2. About the speakers Peter Mika Senior Research Scientist Head
of Semantic Search group at Yahoo! Research inBarcelona Semantic
Search, Web Object Retrieval, NaturalLanguage Processing Tran Duc
Thanh Assistent Professor at AIFB, Karlsruhe Institute ofTechnology
Head of Semantic Search group Semantic Search, Semantic / Linked
Data Management 3. Agenda Introduction (10 min) Semantic Web data
(50 min) The RDF data model Publishing RDF Crawling and indexing
RDF data Query processing (40 min) Ranking (30 min) Result
presentation (15 min) Semantic Search evaluation (15 min) Questions
(5 min) 4. Why Semantic Search? I. We are at the beginning of
search. (Marissa Mayer) Solved large classes of queries, e.g.
navigational Heavy investment in computational power Remaining
queries are hard, not solvable by brute force, and require a deep
understanding of the world and human cognition Background knowledge
and metadata can help to address poorly solved queries 5. Poorly
solved information needs Many of these queries Ambiguous searches
would not be asked by paris hilton users, who learned over Long
tail queriestime what search technology can and can george bush
(and I mean the beer brewernotArizona) in do. Multimedia search
paris hilton sexy Imprecise or overly precise searches jim hendler
pictures of strong adventures people Precise searches for
descriptions countries in africa 32 year old computer scientist
living in barcelona reliable digital camera under 300 dollars 6.
Example: multiple interpretations 7. Why Semantic Search? II. The
Semantic Web is now a reality Large amounts of data published in
RDF Heterogeneous data of varying quality Users who are not skilled
in writing complex queries (e.g. SPARQL) and may not be experts in
the domain Searching data instead or in addition to searching
documents Direct answers Novel search tasks 8. Example: direct
answers in searchPoints of Facetedinterest inInformationsearch for
Information box withVienna,from the Shopping content from
andAustriaKnowledgeresultslinks to Yahoo! Graph Travel Since Aug,
2010, regular search results are Powered by Bing 9. Novel search
tasks Aggregation of search results e.g. price comparison across
websites Analysis and prediction e.g. world temperature by 2020
Semantic profiling recommendations based on particular interests
Semantic log analysis understanding user behavior in terms of
objects Support for complex tasks e.g. booking a vacation using a
combination of services 10. Contextual (pervasive, ambient) search
Yahoo! Connected TV: Widget engine embedded into the TVYahoo!
IntoNow:recognize audio andshow related content 11. Interactive
search and task completion 12. Document retrieval and data
retrieval Information Retrieval (IR) support the retrieval
ofdocuments (document retrieval) Representation based on
lightweight syntax-centric models Work well for topical search Not
so well for more complex information needs Web scale Database (DB)
and Knowledge-based Systems (KB)deliver more precise answers (data
retrieval) More expressive models Allow for complex queries
Retrieve concrete answers that precisely match queries Not just
matching and filtering, but also joins Limitations in scalability
13. Combination of document and dataretrieval Documents with
metadata Metadata may be embedded inside the document Im looking
for documents that mention countries in Africa. Data retrieval
Structured data, but searchable text fields Im looking for
directors, who have directed movies where the synopsis mentions
dinosaurs. 14. Semantic Search Target (combination of) document and
data retrieval Semantic search is a retrieval paradigm that
Exploits the structure/semantics of the data or explicitbackground
knowledge to understand user intent and themeaning of content
Incorporates the intent of the query and the meaning ofcontent into
the search process (semantic models) Wide range of semantic search
systems Employ different semantic models, possibly at different
steps of the search process and in order to support different tasks
15. Semantic models Semantics is concerned with the meaning of the
resources made available for search Various representations of
meaning Linguistic models: models of relationships among words
Taxonomies, thesauri, dictionaries of entity names Inference along
linguistic relations, e.g. broader/narrower terms Conceptual
models: models of relationships among objects Ontologies capture
entities in the world and theirrelationships Inference along
domain-specific relations We will focus on conceptual models in
this tutorial In particular, the RDF/OWL conceptual model for 16.
