Advanced analytical and machine learning methods for analysis of selection and prediction of mortality in commercial swine
No Thumbnail Available
Authors
Meeting name
Sponsors
Date
Journal Title
Format
Thesis
Subject
Abstract
The age of information and the Internet of Things (IoT) has brought forth many exciting opportunities for farmers and researchers in commercial swine production. The amount of data available across all sectors of the industry is rapidly increasing, which requires innovative methods to store, analyze, and derive insights to positively impact producer economic sustainability. In chapter one of this dissertation, we propose a framework for shifting from reactive to proactive livestock management, which is assisted by technologies in artificial intelligence. Further, in chapter two, a novel method known as generation proxy selection mapping (GPSM) was utilized to identify single nucleotide polymorphisms in a commercial pig population that are undergoing significant changes in allele frequency over short time scales (i.e., four to ten years). In chapter three, we developed an algorithm to identify periods of episodic mortality in commercial wean-to-finish pig cohorts, which reveal sequences of days in which mortality is acutely increased relative to population baseline. Lastly, in chapter four, we evaluated various machine learning models to forecast episodic and sporadic mortality in growing pigs. Results from this work can promote evidence-based, data-driven decision making in commercial pig production in real-time.
Table of Contents
DOI
PubMed ID
Degree
Ph. D.