CS464
Introduction to
Machine Learning
K-Nearest Neighbour
Bilkent University
Parametric Methods
Non-parametric Methods
Typically dont make distributional assumptions
Today we will see an example of a non-parametric
model
How would you classify ?
*
Pred
Islamic University of Gaza
Faculty of Engineering
Computer Engineering Department
Information Storage and Retrieval (ECOM 5124)
IR
_
HW 1
Boolean Retrieval
Eng. Mohammed Abdualal
February 14, 2015
Exercise 1.1
Draw the inverted index that would be built f
CS464
Introduction to
Machine Learning
Lecture 11
znur Tatan
Bilkent University
Administrivia
Homework 2 is due Nov 19 Sunday 22:00 pm
Exam on Nov 26 Tuesday 6:00 pm
Everything covered in lectures (including slides and
the board + what we talked) + req
CS464
Introduction to
Machine Learning
Lecture 8
znur Tatan
Bilkent University
Administrivia
Next deadlines on the project:
Progress Presentation Nov 18, 20 2013 in class
Progress Report Nov 21 2013, 5:00 pm
Outline
Estimating Test Error and Model Sel
CS464
Introduction to
Machine Learning
Lecture 10
znur Tatan
Bilkent University
Administrivia
Next deadlines on the project:
Progress Presentation Dec 2, 20 2013 in class
Progress Report Dec 4 2013, 5:00 pm
Decision Trees
Non-linear classifier
Easy to
CS464
Introduction to
Machine Learning
Lecture 7
znur Tatan
Bilkent University
Administrivia
Project Proposal Presentation, Oct 23 Wed, in
class, 5% of the project grade
Upload your ppt, pptx or pdf by Oct 22 10 pm.
Prepare at most 4-to-5 slides.
Proj
CHAPTER 1
GENERATIVE AND DISCRIMINATIVE
CLASSIFIERS:
NAIVE BAYES AND LOGISTIC REGRESSION
Machine Learning
Copyright c 2005, 2010. Tom M. Mitchell. All rights reserved.
*DRAFT OF January 19, 2010*
*PLEASE DO NOT DISTRIBUTE WITHOUT AUTHORS
PERMISSION*
This
CS464
Introduction to
Machine Learning
Lecture 9
znur Tatan
Bilkent University
Administrivia
Next deadlines on the project:
- Progress Report Dec 2 2013, 5:00 pm
Progress Presentation Dec 4, 20 2013 in class
Decision Trees
Non-linear classifier
Easy to
CS464
Introduction to
Machine Learning
Lecture 6
znur Tatan
Bilkent University
Administrivia
Homework is due this Friday 5:00 pm sharp
(Moodle upload will be closed after that).
Detailed Project Information is on Moodle
Project Proposal Presentation, O
CS464
Introduction to
Machine Learning
Nave Bayes (1/3)
Bilkent University
Last Lecture: Density Estimation
2
Outline Today
Nave Bayes Classifier
3
A Bayesian Classifier
Classify the example into the class that is most
probable given the attributes
4
L
CS464
Introduction to
Machine Learning
Nave Bayes (2/3)
Bilkent University
Outline
Continue Nave Bayes Classifier
2
A Bayesian Classifier
This is our classifier the class label that is most probable given
the attributes
3
Can we Reduce the
Number of P
CS 464-1 Term Project Proposal
Project Name: WhatEmotion (Emotion Recognizing from 2D Images)
Group Members: Cansu Demiryrek (21101084), Aye Kbra Ta (21301716), Ozan Zaimolu
(21101335), Berire Gndz (21202089)
Description of the Data: Depending on how the
CS464
Introduction to
Machine Learning
Logistic Regression
Bilkent University
Logistic Regression
Name is somewhat misleading.
It is a technique for classification, not regression.
Linear regression for regression
Outcome variable Y is continuous
Logi
CS464
Introduction to
Machine Learning
Classification Performance Metrics
Bilkent University
Outline
Classification Metrics
Model Selection and Validation
Learning typically involves trying out different
models (algorithms, parameters, feature sets,.etc
%
% Introduction to Matlab
% (adapted from http:/www.stanford.edu/class/cs223b/matlabIntro.html)
%
% Stefan Roth <roth (AT) cs DOT brown DOT edu>, 09/08/2003
%
% Stolen from cs143 for cs129 by
% Patrick Doran <pdoran (AT) cs DOT brown DOT edu>, 01/30/2010
CS464
Introduction to
Machine Learning
Lecture 2
Bilkent University
Administrivia
Please check Moodle for the syllabus and required
readings
Outline
Probability and statistics review
Basic concepts and exercises on the board
Review notes are include
In Class Exercises - Probability and Statistics Review
February 9, 2016
Question 1
In a college classroom of engineering majors, 41 percent of students own a smart phone, 35 percent of
students own a tablet, and 20 percent of students that own a smart pho
CS464
Introduction to
Machine Learning
CS464
Bilkent University
Welcome to CS464!
Logistics
Instructors
znur Tatan
e-mail: [email protected]
Office: EA 429
Office hours: By appointment
Teaching Assistant
Ali Burak nal, Iman Deznabi, Ha
CS464
Introduction to
Machine Learning
Estimation
Bilkent University
Outline
Density Estimation
MLE
MAP
Where do we get these probability estimates?
Density Estimation
How do we learn these probability density
functions?
Density Estimation
A billion
CS464
Introduction to
Machine Learning
Nave Bayes (3/3)
Bilkent University
Gaussian Nave Bayes
What about continuous features?
2
One Dimensional Continuous Feature
Assume Gaussian class conditional densities
3
Continuous Features
Distinguish children
CS464
Introduction to
Machine Learning
Lecture 4
znur Tatan
Bilkent University
Administrivia
Office hours announced included in the syllabus
znur Tatan
Office: EA 429
Office Hour: By appointment
Teaching Assistant
Ahmet cen
e-mail: ahmet. [email protected]
CS464
Introduction to
Machine Learning
Lecture 5
znur Tatan
Bilkent University
Administrivia
Midterm date is set.
Nov 26 Tuesday 18:00 20:00 pm
HW1 is due Oct 11 5:00 pm
Well talk about projects (deadlines and content) at
the end of the second hour
Yo
Lexical Analyzer
Lexical Analyzer reads the source program character by character to produce tokens. Normally a lexical analyzer doesnt return a list of tokens at one shot, it returns a token when the parser asks a token from it.
source program
Lexical A
Concept Learning
Inducing general functions from specific training examples is a main issue of machine learning. Concept Learning: Acquiring the definition of a general category from given sample positive and negative training examples of the category. C
CS464 Introduction to Machine Learning
Fall 2009 Homework 3 Nueral Networks Due Date: November 18, 2009 Q1) a) Design a two-input perceptron that implements the boolean function A B. b) Give the trace of the perceptron learning algorithm for this function
CS464 Introduction to Machine Learning
Fall 2009 Homework 2 Decision Tree Learning Due Date: October 23, 2009
Q1) Give decision trees to represent the following boolean functions: A B A [B C] A XOR B [A B] [C D] Q2) Consider the following set of training
CS464 Introduction to Machine Learning
Fall 2009 Homework 1 Concept Learning Due Date: October 14, 2009
Assume that the following training examples are given: Ex 1 2 3 4 Attrb1 a a b c Attrb2 b b c b Attrb3 b b a b Attrb4 b a a b Attrb5 a a b b TargetAttr