China
Africa
Explore chapters and articles related to this topic
Data Analysis
Published in Marian (Editor-in-Chief) Muste, Dennis A. Lyn, David M. Admiraal, Robert Ettema, Vladimir Nikora, Marcelo H. Garcia, Experimental Hydraulics: Methods, Instrumentation, Data Processing and Management, 2017
Marian (Editor-in-Chief) Muste, Dennis A. Lyn, David M. Admiraal, Robert Ettema, Vladimir Nikora, Marcelo H. Garcia
Another traditional linear approach, termed linear discriminant analysis (LDA), shares similarities with logistic regression in the model equation, but makes stronger assumptions regarding the probability distributions characterizing each group to be classified. The stronger assumptions allow a closed-form solution, and whereas logistic regression can become unstable in problems with perfectly linearly separable groups, the LDA approach will remain stable. LDA also has an appealing geometric interpretation in terms of the centroids of each group and the distances of a point from the centroids. Hastie et al. (2009) compares the two methods and their advantages and weaknesses. Both logistic and LDA are frequently formulated explicitly in terms of Bayes’ theorem, and so may be considered Bayesian in approach.
Affective Natural Interaction Using EEG: Technologies, Applications and Future Directions
Published in Spyrou Evaggelos, Iakovidis Dimitris, Mylonas Phivos, Semantic Multimedia Analysis and Processing, 2017
Charline Hondrou, George Caridakis, Kostas Karpouzis, Stefanos Kollias
Linear discriminant analysis (LDA) is used in statistics, pattern recognition, and machine learning to find a linear combination of features that characterizes or separates two or more classes of objects or events. The resulting combination may be used as a linear classifier, or for dimensionality reduction before later classification. It attempts to express one dependent variable as a linear combination of other features or measurements [27]. For this method, the independent variables have to be normally distributed, and the dependent variable has to be a categorical variable (i.e., the class label). An example of this technique can be seen in [314].
Deep Learning in Brain Segmentation
Published in Saravanan Krishnan, Ramesh Kesavan, B. Surendiran, G. S. Mahalakshmi, Handbook of Artificial Intelligence in Biomedical Engineering, 2021
Linear discriminant analysis (LDA) is a machine learning method that uses a linear combination of features to determine the class of the inputs. The main principle of LDA can be summarized as projecting data from the extracted feature space onto a single dimension space. The single dimension space can then be classified into two classes by thresholding. The major disadvantage of using LDA is the linear nature of the classifier. In most practical cases, the data cannot be separated in the feature space by a linear function.
Rapid seismic damage assessment using machine learning methods: application to a gantry crane
Published in Structure and Infrastructure Engineering, 2023
Qihui Peng, Wenming Cheng, Hongyu Jia, Peng Guo, Kang Jia
Linear discriminant analysis (LDA) and quadratic discriminant analysis (QDA) (Mika, Ratsch, Weston, Scholkopf, & Mullers, 1999) are two classic classifiers with, as their names imply, a linear and a quadratic decision surface, respectively. LDA attempts to express one dependent variable as a linear combination of other features assuming that the variances among group variables are the same across levels of predictors. Analogously, QDA is related to LDA, where it is assumed that the features from each class are normally distributed but without the assumption that the covariance of each class is identical. These methods are attractive with closed-form solutions that can be easily computed and having no hyperparameter to tune. In this paper, however, they do not work well.
Performance Analysis of Artificial Neural Network for Hand Movement Detection from EMG Signals
Published in IETE Journal of Research, 2022
Angana Saikia, Sushmi Mazumdar, Nitin Sahai, Sudip Paul, Dinesh Bhatia
The purpose of discriminant analysis is to classify objects into one of two or more groups based on a set of features that describe the objects. If one can assume that the groups are linearly separable, one can use a linear discriminant model. Linearly separable suggests that the groups can be separated by a linear combination of features that describe the objects. Linear discriminant analysis (LDA) is a commonly used technique for data classification and dimensionality reduction. LDA is also known as Fisher’s discriminant analysis and it searches for those vectors in the underlying space that best discriminate among classes [32,34]. The objective of LDA is to perform dimensionality reduction while preserving as much of the class discriminatory information as possible.
Artificial neural network-based prediction of field permeability of hot mix asphalt pavement layers
Published in International Journal of Pavement Engineering, 2020
M. K. Nivedya, Rajib B. Mallick
Therefore, the 9.5 and the 12.5 mm NMAS mixes were used in the classification problem analysis. Of the various supervised machine learning methods, the Linear Discriminant and the Logistic Regression (LR) methods were found to give the best accuracy, as evident from their confusion matrices, shown in Figure 7. In Linear Discriminant analysis, the data is assumed to have the same covariance and the probability density function is assumed to be normally distributed. However, in logistic regression, no such assumptions on the distribution of data is made. Note that the classification problem results in a model that is similar to that from statistical analysis, with a prediction function that could be used with new data to predict the quality.