Semantic Search a process view Knowledge RepresentationSemantic
Models Keywords Resources FormsQuery NLConstruction Formal language
IR-style matching & ranking DB-style precise
matchingQueryKB-style matching &Processinginferences Query
visualization Document and data Documents Result
presentationPresentation Summarization Implicit feedback Explicit
feedbackQueryIncentivesRefinementDocument Representation 17.
Semantic Search systemsFor data / document retrieval, semantic
searchsystems might combine a range of techniques, rangingfrom
statistics-based IR methods for ranking,database methods for
efficient indexing and queryprocessing, up to complex reasoning
techniques formaking inferences! 18. Example: Information Workbench
Addressing the lifecycle ofinteracting with the Web ofData
Integration of data sources Content generation by the enduser
Search and Exploration Visualization User- Wikipedia
PublishinggeneratedDBpedia, Yago Integrated management
ofEarthquakeheterogeneous data(Data.gov)sources Structured and
unstructuredStructure Published and user-generatedDynamicd Static
and dynamic Open domain 19. Data Sources in the Application Entire
English Wikipedia Data from Linked Open Data DBpedia YAGO Data from
Data.gov (US Government) E.g. live data about earthquakes Many more
20. Semantic Search Hybrid Search: Structured queries combined
withkeywords across structured and unstructured datasources Query
interpretation: Translation of keywords intohybrid queries Keyword
search/query interpretation combinedwith faceted search: iterative
refinement processbased on keywords and operations on facets 21.
Search, Refinement and Navigation
KeywordsQueryTranslationsTermCompletionsVorlesung Knowledge
Discovery - InstitutFacetsAIFB21 22. Result Inspection, Analysis
and Browsing 23. Semantic Web data 24. Data on the Web Data on the
Web is not directly accessible Most web pages are generated from
databases, butformatted for human consumption APIs offer limited
views over data Two solutions Extraction using Information
Extraction (IE) techniques Out of scope for this tutorial Relying
on publishers to expose structured data using standard Semantic Web
formats 25. Information extraction Natural Language Processing
Named entity recognition and disambiguation, sentimentanalysis etc.
Extraction of information about entities Suchanek et al. YAGO: A
Core of Semantic KnowledgeUnifying WordNet and Wikipedia, WWW,
2007. Wu and Weld. Autonomously Semantifying Wikipedia, CIKM2007.
Extraction from HTML tables Cafarella et al. WebTables: Exploring
the Power of Tables onthe Web. VLDB 2008 Wrapper induction
Kushmerick et al. Wrapper Induction for InformationExtractionText
extraction. IJCAI 2007 Filling web forms automatically
(form-filling) Madhavan et al. Googles Deep-Web Crawl. VLDB 2008
26. Semantic Web Sharing data across the Web Standard data model
RDF A number of syntaxes (file formats) RDF/XML, RDFa Powerful,
logic-based languages for schemas OWL, RIF Query languages and
protocols HTTP, SPARQL 27. Resource Description Framework (RDF)
Each resource (thing, entity) is identified by a URI Globally
unique identifiers URLs a subset of URIs Often abbreviated using
namespaces e.g. example:roi = http://example.org/roi RDF represents
knowledge as a set of triples Each triple is a single fact about
the entity (an attribute or arelationship) A set of triples forms
an RDF graphRDF documenttype foaf:Personexample:roinameRoi Blanco
28. Linking resourcesRois homepageFriend-of-a-Friend ontology type
example:roifoaf:Person nameRoi BlancoknowssameAsYahoo!s website
type worksWith#roi2#peter [email protected] 29. Ontologies
Ontologies are the schemas for RDF graphs Define the intended
meaning of certain classes and relationships in a domain e.g. the
FOAF ontology defines the foaf:Person class and relationships such
as foaf:knows Ontologies are published in standard languages such
as OWL Language defines the constructs for modeling and their
meaning e.g. subClassOf, sameAs Tools for editing ontologies such
as Protg, TopBraid Composer Reusing and extending well-known
ontologies helps to interpret unknown data e.g. if X is subClassOf
foaf:Person, and A is of type X, then we know that A is a person
30. Example: schema.org Agreement on a shared set of schemas for
common types of web content Bing, Google, and Yahoo! as initial
supporters Similar in intent to sitemaps.org (2006) Use a single
format to communicate the same information to allthree search
engines Support for microdata schema.org covers areas of interest
to all search engines Business listings (local), creative
works(video), recipes, reviews User defined extensions Each search
engine continues to develop its products 31. Documentation and OWL
ontology 32. Publishing RDF Interlinked RDF documents (Linked Data)
Each document describes a single resource with URIspointing to
related resources Common RDF file formats are RDF/XML and Turtle
Mostly implemented as a wrapper around a database orWeb service
Embedding RDF inside HTML RDFa, microdata SPARQL endpoints Triple
stores are databases for managing RDF data SPARQL is a standard
protocol and query language for accessing triple stores using HTTP
33. Linked Data Grass-roots community effort to (re)publish open
datasets in RDF Centered around the Dbpedia project Directory of
datasets at linkeddata.org, thedatahub.org 34. Metadata in HTML
anno 1995What does this termmean?How is this datarelated? 35.
