DETECTION AND PREDECTION OF AIR POLLUTION USING ML MODELS

Abstract

Governments in both developed and developing countries are
fully aware that air quality control is a crucial responsibility that
must be completed. Conditions such as weather and traffic
congestion, fossil fuel burning, and industrial features such as
power plant emissions all have a substantial impact on
environmental contamination and are thus considered to be
environmental polluting factors. In terms of influence on air
quality, particulate matter (PM 2.5) is the most significant o f a l l
the particulate matter that can be measured, and it deserves more
attention than it now receives. Human health may be negatively
affected when there is an excess of ozone in the air, which is
conceivable when the amount of ozone is high in the atmosphere.
No amount of emphasis can be placed on how vital it is to
monitor its concentration in the atmosphere on a regular basis in
order to effectively control it. In this study, logistic regression is
used to determine if a data sample is contaminated or not
polluted, based on the distribution of the data sample data. It is
possible to estimate future levels of PM2.5 using
authoregression, which is a statistical method that is based on
previously gathered data. Being aware of the amount o f PM2 .5
that will be present in the air in the following years, months, o r
weeks allows us to work toward lowering its concentration to
levels lower than those considered to be hazardous. Ba sed on a
data collection that includes daily atmospheric condi t ions in a
certain city, this technique was developed to attempt to anticipate
PM2.5 levels and identify air quality in a given place.