Noor Azah Samsudin

Nearest neighbour group-based classification

The purpose of group-based classification (GBC) is to determine the class label for a set of test samples, utilising the prior knowledge that the samples belong to same, but unknown class. This can be seen as a simplification of the well studied, but computationally complex, non-sequential compound classification problem. In this paper, we extend three variants of the nearest neighbour algorithm to develop a number of non-parametric group-based classification techniques. The performances of the proposed techniques are then evaluated on both synthetic and real-world data sets and their performance compared with techniques that label test samples individually. The results show that, while no one algorithm clearly outperforms all others on all data sets, the proposed group-based classification techniques have the potential to outperform the individual-based techniques, especially as the (group) size of the test set increases. In addition, it is shown that algorithms that pool information from the whole test set perform better than two-stage approaches that undertake a vote based on the class labels of individual test samples.

E-Code Checker Application

Nowadays, many areas in computer sciences use ontology such as knowledge engineering, software reuse, digital libraries, web on the heterogeneous information processing, semantic web, and information retrieval. The area of halal industry is the fastest growing global business across the world. The halal food industry is thus crucial for Muslims all over the world as it serves to ensure them that the food items they consume daily are syariah compliant. However, ontology has still not been used widely in the halal industry. Today, Muslim community still have problem to verify halal status for halal products in the market especially in foods consisting of E number. In this paper, ontology will apply at E numbers as a method to solve problems of various halal sources. There are various chemical ontology and databases found to help this ontology construction. The E numbers in this chemical ontology are codes for chemicals that can be used as food additives. With this E numbers ontology, Muslim community could identify and verify the halal status effectively for halal products in the market.

A customizable wireless food ordering system with realtime customer feedback

The existence of wireless technology and the emergence of mobile devices enable a simple yet powerful infrastructure for business application. Some early efforts have been made to utilize both technologies in food ordering system implementations. However, the food ordering systems that have been proposed earlier exhibit limitations, primarily in cost effectiveness, allowing customizations and supporting real-time feedback to customers. In this paper, we discuss the design and implementation of a customizable wireless food ordering system with real-time customer feedback for a restaurant (CWOS-RTF). The CWOS-RTF enables restaurant owners to setup the system in wireless environment and update menu presentations easily. Smart phone has been integrated in the CWOS-RTF implementation to facilitate real-time communication between restaurant owners and customers. A preliminary testing suggests that the CWOS-RTF has the potential to eliminate the limitations of existing food ordering systems.

Group-based meta-classification

Virtually all existing classification techniques label one sample at a time. In this paper, we highlight the potential benefits of group based classification (GBC), where the classifier labels a group of homogeneous samples. In this way, GBC can take advantage of the additional prior knowledge that all samples belong to the same, unknown, class. We pose GBC in a generic hypothesis testing framework requiring the selection of an appropriate sample and test statistic. We then evaluate one simple example of GBC on both synthetic and real data sets and demonstrate that GBC may be a promising approach in applications where the test data can be arranged into homogenous subsets.

A Review: Image Analysis Techniques to Improve Labeling Accuracy of Medical Image Classification

Medical images contain the Region of Interest (ROI) from the affected area in human body and provide useful information to support clinical decision-making for diagnostics as well as the treatment planning. Unfortunately, medical image data may contain noise, missing values, inhomogeneous ROI that may give inaccurate diagnostic. Therefore, image analysis techniques are needed to improve the quality of an image. Then, features extraction task will be performed to produce best feature of images which leads to better classification result for accurate diagnostic. Many techniques have been used for image analysis. However, limited review have been done in categorize the list of related techniques for each image analysis task in medical imaging application. Thus, the aims of this paper is to gather and present general overview of image analysis task and their techniques in order to inspire researcher, pathologist or radiologist to adapt it when analyzing different types of medical image. The current study of image analysis task was summarized and discussed in this paper.