Microformats (anno 2003) Mark-up for data in HTML pages Reuse
existing HTML elements (class, rel) Microformats exist for a
limited set of objects Persons, organizations, reviews, events
etc., see microformats.org No formal schemas Limited reuse,
extensibility of schemas Unclear which combinations are allowed No
identifiers for entities No interlinking between entities
Joe Friday
+1-919-555-7878
Area Administrator, Assistant
36. RDFa and RDFa Lite (2008-2012) W3C standards for embedding
RDF data in HTML documents A set of new HTML attributes to be used
in head or body A specification of how to extract the data from
these attributes RDFa is just a syntax, you have to choose
avocabulary separately RDFa family RDFa 1.0 Recommendation since
October, 2008 RDFa 1.1 is a small update on RDFa to make it easier
to use RDFa Primer (Working Draft, May 8, 2012) RDFa 1.1 Lite is a
subset of RDFa 1.1 with the most 37. Example: Facebooks Open
GraphProtocol The Like button provides publishers with a way to
promote their content on Facebook and build communities Shows up in
profiles and news feed Site owners can later reach users who have
liked anobject Facebook Graph API allows 3rd party developers
toaccess the data Open Graph Protocol is an RDFa-based format that
allows to describe the object that the user Likes 38. Example:
Facebooks Open Graph Protocol RDF vocabulary to be used in
conjunction with RDFa Simplify the work of developers by
restricting the freedom in RDFa Activities, Businesses, Groups,
Organizations, People, Places, Products and Entertainment Only HTML
accepted The Rock (1996) ... 39. Microdata HTML5 Currently under
standardization at the W3C Originally part of the HTML5 spec, but
now a separatedocument Comparable to RDFa 1.1 Lite Key
extensibility features (such as multiple types)missing HTML5 also
has a number of semantic elements
My name is , such as , itemprop="name">Neil.
My band is called Four Parts Water.I was born on May 10th
2009.
40. Current state of metadata on the Web 31% of webpages, 5% of
domains contain some metadata Analysis of the Bing Crawl (US crawl,
January, 2012) RDFa is most common format By URL: 25% RDFa, 7%
microdata, 9% microformat By eTLD (PLD): 4% RDFa, 0.3% microdata,
5.4% microformat Adoption is stronger among large publishers
Especially for RDFa and microdata See also P. Mika, T. Potter.
Metadata Statistics for a Large WebCorpus, LDOW 2012 H.Mhleisen,
C.Bizer.Web Data Commons - ExtractingStructured Data from Two Large
Web Corpora, LDOW2012 41. Exponential growth in RDFa dataAnother
five-fold increasebetween October 2010 andJanuary, 2012Five-fold
increase betweenMarch, 2009 andOctober, 2010Percentage of URLs with
embedded metadata in various formats 42. Crawling and indexing 43.
Crawling the Semantic Web Linked Data Similar to HTML crawling, but
the the crawler needs toparse RDF/XML (and others) to extract URIs
to becrawled Semantic Sitemap/VOID descriptions RDFa Same as HTML
crawling, but data is extracted aftercrawling Mika et al.
Investigating the Semantic Gap through QueryLog Analysis, ISWC
2010. SPARQL endpoints Endpoints are not linked, need to be
discovered by othermeans Semantic Sitemap/VOID descriptions 44.
