Isabelle Augenstein, profile picture
Uploaded byIsabelle Augenstein
PDF, PPTX2,030 views

Information Extraction with Linked Data

The document details techniques for information extraction (IE) using linked data, focusing on methods such as named entity recognition (NER), classification, linking, and relation extraction. It highlights the importance of linguistic features and machine learning approaches in developing effective IE systems. The tutorial aims to demonstrate how these methods can annotate and aggregate data to enhance knowledge bases.

Embed presentation

Download as PDF, PPTX
Information Extraction with Linked Data Isabelle Augenstein Department of Computer Science, University of Sheffield, UK i.augenstein@sheffield.ac.uk 2 September 2015 Information Extraction with Linked Data Tutorial, ESWC Summer School 2015
2Why Information Extraction? Isabelle Augenstein
3Why Information Extraction? Isabelle Augenstein semi-structured information unstructured information
4Why Information Extraction? semi-structured information unstructured information How to link this information to a knowledge base automatically?
5Why Information Extraction? semi-structured information unstructured information How to link this information to a knowledge base automatically? Information Extraction!
6 Information Extraction Isabelle Augenstein The Arctic Monkeys almost exclusively played songs from their new album AM at Summerfest 2014 at Miller Lite Oasis in Milwaukee on 25 June 2014.
7 Information Extraction Isabelle Augenstein The Arctic Monkeys almost exclusively played songs from their new album AM at Summerfest 2014 at Miller Lite Oasis in Milwaukee on 25 June 2014. Named Entity Recognition
8 Information Extraction Isabelle Augenstein The Arctic Monkeys almost exclusively played songs from their new album AM at Summerfest 2014 at Miller Lite Oasis in Milwaukee on 25 June 2014. Named Entity Recognition Named Entity Classification (NEC): Arctic Monkeys: mo:MusicArtist AM: mo:SignalGroup Summerfest 2014: mo:Festival Miller Lite Oases: geo:SpatialThing Milwaukee: geo:SpatialThing
9 Information Extraction Isabelle Augenstein The Arctic Monkeys almost exclusively played songs from their new album AM at Summerfest 2014 at Miller Lite Oasis in Milwaukee on 25 June 2014. Named Entity Recognition Named Entity Classification (NEC): Named Entity Linking (NEL): Arctic Monkeys: mo:MusicArtist Arctic Monkeys: mo:artist/ada7a83 ... AM: mo:SignalGroup AM: mo:release-group/a348ba2f-f8b3 … Summerfest 2014: mo:Festival Summerfest 2014: mo:event/3fc3 … Miller Lite Oases: geo:SpatialThing Miller Lite Oases: mo:place/3f26acf … Milwaukee: geo:SpatialThing Milwaukee: mo:area/4dc3fa97-cf9b- …
10 Named Entities: Definition Named Entities: Proper nouns, which refer to real-life entities Named Entity Recognition: Detecting boundaries of named entities (NEs) Named Entity Classification: Assigning classes to NEs, such as PERSON, LOCATION, ORGANISATION, MISC or fine-grained classes such as SIGNAL GROUP Named Entity Linking / Disambiguation: Linking NEs to concrete entries in knowledge base, example: Milwaukee -> LOC: largest city in the U.S. state of Wisconsin -> LOC: Milwaukee, Oregon, named after the city in Wisconsin -> LOC: Milwaukee County, Wisconsin -> ORG: Milwaukee Tool Corp, a manufacturer of electric power tools -> MISC: early codename for what was to become the Macintosh II -> …
11 Relations Isabelle Augenstein The Arctic Monkeys almost exclusively played songs from their new album AM at Summerfest 2014 at Miller Lite Oasis in Milwaukee on 25 June 2014. Named Entity Recognition Relation Extraction foaf:made gn:parentFeature mo:Festival
12 Relations Isabelle Augenstein The Arctic Monkeys almost exclusively played songs from their new album AM at Summerfest 2014 at Miller Lite Oasis in Milwaukee on 25 June 2014. Named Entity Recognition Relation Extraction Temporal Extraction foaf:made gn:parentFeature mo:Festival 2014-06-25
13 Relations Isabelle Augenstein The Arctic Monkeys almost exclusively played songs from their new album AM at Summerfest 2014 at Miller Lite Oasis in Milwaukee on 25 June 2014. Named Entity Recognition Relation Extraction Temporal Extraction foaf:made gn:parentFeature mo:Festival 2014-06-25 Event Extraction Event: mo:Festival: Summerfest 2014 foaf:Agent: Arctic Monkeys time:TemporalEntity: 2014-06-25 geo:SpatialThing: Miller Lite Oasis
