TOURISM
Question 1) What Is Tourism Product And Services? State Its Types And Forms Of
Tourism.
Answer:
Tourism Products and Services
In relation to tourism, very often you will come across terms like tourism products
and services. Here we attempt to defi

Linear Dimensionality Reduction:
Principal Component Analysis
Piyush Rai
Machine Learning (CS771A)
Sept 2, 2016
Machine Learning (CS771A)
Linear Dimensionality Reduction: Principal Component Analysis
1
Dimensionality Reduction
Machine Learning (CS771A)
Li

Learning as Optimization:
Linear Regression
Piyush Rai
Machine Learning (CS771A)
Aug 10, 2016
Machine Learning (CS771A)
Learning as Optimization: Linear Regression
1
Learning as Optimization
Consider a supervised learning problem with training data cfw_(x

Course Logistics and
Introduction to Machine Learning
Piyush Rai
Machine Learning (CS771A)
July 28, 2016
Machine Learning (CS771A)
Course Logistics and Introduction to Machine Learning
1
Course Logistics
Timing and Venue: WF 6:00-7:30pm, RM 101
Course web

Clustering: K -means and Kernel K -means
Piyush Rai
Machine Learning (CS771A)
Aug 31, 2016
Machine Learning (CS771A)
Clustering: K -means and Kernel K -means
1
Clustering
Usually an unsupervised learning problem
Given: N unlabeled examples cfw_x 1 , . . .

Logistic Regression
Piyush Rai
Machine Learning (CS771A)
Aug 17, 2016
Machine Learning (CS771A)
Logistic Regression
1
Recap
Machine Learning (CS771A)
Logistic Regression
2
Linear Regression: The Optimization View
Define a loss function `(yn , f (x n ) = (

Introduction to Generative Models
Piyush Rai
Machine Learning (CS771A)
Sept 23, 2016
Machine Learning (CS771A)
Introduction to Generative Models
1
Generative Model
Defines a probabilistic way that could have generated the data
Each observation x n is assu

CS771: Machine learning: tools, techniques, applications
Assignment #1: Naive Bayes, NN, LDF
Due on: 31-1-2016, 23.59
MM:225
21-1-2016
You are free to use any open source or free platform or library. Even different ones for different
questions. Avoid pric

6
Dimensionality Reduction
The complexity of any classier or regressor depends on the number
of inputs. This determines both the time and space complexity and
the necessary number of training examples to train such a classier or regressor. In this chapter

Decision trees and forests
I
A simple, powerful, explainable classification model.
I
Works for binary and multi-label classification.
I
Also works for regression.
I
Can be used in progressive ensemble mode through decision
forests.
I
Performs very well in

The Story So Far (Mid-Sem Review)
Piyush Rai
Machine Learning (CS771A)
Sept 9, 2016
Machine Learning (CS771A)
The Story So Far (Mid-Sem Review)
1
Topics Seen So Far
Supervised Learning
Distance based methods
Distance from Means
Nearest Neighbor methods
Le

Learning by Asking Questions:
Decision Trees
Piyush Rai
Machine Learning (CS771A)
Aug 5, 2016
Machine Learning (CS771A)
Learning by Asking Questions: Decision Trees
1
A Classification Problem
Indoor or Outdoor ?
Pic credit: Decision Forests: A Unified Fra

our star and its planets are a tiny part of the Milky Way
galaxy. The Milky Way is a huge city of stars, so big that
even at the speed of light, it would take 100,000 years to
cross it. All the stars in the night sky, including our sun, are
just some of t

The story of Kashmir is not much different from the rest of ancient India and followed the same
pattern. Until the 14th century, various Hindu and Buddhist kingdoms ruled the region. From
about 1350-1580 it came under Islamic rule after attack from the Tu

1. Report:
a. Paper Format: Portrait (Landscape only if required for illustrations)
b. Font: Times New Roman (Arial only for headings if desired)
c. Font Size: 12 for body and 14-16 for headings
d. Line Spacing: 1.5
e. Alignment: Justified
f. Illustration

CERTIFICATE OF MEDICAL FITNESS
(TO BE DEPOSITED A T THE TIME OF JOINING)
To be obtained only from Gazetted Government Medical officer/Medical Officer of a Government
Undertaking. (Please note that in no other form this certificate will be accepted. Medica

l,
DIRECTORATE OF TECHNICAL EDUCATION & INDUSTRIAL
TRAINING, PUNJAB.
PUBLIC NOTICE
It is notified for the Kashmiri Migrants that the State
Government after considering the guidelines issued by All lndia Council for
Technical Education and Ministry of Huma

Nonlinear Learning with Kernels
Piyush Rai
Machine Learning (CS771A)
Aug 26, 2016
Machine Learning (CS771A)
Nonlinear Learning with Kernels
1
Recap
Machine Learning (CS771A)
Nonlinear Learning with Kernels
2
Support Vector Machines
Looked at hard and soft

Learning by Computing Distances:
Distance-based Methods and Nearest Neighbors
Piyush Rai
Machine Learning (CS771A)
Aug 3, 2016
Machine Learning (CS771A)
Learning by Computing Distances: Distance-based Methods and Nearest Neighbors
1
Data and Data Represen

Learning via Probabilistic Modeling of Data
Piyush Rai
Machine Learning (CS771A)
Aug 12, 2016
Machine Learning (CS771A)
Learning via Probabilistic Modeling of Data
1
Quick Recap of Last Lecture
Machine Learning (CS771A)
Learning via Probabilistic Modeling

Generative Models for Clustering, GMM, and
Intro to EM
Piyush Rai
Machine Learning (CS771A)
Sept 26, 2016
Machine Learning (CS771A)
Generative Models for Clustering, GMM, and Intro to EM
1
Generative Models
A probabilistic way to think about the data gene

Online Learning via Stochastic Optimization,
Perceptron, and Intro to SVMs
Piyush Rai
Machine Learning (CS771A)
Aug 20, 2016
Machine Learning (CS771A)
Online Learning via Stochastic Optimization, Perceptron, and Intro to SVMs
1
Stochastic Gradient Descent

CS771A/CS771 - Assignment #0
Pre-requisites: Probability and linear algebra
January 6, 2016
1. This is an assignment to revise your pre-requisites.
2. It will not be graded but you have to submit it. Other assignments will be given in parallel which will

Learning theory
I
What is common to the wide variety of learning algorithms and
methods? (classification, regression, clustering, deep netwoks,
matrix factorization etc.)
I
Is it possible to bound the generalization error of a learning
algorithm? How can

CS771: Machine learning: tools, techniques, applications
Assignment #2: SVM, Kernels, Regression
Due on: 16-3-2016, 23.59
MM: 200
07-3-2016
1. Hinge loss is often used as the loss function for maximum margin classification. It is defined as:
L(y) = max(0,

Kernels
Non-linear mapping to feature space
I
Generally L is not linearly separable.
I
Solution: Map input space : X F by non-linear mapping
to another feature space F(usually of higher dimensional).
Find a linear separator in F. Equivalent to a non-line

CS771: Machine learning: tools, techniques and applications
Mid-semester exam
Time: 2 hours
Max marks: 80
15-Feb-2015
1. Answer all 4 questions. The question paper has 2 pages.
2. Your answers should show all calculations for full credit.
3. Please be pre

Neural networks
I
I
I
Neural networks are non-linear statistical models (known in
statistics as project pursuit regression).
The basic neural network consists of 3 layers of units (usually
called neurons) with interconnections between successive
layers. T