Data fusion Ontology matching Widely studied in Semantic Web
research, see e.g. list ofpublications at ontologymatching.org
Unfortunately, not much of it is applicable in a Web context due to
thequality of ontologies Entity resolution Logic-based approaches
in the Semantic Web Studied as record linkage in the database
literature Machine learning based approaches, focusing on
attributes Graph-based approaches, see e.g. the work of Lisa
Getoorare applicable to RDF data Improvements over only attribute
based matching Blending Merging objects that represent the same
real world entity andreconciling information from multiple sources
45. Data quality assessment and curation Heterogeneity, quality of
data is an even larger issue Quality ranges from well-curated data
sets (e.g. Freebase) tomicroformats In the worst of cases, the data
becomes a graph of words Short amounts of text: prone to mistakes
in data entry orextraction Example: mistake in a phone number or
state code Quality assessment and data curation Quality varies from
data created by experts to user-generatedcontent Automated data
validation Against known-good data or using triangulation
Validation against the ontology or using probabilistic models Data
validation by trained professionals or crowdsourcing Sampling data
for evaluation 46. Indexing Search requires matching and ranking
Matching selects a subset of the elements to be scored The goal of
indexing is to speed up matching Retrieval needs to be performed in
milliseconds Without an index, retrieval would require streaming
through the collection The type of index depends on the query model
to support DB-style indexing IR-style indexing 47. IR-style
indexing Index data as text Create virtual documents from data One
virtual document per subgraph, resource or triple typically:
resource Key differences to Text Retrieval RDF data is structured
Minimally, queries on property values are required 48. Horizontal
index structure Two fields (indices): one for terms, one for
properties For each term, store the property on the same position
inthe property index Positions are required even without phrase
queries Query engine needs to support the alignment operator
Dictionary is number of unique terms + number ofproperties
Occurrences is number of tokens * 2 49. Vertical index structure
One field (index) per property Positions are not required But
useful for phrase queries Query engine needs to support fields
Dictionary is number of unique terms Occurrences is number of
tokens Number of fields is a problem for merging, queryperformance
50. Distributed indexing MapReduce is ideal for building inverted
indices Map creates (term, {doc1}) pairs Reduce collects all docs
for the same term:(term, {doc1, doc2} Sub-indices are merged
separately Term-partitioned indices Peter Mika. Distributed
Indexing for Semantic Search, SemSearch 2010. 51. Query Processing
/ Matching 52. Structure Taxonomy of search approaches Query
processing / matching techniques for SemanticSearch Types of
semantic data Formalisms for querying semantic data Approaches
General task: hybrid graph pattern matching Matching keyword query
against text Matching structured query against structured data
Matching keyword query against structured data Matching structured
query against text (a hybrid case) Main tasks, challenges and
opportunities 53. Taxonomy of search approaches (1) The search
problem A collection of resources, called data Information needs
expressed as queries Search is the task of efficiently computing
results from data that are relevant to queries Document data
retrieval vs. structured data retrieval Differences in query and
data representation andmatching Efficiently retrieve structured
data that exactly matchformal information needs expressed as
structuredqueries Effectively rank textual results that match
ambiguous NL /keyword queries to a certain degree (notions
ofrelevance) Semantic search: ranked retrieval of document and 54.
Taxonomy of search approaches (2)Query engines (ofdatabases) Exact
Complete Query Sound Approximate Matching Not complete Not sound
RankedData Best effort Top-k Search engines (stand-alone/database
extensons)Query processing mainly focuses on efficiency of
matchingwhereas ranking deals with degree of matching (relevance)!
