mining functional dependencies from data

18
Ontology Learning Mining Functional Dependencies from Data Hong Yao and Howard J. Hamilton Presented By Stephen Lynn

Upload: ranit

Post on 23-Feb-2016

56 views

Category:

Documents


0 download

DESCRIPTION

Mining Functional Dependencies from Data. Hong Yao and Howard J. Hamilton Presented By Stephen Lynn. Rule Mining. Algorithmic process that takes data as input and yields rules such as: Association Rules Implications Functional dependencies. Overview. Goals/Objectives - PowerPoint PPT Presentation

TRANSCRIPT

Page 1: Mining Functional Dependencies from Data

Ontology Learning

Mining Functional Dependencies from Data

Hong Yao and Howard J. Hamilton

Presented By Stephen Lynn

Page 2: Mining Functional Dependencies from Data

Ontology Learning

Rule Mining

Algorithmic process that takes data as input and yields rules such as:

Association Rules ImplicationsFunctional dependencies

Page 3: Mining Functional Dependencies from Data

Ontology Learning

Overview Goals/Objectives Implication/Functional Dependencies Base Algorithm 4 Pruning Rules Evaluation Analysis

Page 4: Mining Functional Dependencies from Data

Ontology Learning

Goals and Objectives

Design an efficient rule discovery algorithm for mining functional dependencies from a dataset.

Page 5: Mining Functional Dependencies from Data

Ontology Learning

Implication Describes relationship between one specific

combination of attribute-value pairs.Binary DataPropositional Logic

{milk, eggs} → {bread}

Page 6: Mining Functional Dependencies from Data

Ontology Learning

Functional Dependency Describe relationship between all possible

combinations of attribute-value pairs.Disjoint attributesTrue regardless of how many possible attribute valuesantecedent → consequent

postcode → areacode

Page 7: Mining Functional Dependencies from Data

Ontology Learning

Search Space

Page 8: Mining Functional Dependencies from Data

Ontology Learning

Armstrong’s Axioms

Page 9: Mining Functional Dependencies from Data

Ontology Learning

Equivalent Attributes

Page 10: Mining Functional Dependencies from Data

Ontology Learning

Nontrivial Closure

Page 11: Mining Functional Dependencies from Data

Ontology Learning

Base Algorithm Generate all possible antecedents then test with

possible consequents (1 level at a time)

Page 12: Mining Functional Dependencies from Data

Ontology Learning

Pruning Rules

Page 13: Mining Functional Dependencies from Data

Ontology Learning

FD_Mine

Page 14: Mining Functional Dependencies from Data

Ontology Learning

Experimental Summary 15 Datasets from UCI Machine Learning Repository

(2005)

Page 15: Mining Functional Dependencies from Data

Ontology Learning

Results

Page 16: Mining Functional Dependencies from Data

Ontology Learning

Results

Page 17: Mining Functional Dependencies from Data

Ontology Learning

Runtime

Page 18: Mining Functional Dependencies from Data

Ontology Learning

Analysis Strengths

Nicely drawn proofs Weaknesses

Missing good exampleNice to show results with/without pruning

Future WorkFind multivalued dependenciesFind conditional dependenciesData cleaning