Induction and Decision Trees. Artificial Intelligence The design and development of computer systems that exhibit intelligent behavior. What is intelligence?

Slides:



Advertisements
Similar presentations
Data Mining Lecture 9.
Advertisements

The Logic of Intelligence Pei Wang Department of Computer and Information Sciences Temple University.
Huffman code and ID3 Prof. Sin-Min Lee Department of Computer Science.
Decision Tree Approach in Data Mining
Classification: Definition Given a collection of records (training set ) –Each record contains a set of attributes, one of the attributes is the class.
Becerra-Fernandez, et al. -- Knowledge Management 1/e -- © 2004 Prentice Hall Chapter 7 Technologies to Manage Knowledge: Artificial Intelligence.
Classification Techniques: Decision Tree Learning
ID3 Algorithm Abbas Rizvi CS157 B Spring What is the ID3 algorithm? ID3 stands for Iterative Dichotomiser 3 Algorithm used to generate a decision.
About ISoft … What is Decision Tree? Alice Process … Conclusions Outline.
Induction of Decision Trees
PSU CS 370 – Artificial Intelligence Dr. Mohamed Tounsi Artificial Intelligence 1. Introduction Dr. M. Tounsi.
Data Mining with Decision Trees Lutz Hamel Dept. of Computer Science and Statistics University of Rhode Island.
Three kinds of learning
Marakas: Decision Support Systems, 2nd Edition © 2003, Prentice-Hall Chapter Chapter 7: Expert Systems and Artificial Intelligence Decision Support.
1 MACHINE LEARNING TECHNIQUES IN IMAGE PROCESSING By Kaan Tariman M.S. in Computer Science CSCI 8810 Course Project.
Building Knowledge-Driven DSS and Mining Data
Chapter 5 Data mining : A Closer Look.
Artificial Intelligence
Artificial Intelligence
Data Mining Techniques
Data Mining Dr. Chang Liu. What is Data Mining Data mining has been known by many different terms Data mining has been known by many different terms Knowledge.
Data Mining Chun-Hung Chou
1 AI and Agents CS 171/271 (Chapters 1 and 2) Some text and images in these slides were drawn from Russel & Norvig’s published material.
Inductive Logic Programming Includes slides by Luis Tari CS7741L16ILP.
DATA MINING : CLASSIFICATION. Classification : Definition  Classification is a supervised learning.  Uses training sets which has correct answers (class.
Copyright R. Weber Machine Learning, Data Mining ISYS370 Dr. R. Weber.
Notes for Chapter 12 Logic Programming The AI War Basic Concepts of Logic Programming Prolog Review questions.
INTRODUCTION TO MACHINE LEARNING. $1,000,000 Machine Learning  Learn models from data  Three main types of learning :  Supervised learning  Unsupervised.
Introduction to Data Mining Group Members: Karim C. El-Khazen Pascal Suria Lin Gui Philsou Lee Xiaoting Niu.
Decision Trees & the Iterative Dichotomiser 3 (ID3) Algorithm David Ramos CS 157B, Section 1 May 4, 2006.
Introduction to machine learning and data mining 1 iCSC2014, Juan López González, University of Oviedo Introduction to machine learning Juan López González.
1 CO Games Development 2 Week 19 Probability Trees + Decision Trees (Learning Trees) Gareth Bellaby.
1 Knowledge Discovery Transparencies prepared by Ho Tu Bao [JAIST] ITCS 6162.
Decision Trees. Decision trees Decision trees are powerful and popular tools for classification and prediction. The attractiveness of decision trees is.
Data Management and Database Technologies 1 DATA MINING Extracting Knowledge From Data Petr Olmer CERN
