scholarly journals Accessing Imbalance Learning Using Dynamic Selection Approach in Water Quality Anomaly Detection

Symmetry ◽  
2021 ◽  
Vol 13 (5) ◽  
pp. 818
Author(s):  
Eustace M. Dogo ◽  
Nnamdi I. Nwulu ◽  
Bhekisipho Twala ◽  
Clinton Aigbavboa
Keyword(s):  
Water Quality ◽  
Anomaly Detection ◽  
Distribution Systems ◽  
Ensemble Classification ◽  
Imbalanced Datasets ◽  
Dynamic Selection ◽  
Ensemble Models ◽  
Classifier Selection ◽  
Local Accuracy ◽  
Homogeneous Ensemble

Automatic anomaly detection monitoring plays a vital role in water utilities’ distribution systems to reduce the risk posed by unclean water to consumers. One of the major problems with anomaly detection is imbalanced datasets. Dynamic selection techniques combined with ensemble models have proven to be effective for imbalanced datasets classification tasks. In this paper, water quality anomaly detection is formulated as a classification problem in the presences of class imbalance. To tackle this problem, considering the asymmetry dataset distribution between the majority and minority classes, the performance of sixteen previously proposed single and static ensemble classification methods embedded with resampling strategies are first optimised and compared. After that, six dynamic selection techniques, namely, Modified Class Rank (Rank), Local Class Accuracy (LCA), Overall-Local Accuracy (OLA), K-Nearest Oracles Eliminate (KNORA-E), K-Nearest Oracles Union (KNORA-U) and Meta-Learning for Dynamic Ensemble Selection (META-DES) in combination with homogeneous and heterogeneous ensemble models and three SMOTE-based resampling algorithms (SMOTE, SMOTE+ENN and SMOTE+Tomek Links), and one missing data method (missForest) are proposed and evaluated. A binary real-world drinking-water quality anomaly detection dataset is utilised to evaluate the models. The experimental results obtained reveal all the models benefitting from the combined optimisation of both the classifiers and resampling methods. Considering the three performance measures (balanced accuracy, F-score and G-mean), the result also shows that the dynamic classifier selection (DCS) techniques, in particular, the missForest+SMOTE+RANK and missForest+SMOTE+OLA models based on homogeneous ensemble-bagging with decision tree as the base classifier, exhibited better performances in terms of balanced accuracy and G-mean, while the Bg+mF+SMENN+LCA model based on homogeneous ensemble-bagging with random forest has a better overall F1-measure in comparison to the other models.

scholarly journals Experimental Study and Comparison of Imbalance Ensemble Classifiers with Dynamic Selection Strategy

Entropy ◽  
2021 ◽  
Vol 23 (7) ◽  
pp. 822
Author(s):  
Dongxue Zhao ◽  
Xin Wang ◽  
Yashuang Mu ◽  
Lidong Wang
Keyword(s):  
Classification Performance ◽  
Ensemble Classification ◽  
Selection Strategy ◽  
Ensemble Classifiers ◽  
Imbalanced Datasets ◽  
Dynamic Selection ◽  
Imbalanced Classification ◽  
Ensemble Techniques ◽  
Imbalance Learning ◽  
Ensemble Algorithms

Imbalance ensemble classification is one of the most essential and practical strategies for improving decision performance in data analysis. There is a growing body of literature about ensemble techniques for imbalance learning in recent years, the various extensions of imbalanced classification methods were established from different points of view. The present study is initiated in an attempt to review the state-of-the-art ensemble classification algorithms for dealing with imbalanced datasets, offering a comprehensive analysis for incorporating the dynamic selection of base classifiers in classification. By conducting 14 existing ensemble algorithms incorporating a dynamic selection on 56 datasets, the experimental results reveal that the classical algorithm with a dynamic selection strategy deliver a practical way to improve the classification performance for both a binary class and multi-class imbalanced datasets. In addition, by combining patch learning with a dynamic selection ensemble classification, a patch-ensemble classification method is designed, which utilizes the misclassified samples to train patch classifiers for increasing the diversity of base classifiers. The experiments’ results indicate that the designed method has a certain potential for the performance of multi-class imbalanced classification.