55. Query processing for Semantic Search (1) Resources represented
by semantic data ranging from Structured data with well defined
schemas Semi-structured data with incomplete or no schemas Data
that largely comprise text Hybrid / embedded data Information needs
of varying complexity, captured usingdifferent formalisms and
querying paradigms Natural language texts and keywords Form-based
inputs Formal structured queries(Search is end-user oriented
paradigm, requiresnatural, intuitive querying interfaces) Semantic
search: efficiently computing results (queryprocessing) from data
that are relevant to queries(ranking) 56. Query processing for
Semantic Search (2)NL Form- / facet- Structured QueriesKeywords
Questions based Inputs (SPARQL)Ambiquities Query MatchingData RDF
data Semi- OWL ontologies withStructured embedded inStructured
rich, formalRDF datatext (RDFa) RDF datasemanticsAmbiquities:
confidence degree, truth/trust 57. Query processing for Semantic
Search (3)Textual DataStructured query on Keyword query textual
data on textual , e.g. querying data, e.g. Web Search target
Semanticextension for search systems group of differentsearch
users, informationsystems?Unstructuredneeds, and types Structured
of data.StructuredQueryQuery Query processing for on Keyword query
queryon structured search is hybrid semantic structured datadata,
e.g. e.g. standardcombination of techniques!search querying
extensions forinterface fordatabasesdatabases / RDF
storesStructured Data 58. Types of data models (1) Textual
Bag-of-words Represent documents, text in structured data,,
real-world objects (captured as structured data) Lacks structure in
text, e.g. linguistic structure, hyperlinks,
(positionalinformation) Structure in structured data
representationterm (statistics) In combination with combination
Cloud Computing Cloud technologies, promising Computing solutions
for the Technologies management of `big solutions data have
emerged. management Existing industry `big data solutions are able
to industry support complex solutions queries and analytics support
tasks with terabytes of complex data. For example, using a
Greenplum. 59. Types of data models (2) Textual Structured Resource
Description Framework (RDF) Represent real-world objects, services,
applications, .documents Resource attribute values and
relationships betweenresources SchemaPicture creatorPersonBob 60.
Types of data models (3) Textual Structured Hybrid RDF data
embedded in text (RDFa) 61. Types of data models RDFa (1)
The trouble with BobAliceBob is a good friend of mine. We went
to the same university, andalso shared an apartment in Berlin in
2008. The trouble with Bob isthat he takes much better photos than
I do:
Beautiful Sunset by Bob.
adopted from : http://www.w3.org/TR/xhtml-rdfa-primer/ 62. Types
of semantic data RDFa (2)Bob is a good friend of mine. We
contentwent to the same university, andalso shared an apartment in
Berlinin 2008. The trouble with Bob isthat he takes much better
photosthan I do: content adopted from :
http://www.w3.org/TR/xhtml-rdfa-primer/ 63. Types of semantic data
- conclusionSemantic data in general can be conceived as a graph
with text and structured data items as nodes, and edges
representdifferent types of relationships includingexplicit
semantic relationships andvaguely specified ones such as
hyperlinks! 64. Formalisms for querying semantic data (1)Example
information needInformation about a friend of Alice, who sharedan
apartment with her in Berlin and knowssomeone working at KIT.
Unstructured queries Fully-structured queries Hybrid queries:
unstructured + structured 65. Formalisms for querying semantic data
(2)Example information needInformation about a friend of Alice, who
sharedan apartment with her in Berlin and knowssomeone working at
KIT. Unstructured NL Keywordsshared apartment Berlin Alice 66.
Formalisms for querying semantic data (3)Example information
needInformation about a friend of Alice, who sharedan apartment
with her in Berlin and knowssomeone working at KIT. Unstructured
Fully-structured SPARQL: BGP, filter, optional, union, select,
construct, ask, describe PREFIX ns: SELECT ?x WHERE { ?x ns:knows ?
y. ?y ns:name Alice.?x ns:knows ?z. ?z ns: works ?v. ?v ns:name KIT
} 67. Formalisms for querying semantic data(4) Fully-structured
Unstructured Hybrid: content and structure constraints shared
apartment Berlin Alice?x ns:knows ? y. ?y ns:name Alice.?x ns:knows
?z. ?z ns: works ?v.?v ns:name KIT 68. Formalisms for querying
semantic data(5) Fully-structured Unstructured Hybrid: content and
structure constraints shared apartment Berlin Alice?x ns:knows ? y.
?y ns:name Alice.?x ns:knows ?z. ?z ns: works ?v.?v ns:name KIT 69.