14 Relations, Time Expressions and Events: Definition Relations: Two or more entities which relate to one another in real life Relation Extraction: Detecting relations between entities and assigning relation types to them, such as LOCATED-IN Temporal Extraction: Recognising and normalising time expressions: times (e.g. “3 in the afternoon”), dates (“tomorrow”), durations (“since yesterday”), and sets (e.g. “twice a month”) Events: Real-life events that happened at some point in space and time, e.g. music festival, album release Event Extraction: Extracting events consisting of the name and type of event, agent, time and location
15 Summary: Introduction •  Information extraction (IE) methods such as named entity recognition (NER), named entity classification (NEC), named entity linking, relation extraction (RE), temporal extraction, and event extraction can help to add markup to Web pages •  Information extraction approaches can serve two purposes: •  Annotating every single mention of an entity, relation or event, e.g. to add markup to Web pages •  Aggregating those mentions to populate knowledge bases, e.g. based on confidence values and majority voting Milwaukee LOC 0.9 Milwaukee LOC 0.8 Milwaukee ORG 0.4 à Milwaukee LOC Isabelle Augenstein
16 NERC: Methods •  Possible methodologies •  Rule-based approaches: write manual extraction rules •  Machine learning based approaches •  Supervised learning: manually annotate text, train machine learning model •  Unsupervised learning: extract language patterns, cluster similar ones •  Semi-supervised learning: start with a small number of language patterns, iteratively learn more (bootstrapping) •  Gazetteer-based method: use existing list of named entities •  Combination of the above Isabelle Augenstein
17 NERC: Methods Developing a NERC involves programming based around APIs.. Isabelle Augenstein
18 NERC: Methods Developing a NERC involves programming based around APIs.. which can be frustrating at times Isabelle Augenstein
19 NERC: Methods and (at least basic) knowledge about linguistics Isabelle Augenstein
20 Background: Linguistics Sentences Tokens Morphemes Words Sentences Meaning Isabelle Augenstein Lexicon Morphology The farmer hit the donkey. Syntax Lexicon Semantics, Discourse The, farmer, hit, the, donkey, . wait + ed -> waited, cat -> cat wait -> V, cat -> N The (D) farmer (N) hit (V) the (D) donkey (N). NP Every farmer who owns a donkey beats it. ∀x∀y (farmer(x) ∧ donkey(y) ∧ own(x, y) → beat(x, y)) NP
21 Background: NLP Tasks Sentences Tokens Morphemes Words Sentences Meaning Isabelle Augenstein Sentence splitting Tokenisation Lexicon Morphology Syntax Lemmatisation or stemming, part of speech (POS) tagging Chunking, parsing Lexicon Semantics, Discourse Semantic and discourse analysis, anaphora resolution
22 Background: Linguistics Sentences Tokens Morphemes Words Sentences Meaning Isabelle Augenstein New York-basedLexicon Morphology Syntax Lexicon Semantics, Discourse
23 Background: Linguistics Sentences Tokens Morphemes Words Sentences Meaning Isabelle Augenstein New York-based [New, York-based] or [New, York, -, based] Lexicon Morphology Syntax Lexicon Semantics, Discourse
24 Background: Linguistics Sentences Tokens Morphemes Words Sentences Meaning Isabelle Augenstein Lexicon Morphology Syntax Lexicon She’d Semantics, Discourse New York-based [New, York-based] or [New, York, -, based]
25 Background: Linguistics Sentences Tokens Morphemes Words Sentences Meaning Isabelle Augenstein Lexicon Morphology Syntax Lexicon She’d -> she would, she had Semantics, Discourse New York-based [New, York-based] or [New, York, -, based]
26 Background: Linguistics Sentences Tokens Morphemes Words Sentences Meaning Isabelle Augenstein Lexicon Morphology Syntax Lexicon She’d -> she would, she had Semantics, Discourse New York-based [New, York-based] or [New, York, -, based] Time flies like an arrow
27 Background: Linguistics Sentences Tokens Morphemes Words Sentences Meaning Isabelle Augenstein Lexicon Morphology Syntax Lexicon She’d -> she would, she had Semantics, Discourse New York-based [New, York-based] or [New, York, -, based] Time flies(V/N) like(V/P) an arrow
28 Background: Linguistics Sentences Tokens Morphemes Words Sentences Meaning Isabelle Augenstein Lexicon Morphology Syntax Lexicon She’d -> she would, she had Semantics, Discourse New York-based [New, York-based] or [New, York, -, based] Time flies(V/N) like(V/P) an arrow The woman saw the man with the binoculars.
29 Background: Linguistics Sentences Tokens Morphemes Words Sentences Meaning Isabelle Augenstein Lexicon Morphology Syntax Lexicon She’d -> she would, she had Semantics, Discourse New York-based [New, York-based] or [New, York, -, based] Time flies(V/N) like(V/P) an arrow The woman saw the man with the binoculars. -> Who had the binoculars?
