Scribd
Upload
29 views1 page

Existing System

The existing system for flood forecasting is complicated and expensive, relying on satellite images, weather data, and mathematical equations. Current machine learning approaches still have drawbacks that can lead to incorrect flood predictions. The proposed system aims to use different machine learning models like logistic regression, support vector machine, K-nearest neighbors, and decision tree classifier on a dataset of past rainfall and flood data from India to more accurately predict floods.

Uploaded by

gopal
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
29 views1 page

Existing System

The existing system for flood forecasting is complicated and expensive, relying on satellite images, weather data, and mathematical equations. Current machine learning approaches still have drawbacks that can lead to incorrect flood predictions. The proposed system aims to use different machine learning models like logistic regression, support vector machine, K-nearest neighbors, and decision tree classifier on a dataset of past rainfall and flood data from India to more accurately predict floods.

Uploaded by

gopal
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 1

EXISTING SYSTEM

Flood forecasting is highly complicated and expensive. The weather and rainfall is a factor of
predicting the flood. The advanced technology uses simulations supported by physics and
differential equations. The satellite images are used to get the rainfall data. In recent times rapid
urbanization, global climate change and extreme rainfall have resulted in flash floods. In orthodox
methods of flood forecasting, using satellite images and radar also involving mathematical
equations, current weather conditions are detected. At present machine learning technologies are
implemented to detect such kinds of natural disasters. The floods are predicted by considering the
parameters causing the flash flood. There are some drawbacks in machine learning that lead to
wrong predictions of floods. The results cannot be accurate in predicting flash floods.

PROPOSED SYSTEM

The aim of this project is to get all the rainfall, previous flood data of India and from a
dataset containing yearly data. By providing real time input to different models of machine learning,
those are Logistic Regression, Support Vector Machine, K-Nearest Neighbors and Decision Tree
Classifier. The input provided to models are pre-processed and patterns are extracted by getting
maximum accuracy. The data provided is split into a Training set and Test set. It is split in the ratio of
7:3. The all four models are used to predict and by comparing all the results of model and
considering the confusion matrix of all the models the accuracy is determined. The best model is
chosen by comparing the accuracy of each model

You might also like