TEKNOLOGI PENYEDIAAN AIR MINUM UNTUK KEADAAN TANGGAP DARURAT

2018 ◽  
Vol 6 (1) ◽  
Author(s):  
Arie Herlambang
Keyword(s):  
Water Quality ◽  
Emergency Response ◽  
Distribution Systems ◽  
Treatment Plant ◽  
Piping System ◽  
Process Technology ◽  
Treatment Technology ◽  
Disaster Victims ◽  
Ultra Filtration ◽  
Quality Water

In the event of natural disasters such as earthquakes, tsunamis, landslides, floods and droughts, water occupies a key role in disaster relief. The presence of water is important for drinking, cooking and support the refugee areas of environmental sanitation and avoiding disaster victims of diseases waterborn disease. Water problem in disaster conditions may occur partly as a result: the disturbance of water sources because change of water quality, to become turbid or salty, the destruction of a piping system, treatment plant damage, disruption of distribution systems, or the scarcity of water in evacuation areas. Introduction of water quality becomes important to determine which process technology will be used and saved investments in emergency conditions. Priority handling of clean water usually comes first in the refugee areas with communal system, because the need of water for bathing, washing and toilet is big enough, while for a drink in the early events during disaster dominated by bottled water, but for their long-term, they have to boil water. For remote areas and difficult to reach individuals who usually use  system more simple and easily operated. Water Supply Technology for emergency response has the characteristic 1). Able to operate with all sorts of water conditions (flexible adaptable), 2). Can be operated easily, 3). Does not require much maintenance, 4). Little use of chemicals, and 5). Portable and easy removable (Mobile System). Keywords :  Water Quality, Water Treatment Technology, Drinking Water, Emergency Response, filtration, ceramic filtration, Ultra filtration, Reverse Osmosis, Ultraviolet Sterilizer, Ozonizer, Disinfection.

Flat Flexible Thin Milli-electrode Array for Real-time in situ Water Quality Monitoring in Distribution Systems

2017 ◽  
Vol 2017 (4) ◽  
pp. 5598-5617
Author(s):  
Zhiheng Xu ◽  
Wangchi Zhou ◽  
Qiuchen Dong ◽  
Yan Li ◽  
Dingyi Cai ◽  
...  
Keyword(s):  
Water Quality ◽  
Real Time ◽  
Distribution Systems ◽  
Electrode Array ◽  
Water Quality Monitoring ◽  
Quality Monitoring

Sensory aspects of drinking water in contact with epoxy lined copper pipe

2007 ◽  
Vol 55 (5) ◽  
pp. 161-168 ◽  
Author(s):  
T.H. Heim ◽  
A.M. Dietrich
Keyword(s):  
Water Quality ◽  
Drinking Water ◽  
Organic Compounds ◽  
Sensory Analysis ◽  
Water Samples ◽  
Distribution System ◽  
Distribution Systems ◽  
Free Chlorine ◽  
Copper Pipe

Pipe relining via in situ epoxy lining is used to remediate corroded plumbing or distribution systems. This investigation examined the effects on odour, TOC, THM formation and disinfectant demand in water exposed to epoxy-lined copper pipes used for home plumbing. The study was conducted in accordance with the Utility Quick Test, a migration/leaching method for utilities to conduct sensory analysis of materials in contact with drinking water. The test was performed using water with no disinfectant and levels of chlorine and monochloramines representative of those found in the distribution system. Panelists repeatedly and consistently described a “plastic/adhesive/putty” odour in the water from the pipes. The odour intensity remained relatively constant for each of two subsequent flushes. Water samples stored in the epoxy-lined pipes showed a significant increase in the leaching of organic compounds (as TOC), and this TOC was demonstrated to react with free chlorine to form trichloromethane. Water stored in the pipes also showed a marked increase in disinfectant demand relative to the water stored in glass control flasks. A study conducted at a full scale installation at an apartment demonstrated that after installation and regular use, the epoxy lining did not yield detectable differences in water quality.

Suspended particles in the drinking water of two distribution systems

2001 ◽  
Vol 1 (4) ◽  
pp. 237-245 ◽  
Author(s):  
V. Gauthier ◽  
B. Barbeau ◽  
R. Millette ◽  
J.-C. Block ◽  
M. Prévost
Keyword(s):  
Water Quality ◽  
Drinking Water ◽  
Organic Matter ◽  
Distribution System ◽  
Distribution Systems ◽  
Suspended Solids ◽  
Suspended Particles ◽  
Raw Water ◽  
Remote Locations ◽  
Dissolved Matter

The concentrations of suspended particles were measured in the drinking water of two distribution systems, and the nature of these particles documented. The concentrations of particulate matter were invariably found to be small (maximum 350 μg/L). They are globally in the very low range in comparison with dissolved matter concentrations, which are measured in several hundreds of mg/L. Except during special water quality events, such as turnover of the raw water resource, results show that organic matter represents the most important fraction of suspended solids (from 40 to 76%) in treated and distributed water. Examination of the nature of the particles made it possible to develop several hypotheses about the type of particles penetrating Montreal's distribution system during the turnover period (algae skeleton, clays). These particles were found to have been transported throughout the distribution systems quite easily, and this could result in the accumulation of deposits if their surface charge were ever even slightly destabilised, or if the particles were to penetrate the laminar flow areas that are fairly typical of remote locations in distribution systems.

Sign in / Sign up

Export Citation Format

Share Document