30 Background: Linguistics Sentences Tokens Morphemes Words Sentences Meaning Isabelle Augenstein Lexicon Morphology Syntax Lexicon She’d -> she would, she had Semantics, Discourse New York-based [New, York-based] or [New, York, -, based] Time flies(V/N) like(V/P) an arrow The woman saw the man with the binoculars. -> Who had the binoculars? Somewhere in Britain, some woman has a child every thirty seconds.
31 Background: Linguistics Sentences Tokens Morphemes Words Sentences Meaning Isabelle Augenstein Lexicon Morphology Syntax Lexicon She’d -> she would, she had Semantics, Discourse New York-based [New, York-based] or [New, York, -, based] Time flies(V/N) like(V/P) an arrow The woman saw the man with the binoculars. -> Who had the binoculars? Somewhere in Britain, some woman has a child every thirty seconds. -> Same woman or different women?
32 Background: Linguistics Sentences Tokens Morphemes Words Sentences Meaning Isabelle Augenstein Lexicon Morphology Syntax Lexicon She’d -> she would, she had Semantics, Discourse New York-based [New, York-based] or [New, York, -, based] Time flies(V/N) like(V/P) an arrow The woman saw the man with the binoculars. -> Who had the binoculars? Somewhere in Britain, some woman has a child every thirty seconds. -> Same woman or different women? Ambiguities on every level
33 Background: Linguistics Sentences Tokens Morphemes Words Sentences Meaning Isabelle Augenstein Lexicon Morphology Syntax Lexicon She’d -> she would, she had Semantics, Discourse New York-based [New, York-based] or [New, York, -, based] Time flies(V/N) like(V/P) an arrow The woman saw the man with the binoculars. -> Who had the binoculars? Somewhere in Britain, some woman has a child every thirty seconds. -> Same woman or different women? Ambiguities on every level Y U SO AMBIGUOUS?
34 Information Extraction Language is ambiguous.. Can we still build named entity extractors that extract all entities from unseen text correctly? Isabelle Augenstein
35 Information Extraction Language is ambiguous.. Can we still build named entity extractors that extract all entities from unseen text correctly? Isabelle Augenstein
36 Information Extraction Language is ambiguous.. Can we still build named entity extractors that extract all entities from unseen text correctly? However, we can try to extract most of them correctly using linguistic cues and background knowledge! Isabelle Augenstein
37 NERC: Features What can help to recognise and/or classify named entities? •  Words: •  Words in window before and after mention •  Sequences •  Bags of words Summerfest 2014 took place at Miller Lite Oasis in Milwaukee on 25 June 2014. w: Milwaukee w-1: in w-2: Oasis w+1: on w+2: 25 seq[-]: Oasis in seq[+]: on 25 bow: Milwaukee bow[-]: in bow[-]: Oasis bow[+]: on bow[+]: 25 Isabelle Augenstein
38 NERC: Features What can help to recognise and/or classify named entities? •  Morphology: •  Capitalisation: is upper case (China), all upper case (IBM), mixed case (eBay) •  Symbols: contains $, £, €, roman symbols (IV), .. •  Contains period (google.com), apostrophe (Mandy’s), hyphen (speed-o- meter), ampersand (Fisher & Sons) •  Stem or Lemma (cats -> cat), prefix (disadvantages -> dis), suffix (cats -> s), interfix (speed-o-meter -> o) Isabelle Augenstein
39 NERC: Features What can help to recognise and/or classify named entities? •  POS (part of speech) tags •  Most named entities are nouns •  Prokofyev (2014) Isabelle Augenstein 4.1 Part-of-Speech Tags Part-Of-Speech (POS) tags have often been considered as an important discriminative feature for term identification. Many works on key term identification apply either fixed or regular expression POS tag patterns to improve their ef- fectiveness. Nonetheless, POS tags alone cannot produce high-quality results. As can be seen from the overall POS tag distribution graph extracted from one of our collections (see Figure 3), many of the most frequent tag patterns (e.g., JJ NN tagging adjectives and nouns6 ) are far from yielding perfect results. Figure 3: Top 6 most frequent part-of-speech tag comma, while the n-gram “au tion”, which is indeed a valid the beginning or at the end The contingency tables giv trate this: The +punctuation respectively, the counts of the punctuation mark in any of of the n-grams that have no occurrences. From the tables of punctuation marks (+pun an n-gram occurs twice as o valid entities compared to th that the absence of punctuat pens less frequently for the va ones. Table 1: Contingency ta appearing immediately be Va +punctuation 16 punctuation 65 Totals 81 Table 2: Contingency ta appearing immediately af Va
