A machine learning approach for predicting bacteria content in drinking water - A case study for finding a suitable machine learning model - Including requirements and a recommended implementation

Aug 3, 2026

Upload by Skyler

Current drinking-water testing for bacterial contamination is slow, taking up to eight days and during this interval many people may be exposed to disease risks. This master’s thesis investigates whether machine learning can forecast bacteria levels, and how such a model can be designed and implemented. The work is conducted with Nocoli and combines a literature review of eight prior case studies with interviews with industry stakeholders to assess applicability and requirements. Random Forest is recommended for an initial trade-off between accuracy and interpretability.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages76

Reading Time3 h 12 min

AccessFree to download

Introduction

Aim & Research Question

Limitations

Theory

Measuring of bacteria levels in drinking water today

Case company

Designing a theoretical machine learning model

Research methods and data collection

Machine Learning

Classification and Regression

A machine learning approach for predicting bacteria content in drinking water A case study for ﬁnding a suitable machine learning model including requirements and a recommended implementation Master’s thesis in Computer science and engineering ERIC JONSSON Department of Computer Science and Engineering CHALMERS UNIVERSITY OF TECHNOLOGY UNIVERSITY OF GOTHENBURG Gothenburg, Sweden 2023 Master’s thesis 2023 A machine learning approach for predicting bacteria content in drinking water A case study for ﬁnding a suitable machine learning model including requirements and a recommended implementation ERIC JONSSON Department of Computer Science and Engineering Chalmers University of Technology University of Gothenburg Gothenburg, Sweden 2023 A machine learning approach for predicting bacteria content in drinking water A case study for ﬁnding a suitable machine learning model including requirements and a recommended implementation ERIC JONSSON © ERIC JONSSON, 2023 . Supervisor: Dana Dannélls, Department of Swedish, multilingualism, language technology Advisor: Jacob Cahn, Nocoli Examiner: Marina Axelson-Fisk, Applied Mathematics and Statistics Master’s Thesis 2023 Department of Computer Science and Engineering Chalmers University of Technology and University of Gothenburg SE-412 96 Gothenburg Telephone +46 31 772 1000 Typeset in LATEX Gothenburg, Sweden 2023 A machine learning approach for predicting bacteria content in drinking water A case study for ﬁnding a suitable machine learning model including requirements and a recommended implementation ERIC JONSSON Department of Computer Science and Engineering Chalmers University of Technology and University of Gothenburg Abstract The current method for ﬁnding whether drinking water contains bacterial contamination is a very slow process and it can take up to eight days before the results are obtained. During this time, a signiﬁcant proportion of the population has potentially obtained diseases from contaminated water. As a mitigating action, this thesis aimed to understand if machine learning could be a promising method for forecasting the bacteria level and how such a model could be designed. The project was performed in association with a case company called Nocoli, which is spun out of Chalmers Ventures and desired an examination of the potential implementation. A literature review including eight diﬀerent case studies of how machine learning was previously applied in the ﬁeld and three semi-structured interviews with industryspeciﬁc stakeholders were conducted. The research methodology originated from the fact that both an overview of the current industry situation as well as machine learning applicability was required. Moreover, by using an extracted theory of machine learning algorithms for diﬀerent objectives, the case studies were evaluated to ﬁnd patterns that could meet the case companys demands. It was found that machine learning is promising and desired in the industry to improve current operations. The Random Forest algorithm was recommended in the initial stage due to its trade-oﬀ between accuracy and interpretability. Data on bacterial content and other factors including weather was intended as the data source. The recommendation included a 3:1:1 split between training-, validation- , and test sets as well as using a recursive feature selection algorithm. Additionally, a combination of error measures was recommended including Mean Squared Error with an out-of-bag supplement to reduce overﬁtting. Furthermore, although no data could be obtained to evaluate the recommended model, it was concluded that machine learning could have a positive impact on today’s approach and contribute to improved water management and safety by enabling reliable forecasts. Keywords: machine learning, forecasting, drinking water quality, contaminated water, drinking water treatment, escherichia coli prediction, HPC method, Random Forest. Ac

FAQs

Why is a faster method needed for detecting bacteria in drinking water?

The current method can take up to eight days to produce results. During that time, people may be exposed to diseases from contaminated water.

What was the thesis aim regarding machine learning?

To evaluate whether machine learning is promising for forecasting bacteria levels and to outline how such a model could be designed for implementation.

Which machine learning algorithm was recommended and why?

Random Forest was recommended in the initial stage for its trade-off between accuracy and interpretability.

Aug 3, 2026

Upload by Skyler