RDF2Rule PRESENTATION

RDF2Rule
Resource Description Framework
Presented by
Efrah Shakir BITF13E113
Sonia Akhtar BITF13E107
1

Semantic Web
 The Semantic Web is an extension of the current web in which information is
given well-defined meaning, better enabling computers and people to work in
cooperation (Tim Berners-Lee, 2001).
 The study of meaning focused on the relation between signifiers like, words,
phrase, signs and symbol and what they stand for.
2

Semantic web problems
Too Much web information
Around 1,000,000,000 (1×109) resources.
Many different type of resources
• Texts, images, graphics
• Audio, video, multimedia.
• Database, web applications
3

Semantic web problems
Information not Indexable
No common scheme for doing so.
Differing relationship between authors, publisher, info
intermediaries and users
Each community use their own approach
Information not shareable
Difficult to share information about information.
Not common catalog scheme.
4

Main issue
 Metadata
 Information about information
 Structured data about data
5

Second issue
 Language for expressing metadata must be:
 universal (so all can understand)
 –flexible (to incorporate different types)
 –extensible (flexible to custom types)
 –simple (to encourage adoption)
 –modular (so that schemes can be mixed, extended)
6

RDF
 RDF stands for Resource Description Framework
 It is a machine understandable metadata
 RDF is graphical formalism ( + XML syntax + semantics) –for
representing metadata
 for describing the semantics of information in a machine
 accessible way
8

Resource Description Framework(RDF)
 RDF is language for represent the resources:
 A Resource can be any thing.
 Within the context of the web relevance is the given to web
resources, i.e. that any thing that can be located via
URL(Uninform Resource Locater).
 The basic building block is the statements(or triple).
 One of the main application: data integration
9

RD2RDF
 The adaption of the relation model to the web given rise to RDF.
 From Tuples to Triples.
 Any relation data can be represented as triples:
 Row key subject
 Column  property / Relation / literal
 Value  value / object
10
subject valueproperty
statement

RDF2Rules
 Rule Learning Approach
 Learning Rules from RDF Knowledge Bases by Mining Frequent
Predicate Cycles
 First mines frequent predicate cycles (FPCs), a kind of interesting
frequent patterns in knowledge bases, and then generates rules
from the mined FPCs
 It uses the entity type information when generates and evaluates
rules
11

Quality of RDF KB
 enrich the knowledge in an RDF KB,
 information extraction techniques are usually used to extract more entities
and their relations from plain text or semi-structured text.
 expand a KB is to infer new facts from the existing ones by using
inference rules.
 hasChild(A,B) ∧ hasSpouse(A,C) ⇒ hasChild(C,B)
 Entities
 Facts
12

RDF graph
 RDF is a graph-based data model;
 a set of RDF triples constitutes an RDF
graph,
nodes  resources
directed vertices  predicates
13

RDF graph (Nodes)
 three kinds of nodes (resources) in an RDF graph:
 IRI  a global identiﬁer for a resource, such as people, organization and place;
 Literals are basic values including strings, dates and numbers, etc.;
 blank nodes in RDF represent recourses without global identiﬁers.
14

Path and Cycle
 Path
 A path in an RDF KB G = (E,P,T) is a sequence of consecutive entities and predicates
 Cycle
 A cycle in an RDF graph is a special path that starts and ends at the same node.
15

Predicate path and Cycle
 (PREDICATE PATH).
 A predicate path is a sequence of entity variables and predicates .
 (PREDICATE CYCLE).
 A predicate cycle is a special predicate path that starts and ends at the same entity
variable.
16

FPC(Frequent Predicate Cycle)
 the interesting patterns represented by predicate cycles can be used to infer new
facts in KBs
 dbo : spouse(x1,x2) ∧ dbo : children(x2,x3)
⇒ dbo : children(x1,x3)
 dbo : children(x1,x3) ∧ dbo : children(x2,x3)
⇒ dbo : spouse(x1,x2)
17

FPC(Frequent Predicate Cycle)
 The number of its instances that exists in the given RDF KB is called the support
of it. If the support of a predicate path (cycle) is not less than a speciﬁed
threshold, it is called the frequent predicate path (cycle)
 Frequent predicate cycles (FPCs) are patterns that frequently appear in the KB,
 rules generated from FPCs are prone to be reliable.
 so our proposed approach RDF2Rules ﬁrst mines frequent predicate cycles from
RDF KBs, and then generates inference rules from FPCs.
18

RDF INDEXING FOR MINING ALGORITHM
 In-memory indexing structure to support the mining algorithm
instead of using the existing RDF storage systems
 Given a predicate, find all the entity pairs that it connects;
 Given an entity, find all its incident edges and its neighbor entities;
 Given a predicate path, find all of its instances path
19

RDF2 Rule
 RDF2Rules can learn rules very quickly.
 RDF2Rules always gets more rules.
 Quality of Predictions and the running time.
 Remove duplicate rules through FPC.
20

conclusion
 RDF is a simple data model based on graph.
 RDF has a extensible URI-based vocabulary.
 Anyone can make statements about ANY RESOURCE( open world
assumptions).
 Rules are learned by ﬁnding frequent predicate cycles in RDF
graphs.
 Quality of Predictions and the running time.
21

RDF2Rule PRESENTATION

More Related Content

What's hot (20)

Similar to RDF2Rule PRESENTATION (20)

RDF2Rule PRESENTATION

Editor's Notes