Extended Local Mean-Based Nonparametric Classifier for Cervical Cancer Screening

Malignancy associated changes approach is one of possible strategies to classify a Pap smear slide as positive (abnormal) or negative (normal) in cervical cancer screening procedure. The malignancy associated changes (MAC) approach acquires analysis of the cells as a group as the abnormal phenomenon cannot be detected at individual cell level. However, the existing classification algorithms are limited to automation of individual cell analysis task as in rare event approach. Therefore, in this paper we apply extended local-mean based nonparametric classifier to automate a group of cells analysis that is applicable in MAC approach. The proposed classifiers extend the existing local mean-based nonparametric techniques in two ways: voting and pooling schemes to label each patient’s Pap smear slide. The performances of the proposed classifiers are evaluated against existing local mean-based nonparametric classifier in terms of accuracy and area under receiver operating characteristic curve (AUC). The extended classifiers show favourable accuracy compared to the existing local mean-based nonparametric classifier in performing the Pap smear slide classification task.

Extended naïve bayes for group based classification

This paper focuses on extending Naive Bayes classifier to address group based classification problem. The group based classification problem requires labeling a group of multiple instances given the prior knowledge that all the instances of the group belong to same unknown class. We present three techniques to extend the Naive Bayes classifier to label a group of homogenous instances. We then evaluate the extended Naive Bayes classifier on both synthetic and real data sets and demonstrate that the extended classifiers may be a promising approach in applications where the test data can be arranged into homogenous subsets.

A Group-Based Feature Selection Approach to Improve Classification of Holy Quran Verses

Most existing feature selection approach is limited to determine features from a single source of data. In this paper, a feature selection approach is proposed to consider multiple sources of textual data. The proposed GBFS approach is then applied to label Quranic verses based on two major references, the English translation and tafsir (Commentary). The verses were selected from two chapters, Surah Al-Baqarah and Surah Al-Anaam. The verses are classified into three categories: Faith, Worship, and Etiquette. The textual data from the translation and commentary were preprocessed using StringToWord Vector with weighted TF-IDF. Feature selection algorithms: information gain, chi square, Pearson correlation coefficient, relief, and correlation-based were experimented on four classifiers: naive Bayes, libSVM, k-NN, and decision trees (J48). The proposed group-based feature selection approach has shown promising results in terms of Accuracy and Area under Receiver Operating Characteristics (ROC) curve (AUC) by achieving Accuracy of 94.5% and AUC of 0.944.

Medipre: medical diagnosis prediction using rough-regression approximation

Rough set and regression approximations are useful in establishing decision support system for medical diagnostic applications. However, the data elimination strategy for unclassified elements or patients in the medical diagnostic applications remains as a serious issue to be explored, especially with the aim of achieving higher prediction accuracy. This paper presents step-by-step procedure in building rough-regression approximation based on data elimination strategy. A number of data sets is used to examine our proposed approximation. The result has shown that the proposed rough-regression is capable to improve the prediction accuracy if compared with the existing approximations significantly. The proposed approximation can improve the performance of medical diagnosis prediction system. Therefore, it may help inexperienced doctors and patients for preliminary diagnosis.

An Enhanced Possibilistic Programming Model with Fuzzy Random Confidence-Interval for Multi-objective Problem

Mathematical models are established to represent real-world problems. Since the real-world faces various types of uncertainties, it makes mathematical model suffers with insufficient uncertainties modeling. The existing models lack of explanation in dealing uncertainties. In this paper, construction of mathematical model for decision making scenario with uncertainties is presented. Primarily, fuzzy random regression is applied to formulate a corresponding mathematical model from real application of a multi-objective problem. Then, a technique in possibilistic theory, known as modality optimization is used to solve the developed model. Consequently, the result shows that a well-defined multi-objective mathematical model is possible to be formulated for decision making problems with the uncertainty. Indeed, such problems with uncertainties can be solved efficiently with the presence of modality optimization.