Machine Learning based Decision Support System for Post Mortem Inspection of Pig Health

Abstract

This paper presents a decision support system for post mortem inspection of slaughtered pigs to help pig abattoirs in producing quality pork. Valuable information on pig health can be achieved by means of computer application. A noble method for decision making on post mortem inspection of pigs using different machine learning techniques is presented here. It is important to make an accurate decision for pork consumption from the post mortem finding of the pig to prevent consumption of unhealthy meat. The proposed system collects the comprehensive information regarding the post mortem decisions related to pig from the veterinary experts. Different models of Machine Learning Algorithms are trained in this system to perform a comparative study in terms of different performance measures. It is found that the predictive model with Support Vector Machine(SVM) is the best performing model for making a decision on the post mortem health of a pig for the pig datasets. By using the developed predictive machine learning model, it is able to take a decision on normal, partial condemnation or total condemnation of a post mortem pig with high accuracy.