Transcript Semantic Query Languages (sunuş slaytları) File

BBY 464 Semantic Information Management
(Spring 2016)
Semantic Query Languages
Yaşar Tonta & Orçun Madran
[yasartonta, orcunmadran]@gmail.com
Hacettepe University
Department of Information Management
Semantic Web
From Syntactic to Semantic
Interoperability
Search Engines According to Zadeh
• Insufficient
• Works through 2-values logic
• Can’t make inferences
Source: Zadeh, 2005, 2006)
Conversion to Question Answering
Systems
• World’s knowledge
• Relevance (statistical/semantic)
– q: How old is Vera?
– p: Vera is the same age as Irene
– r: Irene is 65
• Making inferences from perception-based
knowledge
– 2-valued logic and probability is not valid
• The main problem is understanding natural
language
Source: Zadeh, 2005, 2006)
• Query formulation
Problems
– Synonymous words (“garbage theory”)
• Lack of semantics
– “Telekom Inc. Turkey Nebi Fışkın Director”
– “Istanbul-based Mobilfon’s Executive Committee
appointed Nebi Fışkın as CEO
• Lack of context
– In which context user seeks information
– COntext INterchange (COIN)
• Presentation of search results
– Users cannot look at the results beyond the first page.
Source: Warren & Davies, 2007, pp. 179-181
Examples
• Google
• Wolfram Alpha
• Swoogle
From Databases to Data Spaces
• Database -> structured
• Data space -> not so structured
– Its primary function is to simplify integration of
heterogeneous data
– e.g., semi-structured data such as XML documents and
text files
– They will be accessible via the same interface as
structured data organized into tables or key/value pairs
– Secondary function of a data space is to simplify data
integration by providing data mapping and semantic
integration facilities for hosted data collections and
external data resources such as relational databases or
files
Source: http://www.streamscape.com/Technology/dataspaces.html
• Database
– Can be queried with SQL (Structured Query
Language)
• Data space
– NoSQL (Not only SQL) databases that specialize in
semi-structured data
Source: http://www.streamscape.com/Technology/dataspaces.html
Structured Query Language (SQL)
Example: SELECT Statement
SELECT *
FROM PROJECT
WHERE Department
=’Finance’ AND MaxHours >
100;
Copyright © 2004
Chapter 6/11
Subqueries
• Subqueries can be extended to include many
levels
• Example
SELECT DISTINCT Name
FROM EMPLOYEE
WHERE EmployeeNumber IN
(SELECT EmployeeNum
FROM ASSIGNMENT
WHERE HoursWorked > 40
AND ProjectID IN
(SELECT ProjectID
FROM PROJECT
WHERE Department = ‘Accounting’));
Chapter 6/12
Copyright © 2004
SPARQL
• SPARQL Protocol and RDF Query Language
• Semantic query language
• Retrieves and manipulates data in RDF format
Source: https://en.wikipedia.org/wiki/SPARQL
Examples
Returns names and emails of every person in the
FOAF dataset
Source: https://en.wikipedia.org/wiki/SPARQL
“What are all the country capitals in Africa?”
Source: https://en.wikipedia.org/wiki/SPARQL
• Web 2.0 is a critical precursor of Semantic Web
and complements RDF
• Semantic Web will be triggered by Web apps
(blogs, wikis, social networks, photo sharing
services, and so on) that require open data and
atomic data containment (Data Spaces)
• Transition from “programmable web” to
“programmable and query-able web of
databases”
• Representing non-RDF data as RDF data by way of
Ontology mapping
How?
• 1. By storing app triples in an RDF Triple Store
• 2. By converting SPARQL queries to SQL and
reformatting results back into RDF from.
RDF Triple Store Implementation
•
•
•
•
•
Data types
Data dictionary (System tables and indexes)
Virtuoso SQL and SPARQL Fusion
Join Operation Algorithms
Data import and index compaction
RDF Data Management
•
•
•
•
Xquery
Xpath
XSLT
XMLS