Syllabus

- Chapter 0: Introduction to ML
- Chapter 1: K nearest neighbors
- Chapter 2: Association Rule Mining
- Chapter 3: Naive Bayes classifier
- Chapter 4: Decision tree for classification and Regression
- Chapter 5: Random Forest
- Chapter 6: Model Performance
- Chapter 7: Artificial Neural Network
- Chapter 8: Introduction to XAI ( Explainable AI—Machine Learning)
- Chapter 9: Convolutional neural network (CNN)

Introduction to ML
Machine learning is a subset of artificial intelligence (AI) that enables computers to learn from data and make
predictions without being explicitly programmed.

Machine learning teaches computers to recognize patterns and make decisions automatically using data and
algorithms.
It can be broadly categorized into three types:
●​ Supervised Learning: Trains models on labeled data to predict or classify data.Over time, it adjusts itself to
minimize errors and improve accuracy.
●​ Unsupervised Learning: Finds patterns or groups in unlabeled data, like clustering or dimensionality
reduction.
●​ Reinforcement Learning: Learns through trial and error to maximize rewards, ideal for decision-making
tasks.

Supervised Learning
Key word: predictions

Logic:
-​ You take a cleaned dataset/labeled data
-​ You split it into 2 subsets
-​ The first subset will be used to train the model
-​ The second subset will be used to test the model
-​ Data is usually split in the ratio of 80:20
-​ Pick a model
-​ Then, there are 2 phases:
-​ Phase 1: training the model, using the first subset
The algorithm learns to identify patterns and relationships between the input and output data.
-​ Phase 2: testing the model, using the second subset
We evaluate the accuracy of the model
-​ If we are satisfied with the accuracy of our model (%), we deploy it
 Types:
The type of supervised ML depends on the type of target variable.

Target variable categorical numerical/ continuous

supervised ML Classification Regression

2 categories: Binary classification - Linear Regression

2< categories: multinominal classification

⚡Logistic regression is a supervised machine learning algorithm

that accomplishes binary classification tasks by predicting the
probability of an outcome, event, or observation.

⚡Naive Bayes algorithm is a supervised machine learning

algorithm based on applying Bayes’ Theorem with the “naive”
assumption that features are independent of each other given the
class label.

Supervised Machine Learning Algorithms

Algorithm Type Purpose Method Use Cases

Linear Regression Regression Predict continuous output values Linear equation minimizing sum of squares of residuals Predicting continuous values

Logistic Regression Classification Predict binary output variable Logistic function transforming linear relationship Binary classification tasks

Model decisions and their Tree-like structure with decisions and outcomes (decision Classification and Regression
Decision Trees Both
possible outcomes theory) tasks

Random forests again are made up of multiple decision

trees that work together to make predictions. Each tree in
Improve classification and Reducing overfitting, improving
Random Forests Both the forest is trained on a different subset of the input
regression accuracy prediction accuracy
features and data. The final prediction is made by
aggregating the predictions of all the trees in the forest

Predict class or value based on k Finding k closest neighbors and predicting based on Classification and Regression
KNN Both
closest neighbors majority or average tasks, sensitive to noisy data

Predict class based on the feature Text classification, spam filtering,

Naive Bayes Classification Bayes’ theorem with feature independence assumption
independence assumption sentiment analysis, medical
Advantages and disadvantages of supervised learning:

Advantages of Supervised Learning

The power of supervised learning lies in its ability to accurately predict patterns and make data-driven decisions
across a variety of applications. Here are some advantages of supervised learning listed below:
1.​ excels in accurately predicting patterns and making data-driven decisions.
2.​ encompasses tasks such as supervised learning classification and supervised learning regression.
3.​ Applications include complex problems like image recognition and natural language processing.

Disadvantages of Supervised Learning

Despite the benefits of supervised learning methods, there are notable disadvantages of supervised learning:
1.​ Overfitting: Models can overfit training data, leading to poor performance on new data due to capturing
noise in supervised machine learning.
2.​ Feature Engineering : Extracting relevant features is crucial but can be time-consuming and requires
domain expertise in supervised learning applications.
3.​ Bias in Models: Bias in the training data may result in unfair predictions in supervised learning algorithms.
4.​ Dependence on Labeled Data: Supervised learning relies heavily on labeled training data, which can be
costly and time-consuming to obtain.

Unsupervised Learning Algorithms

Types:
There are mainly 3 types of algorithms that are used for unsupervised datasets.
●​ Clustering
○​ Kmeans
○​ Hierarchical
○​ DBSCAN
○​ Spectral
●​ Association Rule Learning
●​ Dimensionality Reduction
○​ PCA
○​ …

Challenges of Unsupervised Learning

Here are the key challenges of unsupervised learning:
●​ Noisy Data: Outliers and noise can distort patterns and reduce the effectiveness of algorithms.
●​ Assumption Dependence: Algorithms often rely on assumptions (e.g., cluster shapes), which may not
match the actual data structure.
●​ Overfitting Risk: Overfitting can occur when models capture noise instead of meaningful patterns in the
data.
●​ Limited Guidance: The absence of labels restricts the ability to guide the algorithm toward specific
outcomes.
 ●​ Cluster Interpretability: Results, such as clusters, may lack clear meaning or alignment with real-world
categories.
●​ Sensitivity to Parameters: Many algorithms require careful tuning of hyperparameters, such as the number
of clusters in k-means.
●​ Lack of Ground Truth: Unsupervised learning lacks labeled data, making it difficult to evaluate the accuracy
of results.

Applications of Unsupervised Learning

Unsupervised learning has diverse applications across industries and domains. Key applications include:
●​ Customer Segmentation: algorithms cluster customers based on purchasing behavior or demographics,
enabling targeted marketing strategies.
●​ Anomaly Detection: Identifies unusual patterns in data, aiding fraud detection, cybersecurity, and
equipment failure prevention.
●​ Recommendation Systems: Suggest products, movies, or music by analyzing user behavior and
preferences.
●​ Image and Text Clustering: Groups similar images or documents for tasks like organization, classification,
or content recommendation.
●​ Social Network Analysis: Detects communities or trends in user interactions on social media platforms.
●​ Astronomy and Climate Science: Classifies galaxies or groups weather patterns to support scientific
research

Deep learning
Deep learning is a type of machine learning that teaches computers to perform tasks by learning from examples,
much like humans do. Imagine teaching a computer to recognize cats: instead of telling it to look for whiskers, ears,
and a tail, you show it thousands of pictures of cats. The computer finds the common patterns all by itself and
learns how to identify a cat. This is the essence of deep learning.
In technical terms, deep learning uses neural networks, which are inspired by the human brain. These networks
consist of layers of interconnected nodes that process information. The more layers, the "deeper" the network,
allowing it to learn more complex features and perform more sophisticated tasks.

Lazy Learning
Lazy learning is a type of machine learning that doesn't process training data until it needs to make a
prediction. Instead of building models during training, lazy learning algorithms wait until they encounter
a new query. This method stores and compares training examples when making predictions. It's also
called instance-based or memory-based learning.

One of the most popular lazy learning algorithms is the k-nearest neighbors (k-NN) algorithm. In k-NN,
the k closest training instances to the query point are considered, and their class labels are used to
determine the class of the query. Lazy learning methods excel in situations where the underlying data
distribution is complex or where the training data is noisy.