DWDM

You might also like

Download as pptx, pdf, or txt
Download as pptx, pdf, or txt
You are on page 1of 20

DWDM

Anurag Shankar Maurya


211542
Support Vector Machine
• A Support Vector Machine (SVM) is a discriminative
classifier which intakes training data (supervised
learning), the algorithm outputs an optimal hyperplane
which categorizes new examples.
• Supervised learning algorithm for classification.
• Finds a hyperplane that best separates data points of
different classes.
• Maximizes the margin between the hyperplane(decision
boundary) and the closest data points (support vectors)
Model Training

Prediction New Data ?

Output
(Elipse/ Quardilateral
• Teachable Machine with Google
• What could be drawn to classify the black dots from blue squares?
Linearly Separable Data

• A line drawn between these data points classify the black dots and blue squares.
Non Linear Data
• For the previous data the line, if used as a Hyperplane
– Two black dots also fall in category of blue squares
– Data separation is not perfect
– It tolerates some outliers in the classification
Hyperplane
• A hyperplane is a decision boundary that differentiates the
two classes in SVM.
• A data point falling on either side of the hyperplane can be
attributed to different classes.
Margin
• Margin is the perpendicular distance between the closest
data points and the Hyperplane (on both sides)
• The best optimised line (hyperplane) with maximum
margin is termed as Margin Maximal Hyperplane.
• The closest points where the margin distance is
calculated are considered as the support vectors.
Applications of SVM
• Image Classification: Facial Recognition, Medical Image
Analysis
• Text Classification: Spam Filtering, Document
Classification
• Other Applications: Bioinformatics, Fraud Detection, Stock
Market Predection
Lazy Learners

• Lazy learning is a learning


method in which generalization of
the training data is, in theory,
delayed until a query is made to
the system, as opposed to in
eager learning, where the system
tries to generalize the training
data before receiving queries.
• Lazy learners do less work while training data is given
and more work when classification of a test tuple is given.

• A lazy learner simply stores the training data and only


when it sees a test tuple starts generalization to classify
the tuple based on its similarity to the stored training tuple
• Building a model from a given set of training data
• Applying the model to a given set of testing data
• (Learning From Neighbours)
• Simply stores test data and wait until it gets a test tuple
• i.e., works only when it get new example
• Less Training Time
• More Prediction Time
• Lazy Learning
– works only when get new example (test tuple)

• Eager Learning
– eagerly work on every thing (training)

• ex. (a+b)2 = a2 + 2ab + b2


Model Evaluation and Selection
• Selecting the best possible model is crucial for extracting
accurate and actionable insights from your data
warehouse.
• Data warehouses act as a centralized repository for
historical data, providing a goldmine for training and
testing machine learning models, including SVMs and
Lazy Learners.
• To effectively evaluate and select the best model for your
needs, various metrics come into play:
• Accuracy: This is the most common metric, simply
representing the proportion of predictions that were
correct.
• Precision: This metric focuses on the positive
predictions. It measures the ratio of true positives
(correctly identified positive cases) to all predicted
positives (including false positives).
• Recall: In contrast to precision, recall looks at how well
the model identifies all actual positive cases. It's the ratio
of true positives to all actual positives (including false
negatives).
• F1-score: This metric strikes a balance between
precision and recall, providing a harmonic mean of both.
• By considering these metrics alongside the specific goals
of your data mining task within the data warehouse
environment, you can make an informed decision about
which model (SVM, Lazy Learner, or another) is best
suited to extract the insights you seek from your data.

You might also like