How Solvable Is Intelligence? A brief introduction to AI Dr. Richard Fox Department of Computer Science Northern Kentucky University.
1 Machine Learning 1.Where does machine learning fit in computer science? 2.What is machine learning? 3.Where can machine learning be applied? 4.Should.
Decision Trees. MS Algorithms Decision Trees The basic idea –creating a series of splits, also called nodes, in the tree. The algorithm adds a node to.
I Robot.
Multi-Relational Data Mining: An Introduction Joe Paulowskey.
Data Mining In contrast to the traditional (reactive) DSS tools, the data mining premise is proactive. Data mining tools automatically search the data.
Machine Learning Chapter 5. Artificial IntelligenceChapter 52 Learning 1. Rote learning rote( โรท ) n. วิถีทาง, ทางเดิน, วิธีการตามปกติ, (by rote จากความทรงจำ.
CS690L Data Mining: Classification
27-18 września Data Mining dr Iwona Schab. 2 Semester timetable ORGANIZATIONAL ISSUES, INDTRODUCTION TO DATA MINING 1 Sources of data in business,
L6. Learning Systems in Java. Necessity of Learning No Prior Knowledge about all of the situations. Being able to adapt to changes in the environment.
What is Data Mining? process of finding correlations or patterns among dozens of fields in large relational databases process of finding correlations or.
ID3 Algorithm Michael Crawford.
Data Mining and Decision Trees 1.Data Mining and Biological Information 2.Data Mining and Machine Learning Techniques 3.Decision trees and C5 4.Applications.
Decision Trees Binary output – easily extendible to multiple output classes. Takes a set of attributes for a given situation or object and outputs a yes/no.
Artificial Intelligence, Expert Systems, and Neural Networks Group 10 Cameron Kinard Leaundre Zeno Heath Carley Megan Wiedmaier.
DECISION TREE Ge Song. Introduction ■ Decision Tree: is a supervised learning algorithm used for classification or regression. ■ Decision Tree Graph:
An Introduction Student Name: Riaz Ahmad Program: MSIT( ) Subject: Data warehouse & Data Mining.
CSC 8520 Spring Paula Matuszek DecisionTreeFirstDraft Paula Matuszek Spring,
Of An Expert System.  Introduction  What is AI?  Intelligent in Human & Machine? What is Expert System? How are Expert System used? Elements of ES.
Data Mining By Farzana Forhad CS 157B. Agenda Decision Tree and ID3 Rough Set Theory Clustering.
WHAT IS DATA MINING?  The process of automatically extracting useful information from large amounts of data.  Uses traditional data analysis techniques.
DECISION TREES Asher Moody, CS 157B. Overview  Definition  Motivation  Algorithms  ID3  Example  Entropy  Information Gain  Applications  Conclusion.
WHAT IS DATA MINING?  The process of automatically extracting useful information from large amounts of data.  Uses traditional data analysis techniques.
DATA MINING TECHNIQUES (DECISION TREES ) Presented by: Shweta Ghate MIT College OF Engineering.
Network Management Lecture 13. MACHINE LEARNING TECHNIQUES 2 Dr. Atiq Ahmed Université de Balouchistan.
Prepared by: Mahmoud Rafeek Al-Farra
RESEARCH APPROACH.
TA : Mubarakah Otbi, Duaa al Ofi , Huda al Hakami
AI and Agents CS 171/271 (Chapters 1 and 2)
MACHINE LEARNING TECHNIQUES IN IMAGE PROCESSING
MACHINE LEARNING TECHNIQUES IN IMAGE PROCESSING
AI and Machine Learning
A task of induction to find patterns
A task of induction to find patterns
Artificial Intelligence
Presentation transcript:

Induction and Decision Trees

Artificial Intelligence The design and development of computer systems that exhibit intelligent behavior. What is intelligence? Turing test: Developed in 1950 by Alan Turing (pioneer in computer science) Computer and human in one room Human “interrogator” in another room Interrogator asks questions...human OR computer answers If interrogator cannot tell whether the human or the computer is answering, then the computer is “intelligent”

Classification of AI Systems Knowledge Representation Systems Capture existing expert knowledge and use it to consult end-users and provide decision support Main types: Rule-based expert systems, Case-base reasoning systems, Frame-based knowledge systems, Semantic networks Machine Learning Algorithms that use mathematical or logical techniques for finding patterns in data and discovering or creating new knowledge Main types: Artificial neural networks, genetic algorithms, inductive decision trees, Naïve Bayesian algorithms, Clustering and pattern-recognition algorithms Data mining involves primarily a “machine learning” form of AI

Data Mining Textbook definition: Knowledge discovery in databases Using statistical, mathematical, AI, and machine learning techniques to extract useful information and subsequent knowledge from large databases Key point: identifying patterns in large data sets

5 Microsoft SQL Server Data Mining Algorithms Decision Trees Naïve Bayesian Clustering Sequence Clustering Association Rules Neural Network Time Series

Decision Trees for Machine Learning Based on Inductive Logic Three types of logical structures commonly used in AI systems: Deduction Abduction Induction

Deduction Premise (rule): if p then q Fact (axiom, observation): p Conclude: q This is classical logic (Modus Ponens). If the rule is correct, and the fact is correct, then you know that the conclusion will be correct. We are given the rule

Abduction Premise (rule): if p then q Fact (axiom, observation): q Conclude: p This form of reasoning is a logical fallacy called “affirming the consequent” (Post hoc ergo propter hoc). The conclusion may be wrong, but it is a plausible explanation of the fact, given the rule. Useful for diagnostic tasks. We are given the rule

Induction 1. Observe p and q together n. Observe p and q together Conclude: if p then q This is stereotypical thinking…highly error prone. We create the rule

Example – Baby in the kitchen

ID3 Decision Tree Algorithm “Iterative Dichotomizer” Developed by Ross Quinlan (1979) This is the basis for many commercial induction products The goal of this algorithm is to find rules resulting in YES or NO values. (Therefore, the output of generated rules have 2 possible outcomes) ID3 generates a tree, where each path of the tree represents a rule. The leaf node is the THEN part of the rule, and the nodes leading to it are the ANDS of attribute-value combinations in the IF part of the rule.

ID3 Algorithm Starting Point: an empty tree (this tree will eventually represent the final rules created by ID3) a recordset of data elements (e.g. records from a database) a set attributes (fields), each with some finite number of possible values NOTE: one of the attributes is the “decision” field, with a YES or NO value (or some other 2-valued option...GOOD/BAD, HIGH/LOW, WIN/LOSE, etc.) Output: a tree, where each path of the tree represents a rule

ID3 algorithm If all records in your recordset are positive (i.e. have YES values for their decision attribute), create a YES node and stop (end recursion) If all records in your recordset are negative, create a NO node and stop (end recursion) Select the attribute that best discriminates among the records (using an entropy function) Create a tree-node representing that attribute, with n branches, where n is the number of values for the selected attribute Divide the records of the recordset into subsets subrecordset 1, subrecordset 2,..., subrecordset n corresponding with each value of the selected attribute Recursively apply the algorithm to each subrecordset i, with reduced attribute set (don’t include already used attributes further down the path)

Calculating Entropy Entropy = mixture, chaos We want to pick the attribute with the lowest entropy:  ideally, a particular value for the input attribute leads to ALL yes or ALL no in the outcome attribute…or come as close to this as possible An attribute’s entropy = Where n is the total number of possible values for the attribute and x i is the ith value

Baby’s RecordSet of Oven-Touching Experiences

ID3 Applied to Baby-in-the-Kitchen Which attribute to start with? Based on Entropy measure (assuming log base 2), Touch stove entropy = Mom in kitchen entropy = 1.0 To see this, note that: Probability of touching stove leading to ouch is.67, and not leading to ouch is *.33 =.22 Probability of mom being in kitchen leading to ouch is.5 and mom being in kitchen not leading to ouch is also.5.5 *.5 =.25

Applying the Touch Stove Attribute

Recurse …apply the Mom in Kitchen attribute where needed

Resulting decision rules If Touch_Oven = No then BOO_BOO = No If Touch_Oven = Yes and Mom_In_Kitchen = Yes then BOO_BOO = Yes If Touch_Oven = Yes and Mom_In_Kitchen = No then BOO_BOO = No

Now we’ll do this with Microsoft SQL Server