Formalisms for querying semantic data - conclusionSemantic search
queries can be conceived as graph patterns with nodes referring
totext and structured data items, and edgesreferring to
relationships between these items! 70. Processing hybrid graph
patterns (1) Example information need Information about a friend of
Alice, who shared an apartment with her in Berlin and knows someone
working at KIT.apartment shared Berlin Alice?x ns:knows ?z. ?z ns:
works ?v. ?v ns:name KIT?y ns:name Alice. ?x ns:knows ? ytrouble
with bobFluidOps34Peter sunset.jpgBob is a good friendBeautifulof
mine. We went to Sunsetthe sameGermany Semantic
AliceSearchuniversity, and alsoshared anapartment in Berlinin 2008.
The troubleGermany2009Bobwith Bob is that heThanhtakes much
betterphotos than I do:KIT 71. Processing hybrid graph patterns (2)
Matching hybrid graph patterns against data 72. Matching keyword
query against text Retrieve documents Inverted list (inverted
index)keyword {, , ...} AND-semantics: top-k joinsharedBerlin Alice
shared Berlin Alice D1 D1D1shared=berlin =aliceshared 73. Matching
structured query against structureddata Retrieve data for triple
patterns Index on tables Multiple redundant indexes to cover
different access patterns Join (conjunction of triples) Blocking,
e.g. linear merge join (required sorted input) Non-blocking, e.g.
symmetric hash-join Materialized join indexesPer1 ns:works ?v ?v
ns:name KIT ?x ns:knows ?y. ?x ns:knows ?z.SP-index PO-index?z ns:
works ?v. ?v ns:name KIT = = = Per1 ns:works Ins1 Ins1 ns:name
KITPer1 ns:works Ins1 Ins1 ns:name KIT 74. Matching keyword query
against structured data Retrieve keyword elements Using inverted
indexkeyword {, ,} Exploration / Join Data indexes for triple
lookup Materialized index (paths up to graphs) Top-k Steiner tree
search, top-k subgraph exploration AliceBobKITAlice Bob KIT =
=Alice ns:knows BobInst1 ns:name KITBob ns:works Inst1 75. Matching
structured query against text Based on offline IE (offline see
Peters slides) Based on online IE, i.e., retrieve is as follows
Derive keywords to retrieve relevant documents On-the-fly
information extraction, i.e., phrase pattern matching X name Y
Retrieve extracted data for structured part Retrieve documents for
derived text patterns, e.g. sequence, windows, reg. exp. ?x
ns:knows ?y. ?x ns:knows ??z ns: works ?v. ?v ns:name K knowsname
KIT 76. Matching structured query against text Index Inverted index
for document retrieval and pattern matching Join index inverted
index for storing materialized joins betweenkeywords Neighborhood
indexes for phrase patterns ?x ns:knows ?y. ?x ns:knows ? ?z ns:
works ?v. ?v ns:name K KIT nameknows name KIT 77. Query processing
main tasks Retrieval Documents , data Query elements, triples,
paths, graphs Inverted index,, but also other (B+ tree)Matching
Index documents, triples, materialized paths Join Data Different
join implementations, efficiency depends on availability of indexes
Non-blocking join good for early result reporting and for
unpredictable Linked Data / data 78. Query processing more tasks
More complex queries: disjunction, aggregation, grouping, an Query
alytics Join order optimization Approximate Approximate the search
spaceMatching Approximate the results (matching, join)
Parallelization Top-k Data Use only some entries in the input
streams to produce k results Multiple sources Federation, routing
On-the-fly mapping, similarity join Hybrid 79. Query processing on
the Web - research challenges and opportunities Large amount
ofsemantic data Optimization, parallelizati Data oninconsistent,
redunda Approximationnt, and low quality Hybrid querying and data
Large amount of data managementembedded in text Federation, routing
Large amount of Online schemasourcesmappings Large amount of links
Similarity joinbetween sources 80. Ranking 81. Structure Problem
definition Types of ambiguities Ranking paradigms Model
construction Content-based Structure-based 82. Ranking problem
definition Query Ambiguities arise when representation is
incomplete / imprecise Matching Ambiguities at the level of
elements (content ambiguity) structure between
elementsData(structure ambiguity)Due to ambiguities in the
representation of theinformation needs and the underlying
resources, theresults cannot be guaranteed to exactly match the
query.Ranking is the problem of determining the degree ofmatching
using some notions of relevance. 83. Content ambiguityapartment
shared Berlin Alice?x ns:knows ?z. ?z ns: works ?v. ?v ns:name
KIT?y ns:name Alice. ?x ns:knows ? ytrouble with bobFluidOps34Peter
sunset.jpgBob is a good friendBeautifulof mine. We went to
Sunsetthe sameGermany Semantic AliceSearchuniversity, and
alsoshared anapartment in Berlinin 2008. The
troubleGermany2009Bobwith Bob is that heThanhtakes much
betterphotos than I do:KIT What is meant by Berlin in the query?
