Professional Documents
Culture Documents
I Assignment-2
221910401022 | Haneeth Babu M.
Linear Regression is used for solving Logistic regression is used for solving
Regression problems. Classification problems.
In linear regression, we find the best fit In Logistic Regression, we find the
line, by which we can easily predict the S-curve by which we can classify the
output. samples.
Least square estimation method is used Maximum likelihood estimation method is
for estimation of accuracy. used for estimation of accuracy.
The output for Linear Regression must be The output of Logistic Regression must
a continuous value, such as price, age, be a Categorical value such as 0 or 1, Yes
etc. or No, etc.
Machine learning models can be categorized Neural Networks can be classified into
under two types – supervised and unsupervised feed-forward, recurrent, convolutional, and
learning models. modular Neural Networks.
Since Machine Learning models are adaptive, The first layer is the input layer, followed by a
they are continually evolving by learning through hidden layer, and then finally an output layer.
new sample data and experiences. Thus, the Each layer contains one or more neurons. By
models can identify the patterns in the data. increasing the number of hidden layers within a
Here, data is the only input layer. However, even Neural Network model, you can increase its
in a simple Neural Network model, there are computational and problem-solving abilities.
multiple layers.
Skills required for Machine Learning include Neural networks demand skills like data
programming, probability and statistics, Big Data modelling, Mathematics, Linear Algebra and
and Hadoop, knowledge of ML frameworks, data Graph Theory, programming, and probability
structures, and algorithms. and statistics.
Machine Learning is applied in areas like Neural Networks, on the other hand, are used
healthcare, retail, e-commerce (recommendation to solve numerous business challenges,
engines), BFSI, self-driving cars, online video including sales forecasting, data validation,
streaming, IoT, and transportation and logistics, customer research, risk management, speech
to name a few. recognition, and character recognition, among
other things.