Top Banner
Lecture 2: Introduction to Machine Learning
20
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript

Lecture 2: Introduction to Machine LearningMachine Learning Defnition Field of study that gives computer the ability to learn without being explicitly programmed !rthur "amuel# $%&'( "tudy of algorithms that improve their performance ) at some tas* + with experience , +om Mitchell# $%%-( .ell defned learning tas*: /)# +# ,0+: )lay chec*ers): 1 of games won,: )laying against self.ell Defned Learning +as*2andwriting 3ecognition+as* +: recogni4ing and classifying handwritten words within images)erformance ): percent of words correctly classifed+raining experience ,: a database of written words with given classifcation5uestion"uppose your email program watches which email you do and do not mar* as spam and based on that learn how to better flter spam6 .hat is the tas* in this setting 7lassifying emails as spam or not spam +he number of emails correctly classifying as spam8not spam Labelling emails as spam8 not spam 9on of above: +his is not a machine learning problemMachine Learning !lgorithms"upervised Learning !lgorithms:nsupervised Learning !lgorithms"upervised Learning3ight answers are given for inputs3egression refers to predicting continuous valued output e6g6 price( "upervised Learning7lassifcation refers to predict discrete valued output e6g6 ; or $("upervised LearningMore sophisticated features are::niformity of cell si4e:niformity of cell shape# etc5uestion"uppose you are running a company and want to develop a learning algorithm to address each of two problems:)roblem $: you have large inventory of identical items6

Aiven email labelled as spam8not spam# learn a spam flter Aiven a set of news articles on the web# group them into set of articles about the same storyAiven a database of customer data# automatically discover mar*et segments and group customer into diBerent mar*et segments Aiven a database of patients diagnosed as either having diabetes or not# learn to classify a new patients as either having a diabetes or not6:ngraded !ssignmentInstall @ctave C an open source software or)ractice with:,lementary operation: add# subtract# multiplication# power# divide# etc7onditional operation: eDual# not eDual# greater# greater and eDual to# etcLogical operations: !9D# @3# E@3# etc Fariable assignmentFectors and matrices: defning vectors and matrices# ones# 4eros# rand# eyedoc and help comand