1. School of Health, Sport & Bioscience, University of East London, London, UK
2. Computer Science and DT, ACE, University of East London, University Way, London, UK
Email: u1817367@uel.ac.uk (S.L.); s.sharif@uel.ac.uk (S.S.); u2091940@uel.ac.uk (A.Z.); u1725457@uel.ac.uk (M.S.); e.n.gobena@uel.ac.uk(E.G.); s.cutler@uel.ac.uk (S.C.)
*Corresponding author
Manuscript received August 2, 2024; revised August 19, 2024; accepted October 21, 2024; published January 9, 2025
Abstract—Tick-borne diseases are a significant health risk to humans and animals worldwide. It is important to understand the environmental and climatic factors that contribute to tick occurrence rates in order to reduce the proliferation of tick-borne diseases. Using machine learning and spatial indexing techniques, this study covers tick occurrence rates in Europe over the last 20 years to understand the environmental and climatic factors that contribute to
Ixodes ricinus tick abundance. We used biodiversity databases to study land cover categories, climate, vegetation index, and sociological factors. Areas with agriculture and natural vegetation, especially broad-leaved forests, had the strongest tick correlation. Waterways and pastures also showed significant positive correlations, indicating tick habitats. Ticks have moderate associations with urban green spaces, industrial units, and mixed forests suggesting their presence in ecologically disturbed habitats. Geoclimatic factors namely Normalised Difference Vegetation Index and rainfall, showed weak to negative correlations with tick population, indicating that they were less important than previously assumed. Linear Regression, Decision Tree, Random Forest, and Support Vector Machine were compared. We found that feature set and outlier presence significantly affected model performance. After removing outliers, Linear Regression performed best for land use features, with an R² value of 0.81, Normalised Root Mean Square Error (NRMSE) of 1.56, Scatter Index (SI) of 1.56, and Mean Absolute Percentage Error (MAPE) of 1.22. Outlier exclusion improved the model performance results. This research emphasises the importance of specific land uses in predicting the dynamics of tick populations. Our findings lay the groundwork for focused intervention strategies to reduce the spread of tick-borne diseases using an innovative and intelligent approach, while also emphasising the need for further investigation into the complex interactions between environmental factors and tick abundance.
Keywords—tick-borne diseases,
Ixodes ricinus, linear regression, decision tree, random forest, support vector machine
[PDF]
Cite: Samantha Lansdell, Mhd Saeed Sharif, Abin Zorto, Misaki Seto, Edessa Negera, and Sally Cutler, " Machine Learning-Based Techniques for Assessing Critical Factors for European Tick Abundance," International Journal of Computer Theory and Engineering, vol. 17, no. 1, pp. 13-20, 2025.
Copyright © 2025 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).
