Scribd
Upload
11 views

Supervised Learning

5th Sem ML (TMSL)

Uploaded by

Aryan Shaw
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
11 views

Supervised Learning

5th Sem ML (TMSL)

Uploaded by

Aryan Shaw
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 9

SUPERVISED

LEARNING
CA1

Name : Aryan Kumar Shaw


Roll No : 13031122036
Subject : Machine Learning
Subject Code : ESC(CSBS) 502
Contents

01 02 03
Introduction Types Key Concepts

04 05 06
Common Algorithms Application Challanges
Introduction
What is Supervised Learning
Supervised learning is a category of machine learning that uses
labeled datasets to train algorithms to predict outcomes and
recognize patterns.
Supervised machine learning algorithms make it easier for
organizations to create complex models that can make accurate
predictions. As a result, they are widely used across various industries
and fields, including healthcare, marketing, financial services, and
more.

How does it works


The data used in supervised learning is labeled — meaning that it
contains examples of both inputs (called features) and correct
outputs (labels). The algorithms analyze a large dataset of these
training pairs to infer what a desired output value would be when
asked to make a prediction on new data.
Types of Supervised Learning

Regression algorithms are used to predict a real or continuous


value, where the algorithm detects a relationship between two
or more variables.
A common example of a regression task might be predicting a
salary based on work experience.
Regression
Key Concepts

SL vs USL 1 2 3
When it comes to
understanding the Labelled Training Prediction
difference between Data Phase Phase
supervised learning vs. The dataset used in During training, the Once trained, the
unsupervised learning, supervised learning model uses the labeled model can be used to
the primary difference consists of input-output data to learn the predict the output for
is the type of input pairs. Each input is relationship between new, unseen inputs.
associated with a correct inputs and outputs. This The accuracy of
data used to train the output, known as the involves adjusting the these predictions
model. Supervised label. For example, in a model's parameters to depends on how well
learning uses labeled dataset used to classify minimize the difference the model has
emails as spam or not between the predicted learned from the
training datasets to try
spam, each email (input) outputs and the actual training data.
and teach a model a is labeled as either outputs in the training
specific, pre-defined "spam" or "not spam" data.
goal. (output).
Common Algorithms
Neural Networks Linear Regression
Neural networks process training data Linear regression is used to identify
by mimicking the interconnectivity of the relationship between a
the human brain through layers of dependent variable and one or more
nodes. Each node is made up of inputs, independent variables and is
weights, a bias (or threshold), and an typically leveraged to make
output. If that output value exceeds a predictions about future outcomes.
given threshold, it “fires” or activates
the node, passing data to the next layer
in the network.

Naive Bayes K-Nearest Neighbor


Naive bayes: Naive Bayes is K-nearest neighbor, also known as the
classification approach that adopts KNN algorithm, is a non-parametric
the principle of class conditional algorithm that classifies data points
independence from the Bayes based on their proximity and
Theorem. This means that the association to other available data. It
presence of one feature does not uses Euclidian distance to calculate
impact the presence of another. the distance b/w the data points.
Application
Image and Object Recognition
1 Supervised machine learning algorithms are often trained to classify objects in
images and videos. For example, an algorithm might be used to recognize a
person in an image and automatically tag them on a social media platform.

Spam Detection
2 Identifying and filtering out unwanted emails.
Email services like Gmail use supervised learning to detect and move spam emails to
the spam folder.

Stock Price Forecasting


3 Predicting future stock prices based on historical data.
Example: Financial institutions using models to make investment decisions.

Speech Recognition
4 Converting spoken language into text
Example: Siri and Alexa voice assistants
Challlanges
Although supervised learning can offer businesses advantages, such as deep data
insights and improved automation, there are some challenges when building
sustainable supervised learning models. The following are some of these challenges:

1 2 3 4

Supervised learning Training supervised Datasets can have a Unlike unsupervised


models can require learning models higher likelihood of learning models,
certain levels of can be very time human error, supervised learning
expertise to intensive. resulting in cannot cluster or
structure algorithms learning classify data on its
accurately. incorrectly. own.
THANK
YOU
References
• https://www.ibm.com/topics/supervised-
learning
• https://cloud.google.com/discover/what-
is-supervised-learning?hl=en

You might also like