40Morphology: Penn Treebank POS tags
41Morphology: Penn Treebank POS tags Nouns (all start with N) Verbs (all start with V) Adjectives (all start with J)
42 NERC: Features What can help to recognise and/or classify named entities? •  POS (part of speech) tags •  Most named entities are nouns •  Prokofyev (2014) Isabelle Augenstein 4.1 Part-of-Speech Tags Part-Of-Speech (POS) tags have often been considered as an important discriminative feature for term identification. Many works on key term identification apply either fixed or regular expression POS tag patterns to improve their ef- fectiveness. Nonetheless, POS tags alone cannot produce high-quality results. As can be seen from the overall POS tag distribution graph extracted from one of our collections (see Figure 3), many of the most frequent tag patterns (e.g., JJ NN tagging adjectives and nouns6 ) are far from yielding perfect results. Figure 3: Top 6 most frequent part-of-speech tag comma, while the n-gram “au tion”, which is indeed a valid the beginning or at the end The contingency tables giv trate this: The +punctuation respectively, the counts of the punctuation mark in any of of the n-grams that have no occurrences. From the tables of punctuation marks (+pun an n-gram occurs twice as o valid entities compared to th that the absence of punctuat pens less frequently for the va ones. Table 1: Contingency ta appearing immediately be Va +punctuation 16 punctuation 65 Totals 81 Table 2: Contingency ta appearing immediately af Va
43 NERC: Features What can help to recognise and/or classify named entities? •  Gazetteers •  Retrieved from HTML lists or tables [1] •  Using regular expressions patterns and search engines (e.g. “Popular artists such as * ”) •  Retrieved from knowledge bases [1] https://en.wikipedia.org/wiki/Billboard_200 Isabelle Augenstein
44 NERC: Training Models Extensive choice of machine learning algorithms for training NERCs
45 NERC: Training Models •  Unfortunately, there isn’t enough time to explain machine learning algorithms in detail •  CRFs (conditional random fields) are one of the most widely used algorithms for NERC •  Graphical models, view NERC as a sequence labelling task •  Named entities consist of a beginning token (B), inside tokens (I), and outside tokens (O) took(O) place(O) at(O) Miller(B-LOC) Lite(I-LOC) Oasis(I-LOC) in(O) •  For now, we will rule- and gazetteer-based NERC •  It is fairly easy to write manual extraction rules for NEs, can achieve a high performance when combined with gazetteers •  This can be done with the GATE software (general architecture for text engineering) and Jape rules -> Hands-on session Isabelle Augenstein
46 NLP & ML Software Natural Language Processing: -  GATE (general purpose architecture, includes other NLP and ML software as plugins) -  Stanford NLP (Java) -  OpenNLP (Java) -  NLTK (Python) Machine Learning: -  scikit-learn (Python, rich documentation, highly recommended!) -  Mallet (Java) -  WEKA (Java) -  Alchemy (graphical models, Java) -  FACTORIE, wolfe (graphical models, Scala) -  CRFSuite (efficient implementation of CRFs, Python) Isabelle Augenstein
47 NLP & ML Software Ready to use NERC software: -  ANNIE (rule-based, part of GATE) -  Wikifier (based on Wikipedia) -  FIGER (based on Wikipedia, fine-grained Freebase NE classes) Almost ready to use NERC software: -  CRFSuite (already includes Python implementation for feature extraction, you just need to feed it with training data, which you can also download) Ready to use RE software: -  ReVerb (Open IE, extracts patterns for any kind of relation) -  MultiR (Distant supervision, relation extractor trained on Freebase) Web Content Extraction software: -  Boilerpipe (extract main text content from Web pages) -  Jsoup (traverse elements of Web pages individually, also allows to extract text) Isabelle Augenstein
48 Application: Opinion Mining •  Extracting opinions or sentiments in text •  It’s about finding out what people think University of Sheffield, NLP It's about finding out what people think...
49 Application: Opinion Mining •  Opinion Mining is big business •  Someone just bought an album by a music artist •  Writes a review about it •  Someone else wants to buy an album •  Looks up reviews by fans and music critics •  Music artist and music producer •  Get feedback from fans •  Improve their product •  Improve their marketing strategy
50 Application: Opinion Mining •  “Miley Cyrus's attempts to shock would be more effective if she had songs to back up the posturing.” – The Guardian •  “Bangerz is an Amazing album with great lyrics and we can see the Miley Cyrus' musical evolution. Would love to buy it and I already did. ALBUM OF THE YEAR. Peace” – Rodolfoalmeida3