What is meant by KIT in the query? What is meant by Berlin in the
data?What is meant by KIT in the data? A city with the name Berlin?
a person?A research group? a university? a location? 84. Structure
ambiguityapartment shared Berlin Alice?x ns:knows ?z. ?z ns: works
?v. ?v ns:name KIT?y ns:name Alice. ?x ns:knows ? ytrouble with
bobFluidOps34Peter sunset.jpgBob is a good friendBeautifulof mine.
We went to Sunsetthe sameGermany Semantic AliceSearchuniversity,
and alsoshared anapartment in Berlinin 2008. The
troubleGermany2009Bobwith Bob is that heThanhtakes much
betterphotos than I do:KIT What is the connection betweenWhat is
meant by works? Berlin and Alice? Works at? employed? Friend?
Co-worker? 85. Ambiguity Recall: query processing is matching at
the level ofsyntax and semantics Ambiguities arise when data or
query allow for multipleinterpretations, i.e. multiple matches
Syntactic, e.g. works vs. works at Semantic, e.g. works vs. employ
Aboutness, i.e., contain some elements whichrepresent the correct
interpretation Ambiguities arise when matching elements of
differentgranularities Does i contains the interpretation for j,
given some part(s) of i(syntactically/semantically) match j E.g.
Berlin vs. we went to the same university, and also, weshared an
apartment in Berlin in 2008 Strictly speaking, ranking is performed
after syntactic /semantic matching is done! 86. Features: What to
use to deal with ambiguities?What is meant by Berlin? What is
theconnection between Berlin and Alice? Content features
Frequencies of terms: d more likely to be about aquery term k when
d more often, mentions k(probabilistic IR) Co-occurrences: terms K
that often co-occur form acontextual interpretation, i.e., topics
(clusterhypothesis) Structure features Consider relevance at level
of fields Linked-based popularity 87. Ranking paradigms Explicit
relevance model Foundation: probability ranking principle Ranking
results by the posterior probability (odds) ofbeing observed in the
relevant class: P(w|R) varies in different approaches, e.g.,
binaryindependence model, 2-poisson model, relevancemodel P(D|R) P(
D | R)P(w | R) (1 P(w | N )) P(D/N) w Dw DkP( w | R) P( w | q1
,..., qk ) P ( m) P ( w | m) P(qk | m) m M i 1 88. Ranking
paradigms No explicit notion of relevance: similarity between
thequery and the document model Vector space model (cosine
similarity) Language models (KL divergence)Sim(q, d )Cos(( w1,d
,..., wt , d ), ( w1,q ,..., wk , q ))P(t | q )Sim(q, d ) KL(q || d
) P(t | q ) log( t VP(t | d ) 89. Model construction How to obtain
Relevance models? Weights for query / document terms? Language
models for document / queries? 90. Content-based model construction
Document statistics, e.g. An object is more likely Term
frequencyabout Berlin? Document length When it contains a
Collection statistics, e.g.relatively high number of mentions of
the term Inverse document Berlinfrequency When the number of
Background language mentions of this term inmodels the overall
collection is tfrelatively lowwt , d idf|d |tfP(t | d ) (1 ) P(t |
C ) |d | 91. Structure-based model construction Consider structure
of objects during content-based modeling, i.e., to obtain
structured content-based model Content-based model for structured
objects, documents and for general tuplesP(t | d )f P(t | f )f Fd
An object is more likely about Berlin? When one of its (important)
fields contains a relatively high number of mentions of the term
Berlin 92. Structure-based model construction PageRank Link
analysis algorithm Measuring relative importance of nodes Link
counts as a vote of support The PageRank of a node recursively
depends on the number and PageRank of all nodes that link to it
(incoming links) ObjectRank Types and semantics of links vary in
structured data setting Authority transfer schema graph specifies
connection An object about Berlin is more important than one
another? strengths When a relatively large number of objects are
linked to it Recursively compute authority transfer data graph 93.
