Azah Kamilah Muda

Single channel sEMG muscle fatigue prediction: An implementation using least square support vector machine

Surface electromyogram (sEMG) signal is commonly used for muscle fatigue analysis in clinical rehabilitation studies. Prediction results based on sEMG signals are promising because muscle contradiction can be easily characterized using sEMG signals. However, the prediction results usually deteriorate significantly when noise exist during data acquisition. Noise happens due to many factors ranging from hardware, software to procedure flaws. This investigation is aimed to assess the performance of the Least Square SVM model in predicting muscle fatigue using single channel sEMG signal. The root mean square, median frequency, and mean frequency features were extracted from two sets of raw sEMG signals captured at the multifidus (for low back pain) and flexor carpi radialis (for forearm muscle fatigue) muscles. The proposed LS-SVM technique were used to build the prediction rule-base separately for both the datasets. The implementation, testing and verification were performed in Matlab environment. The k-nearest neighbour and artificial neural network were used as the benchmarking techniques in results comparison and analysis. LS-SVM technique is proven good against the benchmarking techniques on classification accuracy and area under ROC curve. The ANOVA and Tukey HSD post hoc test were used to further validate the significant of the comparison results on both accuracy and AUC measurements.

SOCIFS feature selection framework for handwritten authorship

The uniqueness of shape and style of handwriting can be used to identify the significant features in confirming the author of writing. This paper is meant to propose a novel feature selection framework for Swarm Optimized and Computationally Inexpensive Floating Selection SOCIFS, by exploring existing feature selection frameworks, and compare the performance of proposed feature selection framework against various feature selection methods in Writer Identification in order to find the most significant features. The promising applicability of the proposed framework has been demonstrated in the result and worth to receive further exploration in identifying the handwritten authorship.

PSO and Computationally Inexpensive Sequential Forward Floating Selection in acquiring significant features for handwritten authorship

The uniqueness of shape and style of handwriting can be used to identify the significant features in confirming the author of writing. Acquiring these significant features leads to an important research in Writer Identification domain. This paper is meant to explore the usage of feature selection in Writer Identification in order to find the most significant features. This paper proposes a hybrid feature selection method of Particle Swarm Optimization and Computationally Inexpensive Sequential Forward Floating Selection for Writer Identification. The promising applicability of the proposed method has been demonstrated and worth to receive further exploration in identifying the handwritten authorship.

A comparative study of feature extraction using PCA and LDA for face recognition

Feature extraction is important in face recognition. This paper presents a comparative study of feature extraction using Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA) for face recognition. The evaluation parameters for the study are time and accuracy of each method. The experiments were conducted using six datasets of face images with different disturbance. The results showed that LDA is much better than PCA in overall image with various disturbances. While in time taken evaluation, PCA is faster than LDA.

Embedded scale united moment invariant for identification of handwriting individuality

Past few years, a lot of research on moment functions have been explored in pattern recognition. Several new techniques have been investigated to improve conventional regular moment by proposing the scaling factor of geometrical function. In this paper, integrated scaling formulations of Aspect Invariant Moment and Higher Order Scaling Invariant with United Moment Invariant are presented in Writer Identification to seek the invarianceness of authorship or individuality of handwriting perseverance. Mathematical proving and results of computer simulations are included to verify the validity of the proposed technique in identifying eccentricity of the author in Writer Identification.

Comparing features extraction methods for person authentication using EEG signals

This chapter presents a comparison and analysis of six feature extraction methods which were often cited in the literature, namely wavelet packet decomposition (WPD), Hjorth parameter, mean, coherence, cross-correlation and mutual information for the purpose of person authentication using EEG signals. The experimental dataset consists of a selection of 5 lateral and 5 midline EEG channels extracted from the raw data published in UCI repository. The experiments were designed to assess the capability of the feature extraction methods in authenticating different users. Besides, the correlation-based feature selection (CFS) method was also proposed to identify the significant feature subset and enhance the authentication performance of the features vector. The performance measurement was based on the accuracy and area under ROC curve (AUC) values using the fuzzy-rough nearest neighbour (FRNN) classifier proposed previously in our earlier work. The results show that all the six feature extraction methods are promising. However, WPD will induce large vector set when the selected EEG channels increases. Thus, the feature selection process is important to reduce the features set before combining the significant features with the other small feature vectors set.

Computational Intelligence in Digital Forensics: Forensic Investigation and Applications

Computational Intelligence techniques have been widely explored in various domains including forensics. Analysis in forensic encompasses the study of pattern analysis that answer the question of interest in security, medical, legal, genetic studies and etc. However, forensic analysis is usually performed through experiments in lab which is expensive both in cost and time. Therefore, this book seeks to explore the progress and advancement of computational intelligence technique in different focus areas of forensic studies. This aims to build stronger connection between computer scientists and forensic field experts. This book, Computational Intelligence in Digital Forensics: Forensic Investigation and Applications, is the first volume in the Intelligent Systems Reference Library series. The book presents original research results and innovative applications of computational intelligence in digital forensics. This edited volume contains seventeen chapters and presents the latest state-of-the-art advancement of Computational Intelligence in Digital Forensics; in both theoretical and application papers related to novel discovery in intelligent forensics. The chapters are further organized into three sections: (1) Introduction, (2) Forensic Discovery and Investigation, which discusses the computational intelligence technologies employed in Digital Forensic, and (3) Intelligent Forensic Science Applications, which encompasses the applications of computational intelligence in Digital Forensic, such as human anthropology, human biometrics, human by products, drugs, and electronic devices.

Computational Intelligence in Digital Forensics

Forensic Science has been around for quite some time. Although various forensic methods have been proved for their reliability and credibility in the criminal justice system, their main problem lies in the necessity of highly qualified forensic investigators. In the course of analysis of evidences, forensic investigators must be thorough and rigorous, hence time consuming. Digital Forensic techniques have been introduced to aid the forensic investigators to acquire as reliable and credible results as manual labor to be presented in the criminal court system. In order to perform the forensic investigation using Digital Forensic techniques accurately and efficiently, computational intelligence oftentimes employed in the implementation of Digital Forensic techniques, which has been proven to reduce the time consumption, while maintaining the reliability and credibility of the result, moreover in some cases, it is producing the results with higher accuracy. The introduction of computational intelligence in Digital Forensic has attracted a vast amount of researchers to work in, and leads to emergence of numerous new forensic investigation domains.

Measuring data completeness for microbial genomics database

Poor quality data such as data with missing values (or records) cause negative consequences in many application domains. An important aspect of data quality is completeness. One problem in data completeness is the problem of missing individuals in data sets. Within a data set, the individuals refer to the real world entities whose information is recorded. So far, in completeness studies however, there has been little discussion about how missing individuals are assessed. In this paper, we propose the notion of population-based completeness (PBC) that deals with the missing individuals problem, with the aim of investigating what is required to measure PBC and to identify what is needed to support PBC measurements in practice. This paper explores the need of PBC in the microbial genomics where real sample data sets retrieved from a microbial database called Comprehensive Microbial Resources are used (CMR).

A framework of rough reducts optimization based on PSO/ACO hybridized algorithms

Rough reducts has contributed significantly in numerous researches of feature selection analysis. It has been proven as a reliable reduction technique in identifying the importance of attributes set in an information system. The key factor for the success of reducts calculation in finding minimal reduct with minimal cardinality of attributes is an NP-Hard problem. This paper has proposed an improved PSO/ACO optimization framework to enhance rough reduct performance by reducing the computational complexities. The proposed framework consists of a three-stage optimization process, i.e. global optimization with PSO, local optimization with ACO and vaccination process on discernibility matrix.