51 Application: Opinion Mining Why is opinion mining and sentiment analysis challenging? •  Relatively easy to find sentiment words in sentences, difficult to identify which topic they are about
52 Application: Opinion Mining Why is opinion mining and sentiment analysis challenging? •  Relatively easy to find sentiment words in sentences, difficult to identify which topic they are about •  “The album comes with a free bonus CD but I don't like the cover art much.” Does this refer to the cover art of the bonus CD or the album?
53 Application: Opinion Mining Why is opinion mining and sentiment analysis challenging? •  Relatively easy to find sentiment words in sentences, difficult to identify which topic they are about
54 Application: Opinion Mining Why is opinion mining and sentiment analysis challenging? •  Relatively easy to find sentiment words in sentences, difficult to identify which topic they are about •  Whitney Houston was quite unpopular… University of Sheffield, NLP Whitney Houston wasn't very popular...
55 Application: Opinion Mining Why is opinion mining and sentiment analysis challenging? •  Relatively easy to find sentiment words in sentences, difficult to identify which topic they are about •  Whitney Houston was quite unpopular… or was she? •  Death confuses opinion mining tools University of Sheffield, NLP Or was she?
56 Application: Opinion Mining Why is opinion mining and sentiment analysis challenging? •  It’s not just about finding sentiment words, context is important too •  “It's a great movie if you have the taste and sensibilities of a 5- year-old boy.” •  “It's terrible Candidate X did so well in the debate last night.” •  “I'd have liked the track a lot more if it had been a bit shorter.” •  If sentiment words are neutral, negative or positive depends on domain •  “a long track” vs “a long walk” vs “a long battery life”
57 Application: Opinion Mining Why is opinion mining and sentiment analysis challenging? •  How much should every single opinion be worth? •  experts vs non-experts •  relationship trust •  reputation trust •  spammers •  frequent vs infrequent posters •  “experts” in one area may not be expert in another •  how frequently do other people agree?
58 Application: Opinion Mining Subtopics •  Opinion extraction: extract the piece of text which represents the opinion •  Cyrus has made a 23-song, purposely strange psych-rock record. Make no mistake, some of this album is unlistenable. But Cyrus is also too skilled of an artist to not place some beauty inside this madness, and Miley Cyrus and Her Dead Petz swerves into thoughtful territory when it’s least expected. •  Sentiment classification/orientation: extract the polarity of the opinion (e.g. positive, negative, neutral, or classify on a numerical scale) •  negative: purposely strange, some is unlistenable •  positive: skilled artist, beauty inside madness, thoughful •  Opinion summarisation: summarise the overall opinion about something •  Strange, some unlistenable: negative, skilled artist, beauty, thoughful: positive, Overall 6/10
59 Application: Opinion Mining Subtopics •  Feature-opinion association: given a text with target features and opinions extracted, decide which opinions comment on which features. •  “The tracks are good but not so keen on the cover art” •  Target identification: which thing is the opinion referring to? •  Source identification: who is holding the opinion?
60 Application: Opinion Mining Opinion Mining Resources Bing Liu’s English Sentiment Lexicon •  2006 pos words, 4783 neg words •  Useful properties: includes misspellings, morphological variants, slang •  Available from: http://www.cs.uic.edu/~liub/FBS/opinion-lexicon-English.rar The MPQA Subjectivity Lexicon •  Polarities: positive, negative, both, neutral •  Subjectivity: strongsubj or weaksubj •  Download from: http://mpqa.cs.pitt.edu/lexicons/subj_lexicon/
61 Application: Opinion Mining Opinion Mining Resources WordNet Affect •  Extension of WordNet with affect words •  Useful properties: includes POS categories •  Available from: http://wndomains.fbk.eu/wnaffect.html Hands-on session: Applying standard opinion mining lexicons with GATE •  Spoiler: general purpose lexicons do not always perform well, for better performance, domain- or context-specific lexicons are necessary
62Information Extraction with Linked Data Thank you for your attention!(And thank you to Diana Maynard for allowing me to adapt and reuse her Opinion Mining slides!) Questions? Isabelle Augenstein

