ARTIFICIAL INTELLIGENCE

CHAPTER- 9

Machine Learning
Introduction to Machine Learning (ML)
What is Machine Learning (ML)?
• Machine Learning (ML) is a subset of Artificial Intelligence
(AI) that allows machines to learn from data and make decisions
or predictions without being explicitly programmed.
• Instead of being given specific instructions, an ML model learns
patterns from data and makes predictions based on those
patterns.
How Does ML Work?
•Training: A machine learning model is trained on a dataset that
contains both inputs (features) and the correct outputs (labels).
•Prediction: Once trained, the model can be used to make
predictions on new, unseen data based on what it has learned.
Features of Machine Learning
•Data-Driven:
Machine learning relies heavily on data to learn and make
predictions. The more data provided, the better the model can learn
and perform.
•Model-Based:
ML involves creating models that are used to make predictions.
These models are built using algorithms that analyze and learn from
the data.
•Adaptability:
Machine learning models can adapt to changes in data over time. For
example, a spam email detector can improve as it gets exposed to
more types of spam.
•Automation of Decision Making:
ML enables systems to make decisions or perform tasks
without human intervention. For instance, an ML algorithm
can detect fraudulent activities automatically in a bank
transaction system.
• Generalization:
A good machine learning model can generalize from the training
data to make accurate predictions on new, unseen data.
Needs of Machine Learning
•Handling Large and Complex Data:
Traditional programming struggles with large amounts of complex data. ML
can process and extract valuable insights from big datasets that are too
complex for human analysis.
•Example: Predicting customer behavior based on historical data.
•Automation:
ML automates repetitive tasks, saving time and reducing human errors. For
example, ML models can automatically recommend products to customers
based on their browsing history.
•Real-Time Predictions:
Machine learning can provide predictions or insights in real-time. This is
important in areas like fraud detection, where rapid responses are crucial.
•Improvement Over Time:
•ML models get better as they process more data. Unlike static
systems, ML systems can continuously learn and adapt.
•Data-Driven Decisions:
•Machine learning allows businesses and organizations to make data-
driven decisions, leading to better outcomes. For example, predictive
analytics can help forecast sales trends.
Classification of Machine Learning
Machine learning can be classified into three main types based on how they learn
from the data:
1.Supervised Learning:
1. Definition: In supervised learning, the model is trained on a labeled dataset,
which means that both the input data and the correct output (label) are
provided.
2. Goal: The model learns a mapping from inputs to outputs so that it can
predict the output for new, unseen data.
3. Examples:
1. Classification: Identifying whether an email is spam or not (binary
classification).
2. Regression: Predicting house prices based on features like size, location,
etc.
4. Common Algorithms:
1. Linear Regression
2. Logistic Regression
3. Decision Trees
4. Support Vector Machines (SVM)
2. Unsupervised Learning:
•Definition: In unsupervised learning, the model is given input data without
explicit labels. The goal is to find hidden patterns or structures in the data.
•Goal: Discover relationships or groupings in data without predefined categories.
•Examples:
• Clustering: Grouping customers by purchasing behavior.
• Dimensionality Reduction: Reducing the number of features in a dataset
while maintaining the important information (e.g., PCA).
Common Algorithms: K-Means Clustering
Principal Component Analysis (PCA)
3. Reinforcement Learning:
•Definition: In reinforcement learning, an agent learns by interacting with its
environment and receiving feedback in the form of rewards or penalties.
•Goal: The agent's objective is to maximize the cumulative reward over time by
choosing actions that lead to the best outcomes.
•Examples:
• Game Playing: Teaching a computer to play chess or video games by
rewarding good moves.
• Robotics: Teaching robots to navigate a maze or perform tasks like
picking objects.
•Common Algorithms:
• Q-Learning
• Deep Q-Network (DQN)
• Policy Gradient Methods
Applications of Machine Learning
Machine learning has a wide range of applications in various industries:
1.Healthcare:
1. Disease Prediction: ML models are used to predict diseases based on
patient data (e.g., predicting cancer, diabetes).
2. Medical Image Analysis: ML algorithms analyze medical images like X-
rays or MRIs to detect conditions such as tumors or fractures.
2.Finance:
1. Fraud Detection: ML models detect unusual patterns in transaction data to
identify fraudulent activities.
2. Credit Scoring: ML helps in determining whether a person is eligible for a
loan based on their financial history.
3. Retail:
•Recommendation Systems: ML models recommend products to customers
based on their purchase history (e.g., Amazon’s product recommendations).
•Inventory Management: ML helps predict demand for products to optimize
inventory levels.
4. Transportation:
•Autonomous Vehicles: Self-driving cars use ML to navigate and make
decisions based on sensor data and real-time environment analysis.
•Traffic Prediction: ML algorithms predict traffic patterns to help reduce
congestion and optimize traffic flow.
5. Social Media:
•Sentiment Analysis: ML algorithms analyze text data (like tweets or reviews) to
determine the sentiment (positive, negative, neutral).
•Personalized Content: ML helps tailor content shown to users based on their
preferences and past behavior.
6. Natural Language Processing (NLP):
•Speech Recognition: ML models are used to convert spoken language into text
(e.g., voice assistants like Siri).
•Text Classification: Categorizing emails, tweets, or articles into different topics.
7.Manufacturing:
•Predictive Maintenance: ML predicts when a machine is likely to fail so that it
can be serviced in advance, reducing downtime.
8. Entertainment:
•Movie/TV Show Recommendations: Streaming platforms like Netflix use ML to
recommend shows and movies based on past user preferences.
Summary
•Machine Learning is a powerful tool that enables machines to learn from data
and make decisions without explicit programming.
•It is classified into three main types: Supervised Learning, Unsupervised
Learning, and Reinforcement Learning.
•Machine Learning is widely used across various fields like healthcare, finance,
retail, transportation, and more.
•The need for ML arises from the ability to handle large datasets, automate
decision-making, and improve systems over time.
www.paruluniversity.ac.in