Development of a Model for the Prediction of Lumpy Skin Diseases using Machine Learning Techniques

Citations Over TimeTop 17% of 2023 papers

Abstract

Lumpy skin diseases virus (LSDV) is a dangerous and contagious diseases that are mostly common in Sub-Saharan African, South Eastern Europe, South Asia and as well as Middle East, China. LSDV is transmitted through blood sucking insects which are double stranded DNA virus and belong to the family of Capri poxvirus genus family. The recent study proved and clarified that lumpy skin diseases viruses (LSDV) affected mostly cattle and buffalo in Africa, Asia and Europe with population of 29 966, 8 837 and 2 471 outbreaks respectively, between the years 2005 – 2021. Different machine learning approaches have been adopted for the prediction of lumpy skin diseases. An enhanced model was developed to improve the predictive performance of existing model and also, compared the performance of stacked ensemble of single classifiers with respect to optimized artificial neural network. The implementation was done with python 3.7 on Core i5, 16G RAM Intel hardware. The single classifiers are decision tree (DT), k-nearest neighbor, random forest (RF) and support vector machine (SVM). A feature wiz feature selection technique was adopted on lumpy skin diseases dataset coupled with the parameters tuning of the model before classification. Both stacked ensemble and optimized artificial neural network model outperformed the existing model. Stacked ensemble model gives accuracy, precision, f1-score and recall of 97.69%, 98.44%, 98.93% and 98.68% respectively. The results also showed that optimized artificial neural networks of 200 epochs outperformed stacked ensemble classifiers with accuracy of 98.89% and 98.66% of training and validation respectively. The developed model in a real world would assist in reducing the occurrence of lumpy skin diseases.

Related Papers

• → Variable importance‐weighted Random Forests(2017)130 cited
• → Empirical evaluation of feature selection methods in classification(2010)46 cited
• → Feature Selection Algorithm based on Random Forest applied to Sleep Apnea Detection(2019)25 cited
• RANDOM FORESTS:AN IMPORTANT FEATURE GENES SELECTION METHOD OF TUMOR(2009)
• → Optimizing Heart Disease Classification: Exploring the Impact of Feature Selection and Performance of Machine Learning Algorithms(2023)