More Related Content

PDF
Natural Language Processing for the Semantic Web
byIsabelle Augenstein
 
PPT
Natural language procssing
byRajnish Raj
 
PPTX
Different Semantic Perspectives for Question Answering Systems
byAndre Freitas
 
PPTX
Text Data Mining
byKU Leuven
 
PPTX
Knowledge acquisition using automated techniques
byUniversity of Melbourne, Australia
 
PDF
From Linked Data to Semantic Applications
byAndre Freitas
 
PPT
Download
bybutest
 
PPT
Download
bybutest
 
Natural Language Processing for the Semantic Web
byIsabelle Augenstein
 
Natural language procssing
byRajnish Raj
 
Different Semantic Perspectives for Question Answering Systems
byAndre Freitas
 
Text Data Mining
byKU Leuven
 
Knowledge acquisition using automated techniques
byUniversity of Melbourne, Australia
 
From Linked Data to Semantic Applications
byAndre Freitas
 
Download
bybutest
 
Download
bybutest
 

Similar to Information Extraction with Linked Data

PDF
IE: Named Entity Recognition (NER)
byMarina Santini
 
PDF
Information Extraction
byRubén Izquierdo Beviá
 
PDF
Entity Linking
bykrisztianbalog
 
PPTX
Semantic Perspectives for Contemporary Question Answering Systems
byAndre Freitas
 
PPT
sobha-ner.ppt named entity recognition model
byrajubandam694
 
PPT
Text Analytics: Yesterday, Today and Tomorrow
byTony Russell-Rose
 
PPT
Relationships at the Heart of Semantic Web: Modeling, Discovering, Validating...
byArtificial Intelligence Institute at UofSC
 
PDF
Knowledge Extraction and Linked Data: Playing with Frames
byValentina Presutti
 
PPTX
Semantic engagement
bySTIinnsbruck
 
PDF
Semantic engagement handouts
bySTIinnsbruck
 
PPT
Introduction
bysriniefs
 
PDF
Effective Semantics for Engineering NLP Systems
byAndre Freitas
 
PDF
Tutorial: Text Analytics for Security
byTao Xie
 
PPTX
Mining Web content for Enhanced Search
byRoi Blanco
 
PPTX
HotSoS16 Tutorial "Text Analytics for Security" by Tao Xie and William Enck
byTao Xie
 
