Ensemble Methods in Environmental Data Mining

Abstract

Environmental data mining is the nontrivial process of identifying valid, novel, and potentially useful patterns in data from environmental sciences. This chapter proposes ensemble methods in environmental data mining that combines the outputs from multiple classification models to obtain better results than the outputs that could be obtained by an individual model. The study presented in this chapter focuses on several ensemble strategies in addition to the standard single classifiers such as decision tree, naive Bayes, support vector machine, and k-nearest neighbor (KNN), popularly used in literature. This is the first study that compares four ensemble strategies for environmental data mining: (i) bagging, (ii) bagging combined with random feature subset selection (the random forest algorithm), (iii) boosting (the AdaBoost algorithm), and (iv) voting of different algorithms. In the experimental studies, ensemble methods are tested on different real-world environmental datasets in various subjects such as air, ecology, rainfall, and soil.

A framework for predicting employee health risks using ensemble model

“…As stated by Tukey, using two regression models with the first model fitted to the data and the second to the right the errors from the first model would enhance the prediction outcome. Ensemble learning refers to a combination of learners who are trained to solve the same problem (Tuysuzoglu et al, 2017). The fundamental idea of ensemble learning was to combine weak learners into strong learners, with the ability to provide better generalization errors while reducing the over-fitting of outputs (Tuysuzoglu et al, 2017).…”

“…The focus is consistently driving towards identifying an algorithm that can perform accurate predictions or increase predictive accuracy-but there should be more focus put into solving the underlying issue of practicality and interpretability (Bates et al, 2014). Ensemble learning refers to a combination of learners who are trained to solve the same problem (Tuysuzoglu et al, 2017). It is a machine learning technique whereby the predictions are combined into a single output that potentially has a better performance than an individual model (Tuysuzoglu et al, 2017).…”

“…Ensemble learning refers to a combination of learners who are trained to solve the same problem (Tuysuzoglu et al, 2017). It is a machine learning technique whereby the predictions are combined into a single output that potentially has a better performance than an individual model (Tuysuzoglu et al, 2017). A practical framework to build ensemble models which can be applied by practitioners who are not experts is unclear while a comprehensive ensemble system framework may be too complex to understand (Abdunabi, 2016).…”

Classification of Firm External Audit Using Ensemble Support Vector Machine Method

“…The ensemble method can reduce classification errors effectively, and is believed to perform well compared to the use of a single classifier. The main idea of the ensemble method is to combine several sets of models that solve a similar problem to obtain a more accurate model [14]. Compared to an individual classifier, they only learn and train a set of data only.…”

“…Research [11] uses and compares three classification methods: Support Vector Machine (SVM), Decision Trees (DT), and Neural Network (NN) to predict auditor choices, based on evaluation of 10-Fold cross validation, Decision Tree outperforms two other classification methods , achieving an average accuracy rate of 83.73%. In addition, the classification method to provide good performance enhancements can use an ensemble model based on some training to solve the same problem and then the output of a single classification is combined into one classifier [12], [13], [14].…”