Nazrina Aziz

Robust influence angle for clustering mixed data sets

The great importance in attempting to identify clusters of observations which may be present in a data is how close observations are to each other. Two observations are close when their dissimilarity is small. Some traditional distance functions cannot capture the pattern dissimilarity among the observations. The other demand is the dissimilarity measurement should have the ability to deal with a variety of data types. This article proposed a new dissimilarity measure namely, Robust Influence Angle (RIA) based on eigenstructure of the covariance matrix and robust principal component score. The proposed measurement is able to identify cluster of observation and it also has the ability to handle data set with mixed variables.

Survival analysis of colorectal cancer patients with tumor recurrence using global score test methodology

Colorectal cancer is the third and the second most common cancer worldwide in men and women respectively, and the second in Malaysia for both genders. Surgery, chemotherapy and radiotherapy are among the options available for treatment of patients with colorectal cancer. In clinical trials, the main purpose is often to compare efficacy between experimental and control treatments. Treatment comparisons often involve several responses or endpoints, and this situation complicates the analysis. In the case of colorectal cancer, sets of responses concerned with survival times include: times from tumor removal until the first, the second and the third tumor recurrences, and time to death. For a patient, the time to recurrence is correlated to the overall survival. In this study, global score test methodology is used in combining the univariate score statistics for comparing treatments with respect to each survival endpoint into a single statistic. The data of tumor recurrence and overall survival of colorectal ...

An optimal inventory control in hybrid manufacturing/remanufacturing system with deteriorating and defective items

An optimal control model in a manufacturing/remanufacturing-inventory system, with defective, deteriorating, returned and disposed items was developed. In this model, the time of deterioration, and the percentages of defective, returned and disposed items were assumed random variables that follow the gamma distribution, beta rectangular, four-parameter generalised beta (FPGB) and Kumaraswamy distributions, respectively. The total of returned items from customers and defective items divided into the remanufacturing and disposed items. An optimality conditions were derived from the dynamic of the manufacturing/remanufacturing inventory level. The explicit solution under continuous-review policy was achieved by using the Pontryagin maximum principle. Also, simulation and sensitivity analysis results were illustrated numerically. The numerical results suggested that the model can help firms to maintain a balance between the manufacturing and remanufacturing rates to hedge demand and inventory levels. Also, the manufacturing rate was positively related to the disposal rate.

Deterministic SLIR model for tuberculosis disease mapping

Tuberculosis (TB) occurs worldwide. It can be transmitted to others directly through air when active TB persons sneeze, cough or spit. In Malaysia, it was reported that TB cases had been recognized as one of the most infectious disease that lead to death. Disease mapping is one of the methods that can be used as the prevention strategies since it can displays clear picture for the high-low risk areas. Important thing that need to be considered when studying the disease occurrence is relative risk estimation. The transmission of TB disease is studied through mathematical model. Therefore, in this study, deterministic SLIR models are used to estimate relative risk for TB disease transmission.

Heterogenous customer satisfaction index for evaluating university food service

This paper aims to measure the performance of university food service based on students’ perception. Two cafeterias were chosen for comparison: one located at student residential hall (Cafe 1) and another at the university administration centre (Cafe 2). By considering the components of importance and satisfaction, the Heterogeneous Customer Satisfaction Index-HCSI was computed to measure the performance of quality items in both cafeterias. Stratified sampling method was used to select 278 students and the DINESERVE instrument was used to assess customer perception on service quality. The findings show that the customer rate these two cafeterias as quite satisfied only, with the HCSI for Cafe 1 slightly higher than that for Cafe 2.

Tuberculosis disease mapping in Kedah using standardized morbidity ratio

This paper presents the results of relative risk estimation that applied to TB data in Kedah using the most common approach, Standardized Morbidity Ratio (SMR). Disease mapping has been recognized as one of the methods that can be used by government and public health in order to control diseases since it can give a clear picture of the risk areas. To get good disease mapping, relative risk estimation is an important issue that need to be considered. TB risk areas will be recognized through the map. From the result, Kulim shows the lowest risk areas of contracting TB while Kota Setar has the highest risk area.This paper presents the results of relative risk estimation that applied to TB data in Kedah using the most common approach, Standardized Morbidity Ratio (SMR). Disease mapping has been recognized as one of the methods that can be used by government and public health in order to control diseases since it can give a clear picture of the risk areas. To get good disease mapping, relative risk estimation is an important issue that need to be considered. TB risk areas will be recognized through the map. From the result, Kulim shows the lowest risk areas of contracting TB while Kota Setar has the highest risk area.

The performance of smoothed location model with PCA+Indicator MCA and PCA+Adjusted MCA

: Smoothed location model (SLM) is one of the discriminant analysis that can be used to deal with mixtures of continuous and binary variables simultaneously. However, SLM facing the problem in estimating parameters when the there is a large number of binary variables considered in the study. Thus, two variable extraction techniques, principal component analysis (PCA) and multiple correspondence analysis (MCA) are conducted together with SLM in order to solve the problems of many empty cells and parameters estimation. Simulation results showed that SLM along with PCA+Adjusted MCA performed better than SLM with PCA+ Indicator MCA even when the number of extracted binary is large.

Relative importance index (RII) in ranking of procrastination factors among university students

Procrastination is the action of delaying or postponing something such as making a decision or starting or completing some tasks or activities. According to previous studies, students who have a strong tendency to procrastinate get low scores in their tests, resulting in poorer academic performance compared to those who do not procrastinate. This study aims to identify the procrastination factors in completing assignments among three groups of undergraduate students. The relative importance of procrastination factors was quantified by the relative importance index (RII) method prior to ranking. A multistage sampling technique was used in selecting the sample. The findings revealed that ‘too many works in one time’ is one of the top three factors contributing to procrastination in all groups.

Clustering high dimensional data using RIA

Clustering may simply represent a convenient method for organizing a large data set so that it can easily be understood and information can efficiently be retrieved. However, identifying cluster in high dimensionality data sets is a difficult task because of the curse of dimensionality. Another challenge in clustering is some traditional functions cannot capture the pattern dissimilarity among objects. In this article, we used an alternative dissimilarity measurement called Robust Influence Angle (RIA) in the partitioning method. RIA is developed using eigenstructure of the covariance matrix and robust principal component score. We notice that, it can obtain cluster easily and hence avoid the curse of dimensionality. It is also manage to cluster large data sets with mixed numeric and categorical value.

Variables extraction on large binary variables in discriminant analysis based on mixed variables location model

The natural performance of the location model is a potential tool for allocating an object into one of the two observed groups involving mixtures of continuous and binary variables. In constructing location model, continuous variable is used to estimate parameters while binary variable is utilized to create segmentation in each group. Such segmentation is called as multinomial cells. Basically, the multinomial cells will grow exponentially according to the number of the binary variable. These multinomial cells will become empty when there is no object can be assigned into some of them. Then the occurring of empty cells will lead to unreliable parameter estimation. Consequently, the construction of the discriminant rule based on location model is impossible. Therefore, this paper attempts to discuss how the location model based on maximum likelihood estimation can be constructed even dealing with many measured binary variables. In other word, how is location model able to deal with the issue of many empty ...