A multi-class classification system for continuous water quality monitoring

Abstract

The issue addressed in this exposition is the classification of multivariate data collected through different sensors for water quality monitoring. Multivariate data are sequences that have various attributes in every instance of the sequences. A few endeavours exist to address this issue; however, none of them has given full emphasis on continuous dataset. Another solution for this issue is to reduce the instances to a single attribute while losing significant information. Different arrangements address both the multivariate and the sequential part of the data yet give an un-versatile solution. The proposed algorithm is not only able to monitor continuous water quality, but it also produces a better classification model for other continuous datasets as well. Instead of decreasing the attributes of the dataset, we introduce three additional reference indicators which are dependent on the actual attributes. We compare the classification accuracy of our proposed algorithm with standard classification models. The proposed method gives better classification accuracy compared to existing methods.

Machine Learning Approaches for Predicting Health Risk of Cyanobacterial Blooms in Northern European Lakes

“…Accuracy is defined as the ratio of all True Positives of the matrix divided by the sum of all instances in the dataset and it expresses the ability of the model to correctly identify Low, Medium and High instances. Kappa statistics considers the fact that some of the correct predictions may be identified as such by chance, so it adjusts the reported model accuracy by considering the effect of randomness in correct predictions [61]. Comparing the four implemented machine learning algorithms, in Figure 6 we see that both for accuracy and Kappa statistics, the RF performs best.…”

“…(a) (b) To further analyze the modeling results, we compared in detail the performance of the four algorithms by using three metrics: precision, recall and specificity [61]. Precision is defined for each class (Low, Medium and High) as the ratio of True Positives by the sum of True Positives and False Positives for that class.…”

An implentation of IoT for environmental monitoring and its analysis using k-NN algorithm

“…IoT is produced a lot from applications such as to conduct human health monitoring in real time which is better known as e-Health [2], military [3], smart city [4], [5], agriculture [6], [7], and so on [8]. The various applications of IoT technology are used to make predictions using various methods such as fuzzy logic [9]- [11], support vector machine (SVM) [12], decision tree [13], and so on.…”

Environmental informatics for modelling of freshwater ecosystems

“…Kappa statistics considers the fact that some of the correct predictions may be identified as such by chance, so it adjusts the reported model accuracy by considering the effect of randomness in correct predictions (Shakhari and Banerjee, 2019). Comparing the four implemented machine learning algorithms, in Figure 4.6 we see that both for accuracy and Kappa statistics, the RF performs best.…”

“…According to (Landis and Koch, 1977), a model is considered to produce accurate predictions when Kappa exceeds 60%, which in our case is succeeded. To further analyze the modelling results, the performance of the four algorithms was compared in detail by using three metrics: precision, recall and specificity (Shakhari and Banerjee, 2019). Precision is defined for each class (Low, Medium and High) as the ratio of True Positives by the sum of True Positives and False Positives for that class.…”