You are on page 1of 1

Department of Information Technology

IOE Mini Project Poster Academic Year 2021-22

ITA08 - Rainfall Prediction System


Shravani Kadam, Bhishmesh Chaudhari, Dhanesh Salgaonkar. Guide – Prof. Santosh Tamboli

1. Abstract 3. Problem Statement


In modern times, global warming is affecting the entire Rainfall forecasting is very important because heavy and
world, having a major impact on humankind and irregular rainfall can have many impacts like destruction
accelerating climate change. As a result, the atmosphere of crops and farms, damage of property so a better
forecasting model is essential for an early warning that
and ocean are warming, sea levels are rising, and floods
can minimize risks to life and property and also
and droughts are occurring. One of the major managing the agricultural farms in better way.
consequences of this is uneven rainfall / precipitation. Moreover, one of the major focuses of climate change Rainy Day Page
Precipitation forecasting now-a-days is a tedious task that study is to understand whether is there any extreme
is being considered by most of the major world authorities. changes in the occurrence and frequency of heavy 6. Conclusion
Precipitation is a climatic factor that affects the various rainfall events. Rainfall prediction is a challenging task
human activities on which they depend. Like Agricultural and the results should be accurate. There are many We explored and applied several preprocessing steps
hardware devices for predicting rainfall by using the and implemented various machine learning
production, construction, energy production and tourism.
weather conditions like temperature, humidity, pressure. algorithms. Further we carried out a comparative
This makes rainfall a serious problem and requires better These traditional methods cannot work in an efficient study to understand the overall performance of the
rainfall forecasts. For these various reasons, accurate way so by using machine learning techniques we can algorithms and select the most accurate and effective
forecasting of rainfall is paramount. There are many ways produce accurate results. We can just do it by having the algorithm to implement. Out of the seven algorithms
to predict it, but the one that is chosen for this project is to historical data analysis of rainfall and can predict the implemented, Catboost classifier proved to be the
observe and collect previous year's rainfall data which is rainfall for future seasons. We can apply many most efficient ml model with an accuracy of 91.12%
techniques like classification, regression according to and AuC score of 0.89. We have also developed an
collected over 10 years of rainfall measurements. Next, we
the requirements and also, we can calculate the error user-friendly front end system to accept different
forecast rainfall for the next day. Therefore, this project
between the actual and prediction and also the accuracy. input parameters to observe how the input data
seeks to optimize results and find suitable machine Different techniques produce different accuracies so it is affects the model prediction and how accurately it
learning models for predicting rainfall. We will also important to choose the right algorithm and model it will predict the rainfall for the next day. A well-
compare the machine learning algorithms and methods according to the requirements. Accurate ML forecasting developed front end and interactive dashboard
used in machine learning. Methodologies include logistic model could give early alerts of severe weather to help provides easy and better understanding of the data
regression, support vector classification, random forest, prevent natural disasters and destruction. Hence, there is and the built system. We concluded that Australia's
needs to develop ML algorithms capable in predicting rainfall pattern is irregular and uncertain. Using
catboost, and xgboost. Compare each of these algorithms
rainfall with acceptable level of precision. Feature extraction and data visualization through
individually, taking into account the parameters that apply
PowerBI dashboard we were able to recognize and
to machine learning, such as accuracy.
4. Implementation figure out relationship among the data and identify
the factors which have significant effect on the
rainfall received in the Australian Regions.
2. Introduction/Theory
References
Rainfall forecasting is one of the challenging and uncertain
tasks that has a major impact on human society. Timely [1] Kala, A., & Vaidyanathan, S. G. (2018, July).
Prediction of rainfall using artificial neural network.
and accurate forecasts help you proactively reduce human
2018 International Conference on Inventive Research
and financial losses. This is the cause of natural disasters in Computing Applications (ICIRCA).
such as floods and droughts that people all around the [2] Grace, R. K., & Suganya, B. (2020, March).
world face each year. As global warming progresses, Machine Learning based Rainfall Prediction. 2020
rainfall detection and prediction has become a major issue 6th International Conference on Advanced
in countries where appropriate technology is not available, Computing and Communication Systems (ICACCS).
[3] Ahmed, H. A. Y., & Mohamed, S. W. A. (2021,
and if done correctly, it can serve a variety of purposes
February 26). Rainfall Prediction using Multiple
such as agriculture, health and drinking. The accuracy of
Proposed System Architecture Linear Regressions Model. 2020 International
precipitation forecasts is very important where the Conference on Computer, Control, Electrical, and
economy is heavily dependent on agriculture. Due to the Electronics Engineering (ICCCEEE).
dynamic nature of the atmosphere, statistical methods do
5. Results [4] Basha, C. Z., Bhavana, N., Bhavya, P., & V, S.
not provide excellent accuracy in predicting precipitation. (2020, July). Rainfall Prediction using Machine
The non-linearity of precipitation data makes machine Learning & Deep Learning Techniques. 2020
International Conference on Electronics and
learning and AI better techniques. Predictions help people
Sustainable Communication Systems (ICESC).
take precautions, and predictions must be accurate. The [5] Vasantha, B., Tamilkodi, R., & kiran, L. V. (2019,
purpose of this project is to provide non-experts with easy March). Rainfall pattern prediction using real time
access to the techniques and approaches used in the field global climate parameters through machine learning.
of precipitation prediction, to provide comparative studies 2019 International Conference on Vision Towards
between various machine learning techniques, and to Landing page of the Website Emerging Trends in Communication and Networking
provide early precipitation prediction and to identify the (ViTECoN).
[6] Samad, A., Bhagyanidhi, Gautam, V., Jain, P.,
best machine learning algorithm for you. The objective of
Sangeeta, & Sarkar, K. (2020, October 30). An
this system is to understand & analyze the data collected approach for rainfall prediction using long short term
for rainfall for all the states and derive conclusive memory neural network. 2020 IEEE 5th International
information & statistics regarding the rainfall pattern Conference on Computing Communication and
across the country as well as to design an interactive and Automation (ICCCA).
user-friendly web interface for the system for user’s ease [7] Oswal, N. (2021). Predicting Rainfall using
Machine Learning Techniques. Institute of Electrical
of understanding the results obtained from the input data
and Electronics Engineers (IEEE).
provided. Predictor Page

You might also like