The impact of Machine Learning for big data science on the successful rehabilitation of endangered species
Presentation Type
Poster Presentation
In Person or Zoom Presentation
In-Person
Campus
Daytona Beach
Status
Student
Faculty/Staff Department
Honors
Student Year and Major
2, Data Science
Presentation Description/Abstract
Endangered species conservation is vital to protecting Earth’s ecosystems and maintaining biodiversity. The successful conservation of wildlife is reliant on factors like predicting ecosystem reactions to different stimuli, productive allocation and coordination of resources, and observation of regional and global trends, all of which can be accomplished through big data analysis methods with the help of artificial intelligence and machine learning. However, these tools have a significant and unavoidable environmental cost, specifically in terms of carbon emissions. Thus, it is important to determine the scale of positive impact it has on wildlife conservation and biodiversity in terms of carbon emission offset cost and resource cost saved through algorithm use to ensure that the benefits outweigh the costs before continuing the use of this technology.
To accomplish this, the study will focus on the following objectives through the analysis of 10 case study machine learning algorithms: 1) Determine and compare one- time and training, and per- application environmental cost in terms of carbon emission for machine learning algorithms, 2) Determine and compare resource cost saved of machine learning algorithms through manual processing time and labor costs, and 3) Determine positive impact of algorithms by calculating the carbon emission offset cost compared to resource cost saved. Results can be used to determine projected environmental profitability, which will show if algorithms have achieved a net positive environmental impact after reasonable use and allow it to be determined if they should continue to be utilized in endangered species preservation efforts.
Keywords
Machine learning, Artificial Intelligence, Carbon offset, Probability prediction, Observed data classification, Manual processing
The impact of Machine Learning for big data science on the successful rehabilitation of endangered species
Endangered species conservation is vital to protecting Earth’s ecosystems and maintaining biodiversity. The successful conservation of wildlife is reliant on factors like predicting ecosystem reactions to different stimuli, productive allocation and coordination of resources, and observation of regional and global trends, all of which can be accomplished through big data analysis methods with the help of artificial intelligence and machine learning. However, these tools have a significant and unavoidable environmental cost, specifically in terms of carbon emissions. Thus, it is important to determine the scale of positive impact it has on wildlife conservation and biodiversity in terms of carbon emission offset cost and resource cost saved through algorithm use to ensure that the benefits outweigh the costs before continuing the use of this technology.
To accomplish this, the study will focus on the following objectives through the analysis of 10 case study machine learning algorithms: 1) Determine and compare one- time and training, and per- application environmental cost in terms of carbon emission for machine learning algorithms, 2) Determine and compare resource cost saved of machine learning algorithms through manual processing time and labor costs, and 3) Determine positive impact of algorithms by calculating the carbon emission offset cost compared to resource cost saved. Results can be used to determine projected environmental profitability, which will show if algorithms have achieved a net positive environmental impact after reasonable use and allow it to be determined if they should continue to be utilized in endangered species preservation efforts.