling and background subtraction performance for object detection

Moving object detection in video applications is usually performed based on techniques such as background subtraction, optical flow and temporal differencing. The most popular literature technique approach to detect moving object from video sequences is background subtraction. This approach utilized mathematical model of static background and comparing it with every new frame of video sequence. In this paper, background subtraction technique using Mixture of Gaussian (MoG) method is conducted for detection of moving object at outdoor environment. Focus is specified at the five parameters of MoG namely background component weight threshold (TS), standard deviation scaling factor (D), user-define learning rate (α), Total number of Gaussian components (K) and Maximum number of components M in the background model (M) to give significant impact in producing the optimize background subtraction process. Experimental results showed that by varying each of the parameter can produce acceptable results that enable us to propose suitable parameter range of each parameter for detection of moving object in an outdoor environment.

An efficient false alarm reduction approach in HTTP-based botnet detection

In recent years, bots and botnets have become one of the most dangerous infrastructure to carry out nearly every type of cyber-attack. Their dynamic and flexible nature along with sophisticated mechanisms makes them difficult to detect. One of the latest generations of botnet, called HTTP-based, uses the standard HTTP protocol to impersonate normal web traffic and bypass the current network security systems (e.g. firewalls). Besides, HTTP protocol is commonly used by normal applications and services on the Internet, thus detection of the HTTP botnets with a low rate of false alarms (e.g. false negative and false positive) has become a notable challenge. In this paper, we review the current studies on HTTP-based botnet detection in addition to their shortcomings. We also propose a detection approach to improve the HTTP-based botnet detection regarding the rate of false alarms and the detection of HTTP bots with random patterns. The testing result shows that the proposed method is able to reduce the false alarm rates in HTTP-based botnet detection successfully.

Monitoring of Watermelon Ripeness Based on Fuzzy Logic

The aim of this study is to monitor watermelon ripeness based on image processing technique and fuzzy logic as classifier. The RGB color technique is utilized as the extracted features for the watermelon’s rind. Further, the extracted feature is classified using fuzzy logic system to determine the ripeness level of the watermelon. In addition, the same set of watermelons is graded by both human expert and fuzzy logic system for comparison purpose. Results attained demonstrate the ability of the proposed method in grading and classifying ripeness of watermelons.

Trajectory zero phase error tracking control using comparing coefficients method

This paper presents the studies on trajectory zero phase error tracking control without factorisation of zeros polynomial where the controller parameters are determined using comparing coefficients methods. The controller was applied to two types of third-order non-minimum phase plant. The first plant was having a zero outside and far from the unity circle. Another plant was having a zero outside and near to the unity circle. Simulation and experimental results will be presented to discuss its tracking performance.

Enhancement of background subtraction techniques using a second derivative in gradient direction filter

A new approach was proposed to improve traditional background subtraction (BGS) techniques by integrating a gradient-based edge detector called a second derivative in gradient direction (SDGD) filter with the BGS output. The four fundamental BGS techniques, namely, frame difference (FD), approximate median (AM), running average (RA), and running Gaussian average (RGA), showed imperfect foreground pixels generated specifically at the boundary. The pixel intensity was lesser than the preset threshold value, and the blob size was smaller. The SDGD filter was introduced to enhance edge detection upon the completion of each basic BGS technique as well as to complement the missing pixels. The results proved that fusing the SDGD filter with each elementary BGS increased segmentation performance and suited postrecording video applications. Evidently, the analysis using Fscore and average accuracy percentage proved this, and, as such, it can be concluded that this new hybrid BGS technique improved upon existing techniques.

Heat exchanger modeling using NARX model with binary PSO-based structure selection method

This paper explores the application of Non-Linear Autoregressive Model with Exogenous Inputs (NARX) system identification of heat exchanger system. Model structure selection was performed using the Binary Particle Swarm Optimization (BPSO) algorithm. The application of BPSO for model structure selection represents each particles position as binary values, which were used to select a set of regressors from the regressor matrix. Parameter estimation was then performed using Householder-based QR factorization method. Tests performed on the heat exchanger system defined the model with a maximum lag of five, while fulfilling all model validation criterions.

Feature selection of breast cancer based on Principal Component Analysis

Machine learning is a branch of artificial intelligence that employs a variety of statistical, probabilistic and optimization techniques and allows computer to learn from past examples and detect patterns from large data sets, which is particularly well-suited to assist medical practitioners in diagnosis of disease based on a variety of test results. Therefore, in this research, we deemed further by developing feature extraction algorithm based on Principal Component Analysis (PCA) and Artificial Neural Network (ANNs) as classifier as the optimal tool to enhance the classification of benign or malignant based on the Wisconsin Breast Cancer Database. In addition, the three rules of thumb of PCA namely the Scree Test, Cumulative Variance and the KG rule are employed as feature selection. An ensemble of the reduced datasets based on these rules is used as the inputs to ANN classifier with back propagation algorithm. Initial results showed that this approach is able to discriminate between the normal and breast cancer patients.

Classification of Parkinson's disease based on Multilayer Perceptrons Neural Network

Parkinsons disease (PD) is the second commonest late life neurodegenerative disease after Alzheimers disease. It is prevalent throughout the world and predominantly affects patients above 60 years old. It is caused by progressive degeneration of dopamine containing cells (neurons) within the deep structures of the brain called the basal ganglia and substantia nigra. Therefore, accurate prediction of PD need to be done in order to assist medical or bio-informatics practitioners for initial diagnose of PD based on variety of test results. This paper described the analysis conducted based on two training algorithms namely Levenberg-Marquardt (LM) and Scaled Conjugate Gradient (SCG) of Multilayer Perceptrons (MLPs) Neural Network in diagnosing PD. The dataset information of this project has been taken form the Parkinson Disease Data Set. Results attained confirmed that the LM performed well with accuracy rate of 92.95% while SCG obtained 78.21% accuracy.

Reduced set support vector machines: Application for 2-dimensional datasets

This paper presents the performance of the reduced set (RS) method to approximate the decision boundary for standard support vector machines (SVM) classifier without affecting its generalization performance. The main focus of this work is to demonstrate the capability of the RS method such that even with fewer set of vectors, the generalization performance is not affected. In evaluating the RS method performance, decision boundaries obtained using RS method were benchmarked against the decision boundaries obtained from the standard SVM using sequential minimal optimization (SMO) method. Specifically, the generalization ability of the two methods is not evaluated since the main objective is to analyze the effect of reduced set vector in producing approximation of SVM decision rules. Results obtained demonstrated that the SVM classifier using RS method is comparable with the standard SVM using SMO method. In fact, the RS method is better since it uses fewer set of vectors to produce similar decision boundaries while maintaining the generalization performances.