A Comparative Study of Machine Learning and Deep Learning Techniques for Prediction of Co2 Emission in Cars
This work addresses the need for more reliable CO2 emission estimates to inform environmental policies and vehicle regulations, but it appears incremental as it builds on existing AI methods without introducing a fundamentally new approach.
The study tackled the problem of inaccurate CO2 emission predictions for cars by comparing machine learning, deep learning, and ensemble techniques, achieving improved accuracy with specific models outperforming others, though no concrete numbers were provided in the abstract.
The most recent concern of all people on Earth is the increase in the concentration of greenhouse gas in the atmosphere. The concentration of these gases has risen rapidly over the last century and if the trend continues it can cause many adverse climatic changes. There have been ways implemented to curb this by the government by limiting processes that emit a higher amount of CO2, one such greenhouse gas. However, there is mounting evidence that the CO2 numbers supplied by the government do not accurately reflect the performance of automobiles on the road. Our proposal of using artificial intelligence techniques to improve a previously rudimentary process takes a radical tack, but it fits the bill given the situation. To determine which algorithms and models produce the greatest outcomes, we compared them all and explored a novel method of ensembling them. Further, this can be used to foretell the rise in global temperature and to ground crucial policy decisions like the adoption of electric vehicles. To estimate emissions from vehicles, we used machine learning, deep learning, and ensemble learning on a massive dataset.