A COMPARATIVE STUDY OF THE MULTIPLE LOGISTIC REGRESSION, LINEAR DISCRIMINANT ANALYSIS AND QUADRATIC DISCRIMINANT FOR ESTIMATING THE MISCLASSIFICATION ERROR RATE OF INFANT BIRTH OUTCOME

This study examined comparison of the Multiple logistic regression, Linear discriminant analysis and Quadratic discriminant in estimating the infant birth outcome and misclassification error rate of birth outcomes with factors of infant mortality in Anambra State, Nigeria. The birth outcomes of interest were the Neonatal death, Still birth and Alive. Secondary source of data were obtained from the records department of General Hospital Onitsha from 2007-2016. The data comprises of Status of infant birth, Mothers parity, Age of mother, Weight of baby, Mothers Education Status, Number of Bookings before gestation and Gestation Age. The data analysis is performed using R-software. The result of the findings from the multiple logistic regression showed that Mothers Education Status (MES) and Booking contributed significantly on the logistic model while factors of Parity, Sex, Age of Mother (AOM), Year, GA and Birth Weight (BW) were found to be insignificant on birth outcomes. Also observed that the misclassification error rate for birth outcome for the said approach is found to be 0.5992 (59.92%). More so, findings of the study equally showed that the prior probabilities of the groups for the linear and quadratic discriminant analysis were 0.228503, 0.40168 and 0.36981 for Alive, Neonatal death and Still birth respectively. Further findings revealed that the Mothers Education Status and Bookings Status have the greatest impact for first and second linear function respectively. In addition, the result of the misclassification error rate for birth outcome using the linear discriminant analysis is 0.5931 (59.31%). The misclassification error rate for birth outcome based on quadratic discriminant analysis is 0.5956 (59.56%). Based on the findings of this study, linear discriminant approach is the best alternative in estimating misclassification error rate of infant birth outcome followed by quadratic discriminant analysis and the least is multiple logistic regression. The findings clearly confirmed that the linear discriminant analysis is the best with misclassification error rate of 59.31%.

Error-rate estimation in discriminant analysis of non-linear longitudinal data: A comparison of resampling methods

Consider longitudinal observations across different subjects such that the underlying distribution is determined by a non-linear mixed-effects model. In this context, we look at the misclassification error rate for allocating future subjects using cross-validation, bootstrap algorithms (parametric bootstrap, leave-one-out, .632 and [Formula: see text]), and bootstrap cross-validation (which combines the first two approaches), and conduct a numerical study to compare the performance of the different methods. The simulation and comparisons in this study are motivated by real observations from a pregnancy study in which one of the main objectives is to predict normal versus abnormal pregnancy outcomes based on information gathered at early stages. Since in this type of studies it is not uncommon to have insufficient data to simultaneously solve the classification problem and estimate the misclassification error rate, we put special attention to situations when only a small sample size is available. We discuss how the misclassification error rate estimates may be affected by the sample size in terms of variability and bias, and examine conditions under which the misclassification error rate estimates perform reasonably well.

Adaptif Range-Constrained Otsu Untuk Pemilihan Threshold Secara Otomatis Pada Histogram Citra Dengan Variansi Kelas Yang Tidak Seimbang

Abstrak Image Thresholding merupakan proses segmentasi untuk pemisahkan foreground dan background pada citra dengan cara membagi histogram citra menjadi dua kelas. Beberapa metode thresholding seperti Otsu dan Range-constrained Otsu menggunakan nilai variansi dari histogram untuk mendapatkan titik threshold, namun ketika menangani citra yang memiliki nilai variansi kelas foreground dan background tidak seimbang titik threshold yang dihasilkan kurang tepat. Paper ini mengusulkan metode Adaptif Range-constrained Otsu untuk mengatasi permasalahan variansi kelas yang tidak seimbang dengan cara mencari kelas yang memiliki nilai variansi lebih besar, untuk mendapatkan titik threshold yang lebih tepat. Pengujian menggunakan 22 NDT image dengan evaluasi misclassification error rate dan metode perankingan menunjukkan metode ini menghasilkan rerata ME 0.1153. Sedangkan Otsu sebesar 0.1746. Nilai rerata ranking 3.55, selisih 0.05 dibanding Kittler III. Hasil ini menunjukkan metode yang diusulkan kompetitif, terutama untuk segmentasi citra yang memiliki variansi kelas tidak sama. Kata kunci: segmentasi, thresholding, histogram, Otsu, Range-constrained. Abstract Image thresholding is segmentation process for separating foreground and background of an image by dividing image histogram into two classes. Several thresholding methods like Otsu and Rangeconstrained Otsu using the variance value of the histogram to get the threshold point, but when handling images that have unbalance class variance of the foreground and background produce less accurate threshold point. This paper proposes a method Adaptive Range-constrained Otsu to solve unbalance class variance problem by finding a class that has greater variance value to obtain more accurate threshold point. NDT testing using 22 images with misclassification error rate evaluation and ranking methods shows that this method results ME average of 0.1153, while Otsu method results 0.1746. The rankings mean value is 3.55, which has the difference of 0.05 when compared with Kittler III. These results show that the proposed method is competitive, especially for image segmentation with different class variance. Key word: segmentasi, thresholding, histogram, Otsu, Range-constrained.

MEASURING THE PERFORMANCE OF ORDINAL CLASSIFICATION

Ordinal classification is a form of multiclass classification for which there is an inherent order between the classes, but not a meaningful numeric difference between them. The performance of such classifiers is usually assessed by measures appropriate for nominal classes or for regression. Unfortunately, these do not account for the true dimension of the error. The goal of this work is to show that existing measures for evaluating ordinal classification models suffer from a number of important shortcomings. For this reason, we propose an alternative measure defined directly in the confusion matrix. An error coefficient appropriate for ordinal data should capture how much the result diverges from the ideal prediction and how "inconsistent" the classifier is in regard to the relative order of the classes. The proposed coefficient results from the observation that the performance yielded by the Misclassification Error Rate coefficient is the benefit of the path along the diagonal of the confusion matrix. We carry out an experimental study which confirms the usefulness of the novel metric.

A New Information Entropy-Based Ant Clustering Algorithm

Ant-based clustering is a heuristic clustering method that draws its inspiration from the behavior of ants in nature. We revisit these methods in the context of a concrete application and introduce some modifications that yield significant improvements in terms of both quality and efficiency. In this paper, we propose a New Information Entropy-based Ant Clustering (NIEAC) algorithm. Firstly, we apply new information entropy to model behaviors of agents, such as picking up and dropping objects. The new entropy function led to better quality clusters than non-entropy functions. Secondly, we introduce a number of modifications that improve the quality of the clustering solutions generated by the algorithm. We have made some experiments on real data sets and synthetic data sets. The results demonstrate that our algorithm has superiority in misclassification error rate and runtime over the classical algorithm.