A long short term memory based model for green house climate prediction

Abstract

One of the most burning issues of the modern era has become the problem of climate change. This paper will focus on determining the historical trends of average temperature of the world and analyzing the major factors which cause these trends. Since the previous hundred years of history are testified by the data provided by credible sources, such as the concentration of greenhouse gasses, deforestation rates, and fossil power consumption, the research attempts to identify the connectivity between human-made activity and the increase in the global temperatures. The project is able to spot trends and correlation between the variables through both exploratory data analysis and statistical modeling and thus display the importance of anthropogenic aspects in the climate change.
The analysis involves application of different statistical tech- niques, machine learning techniques to gain insights, and predict trends in temperature. Such methods as a correlation analysis, linear regression, and multivariate regression, and time series forecasting models are used to measure how various factors could influence global temperature changes. Another aspect that the project considers is the advanced forecasting tools that may be used to determine the future changes in temperature on the ground of the current trajectories.
Findings of the research give a better idea of the trend that temperature beckons and the degree to which it is affected by, on the one hand, environmental factors and, on the other hand, manmade factors. The results highlight the emphasis of greenhouse gas emission among other factors of global warming. Furthermore, predictive models created in the current study provide insight into prospective future events which would imply the necessity to implement sustainable policies which would prevent the occurrence of adverse effects.
All in all, the given work has the potential to advance the knowledge of climate dynamics by integrating the data-driven analysis with sound modeling tools. It can provide a basis of further work and can work towards the creation of awareness and make information that can be used in decision making when it comes to combating and adapting to climate change.