Taxonomy of ranking approaches Explicitly vs. non-explicitly
relevance-based Content-based ranking Structure-based ranking
Content- and-structure-based ranking 94. Result Presentation 95.
Search interface Input and output functionality helping the user to
formulate complex queries presenting the results in an intelligent
manner Semantic Search brings improvements in Query formulation
Snippet generation Suggesting related entities Adaptive and
interactive presentation Presentation adapts to the kind of query
and results presented Object results can be actionable, e.g. buy
this product Aggregated search Grouping similar items, summarizing
results in various ways Filtering (facets), possibly across
different dimensions Task completion Help the user to fulfill the
task by placing the query in a task context 96. Query formulation
Snap-to-grid: suggest the most likely interpretation of thequery
Given the ontology or a summary of the data While the user is
typing or after issuing the query Example: Freebase suggest,
TrueKnowledge 97. Enhanced results/Rich Snippets Use mark-up from
the webpage to generate search snippets Originally invented at
Yahoo! (SearchMonkey) Google, Yahoo!, Bing, Yandex now consume
schema.org markup Validators available from Google and Bing 98.
Other result presentation tasks Select the most relevant resources
within an RDF document Penin et al. Snippet Generation for Semantic
Web Search Engines, ASWC 2010 For each resource, rank the
properties to be displayed Natural Language Generation (NLG)
Verbalize, explain results 99. Aggregated search: facets 100.
Aggregated search: Sig.ma 101. Related entities Related actors and
movies 102. Adaptive presentation:housing search 103. Evaluation
104. Semantic Search challenge (2010/2011) Two tasks Entity Search
Queries where the user is looking for a single real world object
Pound et al. Ad-hoc Object Retrieval in the Web of Data, WWW2010.
List search (new in 2011) Queries where the user is looking for a
class of objects Billion Triples Challenge 2009 dataset Evaluated
using Amazons Mechanical Turk Halpin et al. Evaluating Ad-Hoc
Object Retrieval, IWEST2010 Blanco et al. Repeatable and Reliable
Search SystemEvaluation using Crowd-Sourcing, SIGIR2011 105.
Evaluation form 106. Catching the bad guys Payment can be rejected
for workers who try to game the system An explanation is commonly
expected, though cheaters rarelycomplain We opted to mix control
questions into the real results Gold-win cases that are known to be
perfect Gold-loose cases that are known to be bad Metrics Worker
and std. dev onReal Avg.Known gold-win and gold-loose resultsKnown
Good TotalTime to Time to completebadNcomplete N MeanN Mean NMean
(sec) badguy20 2.556200 2.73820 2.684240 29.6 goodguy 13 1130
2.03813 3156 95 whoknows 11 211.571232483.5 107. Other evaluations
TREC Entity Track Related Entity Finding Entities related to a
given entity through a particular relationship Retrieval over
documents (ClueWeb 09 collection) Example: (Homepages of) airlines
that fly Boeing 747 Entity List Completion Given some elements of a
list of entities, complete the list Question Answering over Linked
Data Retrieval over specific datasets (Dbpedia andMusicBrainz) Full
natural language questions of different forms Correct results
defined by an equivalent SPARQL query Example: Give me all actors
starring in Batman Begins. 108. Resources 109. Resources Books
Ricardo Baeza-Yates and Berthier Ribeiro-Neto. Modern Information
Retrieval. ACM Press. 2011 Survey papers Thanh Tran, Peter Mika.
Survey of Semantic Search Approaches. Under submission, 2012.
Conferences and workshops ISWC, ESWC, WWW, SIGIR, CIKM, SemTech
Semantic Search workshop series Exploiting Semantic Annotations in
Information Retrieval (ESAIR) Entity-oriented Search (EOS)
workshop