PDF
Context-Enhanced Adaptive Entity Linking
byGiuseppe Rizzo
 
PDF
Mini seminar presentation on context-based NED optimization
byFilip Ilievski
 
PPTX
Text analysis-semantic-search
byDiana Maynard
 
PDF
Babak Rasolzadeh: The importance of entities
byZoltan Varju
 
ODP
NIF 2.0 Tutorial: Content Analysis and the Semantic Web
bySebastian Hellmann
 
IE: Named Entity Recognition (NER)
byMarina Santini
 
Information Extraction
byRubén Izquierdo Beviá
 
Entity Linking
bykrisztianbalog
 
Semantic Perspectives for Contemporary Question Answering Systems
byAndre Freitas
 
sobha-ner.ppt named entity recognition model
byrajubandam694
 
Text Analytics: Yesterday, Today and Tomorrow
byTony Russell-Rose
 
Relationships at the Heart of Semantic Web: Modeling, Discovering, Validating...
byArtificial Intelligence Institute at UofSC
 
Knowledge Extraction and Linked Data: Playing with Frames
byValentina Presutti
 
Semantic engagement
bySTIinnsbruck
 
Semantic engagement handouts
bySTIinnsbruck
 
Introduction
bysriniefs
 
Effective Semantics for Engineering NLP Systems
byAndre Freitas
 
Tutorial: Text Analytics for Security
byTao Xie
 
Mining Web content for Enhanced Search
byRoi Blanco
 
HotSoS16 Tutorial "Text Analytics for Security" by Tao Xie and William Enck
byTao Xie
 
Context-Enhanced Adaptive Entity Linking
byGiuseppe Rizzo
 
Mini seminar presentation on context-based NED optimization
byFilip Ilievski
 
Text analysis-semantic-search
byDiana Maynard
 
Babak Rasolzadeh: The importance of entities
byZoltan Varju
 
NIF 2.0 Tutorial: Content Analysis and the Semantic Web
bySebastian Hellmann
 

More from Isabelle Augenstein

PDF
Distant Supervision with Imitation Learning
byIsabelle Augenstein
 
PDF
Explainability for NLP
byIsabelle Augenstein
 
PDF
Towards Explainable Fact Checking
byIsabelle Augenstein
 
PDF
Learning to read for automated fact checking
byIsabelle Augenstein
 
PDF
1st Workshop for Women and Underrepresented Minorities (WiNLP) at ACL 2017 - ...
byIsabelle Augenstein
 
PDF
Automatically Detecting Scientific Misinformation
byIsabelle Augenstein
 
PPTX
Weakly Supervised Machine Reading
byIsabelle Augenstein
 
PDF
Learning with limited labelled data in NLP: multi-task learning and beyond
byIsabelle Augenstein
 
PDF
SemEval 2017 Task 10: ScienceIE – Extracting Keyphrases and Relations from Sc...
byIsabelle Augenstein
 
PDF
Extracting Relations between Non-Standard Entities using Distant Supervision ...
byIsabelle Augenstein
 
PDF
Accountable and Robust Automatic Fact Checking
byIsabelle Augenstein
 
PDF
Towards Explainable Fact Checking (DIKU Business Club presentation)
byIsabelle Augenstein
 
PDF
Determining the Credibility of Science Communication
byIsabelle Augenstein
 
PDF
Machine Reading Using Neural Machines (talk at Microsoft Research Faculty Sum...
byIsabelle Augenstein
 
PDF
Multi-task Learning of Pairwise Sequence Classification Tasks Over Disparate ...
byIsabelle Augenstein
 
PDF
Tracking False Information Online
byIsabelle Augenstein
 
PPTX
Beyond Fact Checking — Modelling Information Change in Scientific Communication
byIsabelle Augenstein
 
PDF
1st Workshop for Women and Underrepresented Minorities (WiNLP) at ACL 2017 - ...
byIsabelle Augenstein
 
PDF
What can typological knowledge bases and language representations tell us abo...
byIsabelle Augenstein
 
PDF
USFD at SemEval-2016 - Stance Detection on Twitter with Autoencoders
byIsabelle Augenstein
 
Distant Supervision with Imitation Learning
byIsabelle Augenstein
 
Explainability for NLP
byIsabelle Augenstein
 
Towards Explainable Fact Checking
byIsabelle Augenstein
 
Learning to read for automated fact checking
byIsabelle Augenstein
 
1st Workshop for Women and Underrepresented Minorities (WiNLP) at ACL 2017 - ...
byIsabelle Augenstein
 
Automatically Detecting Scientific Misinformation
byIsabelle Augenstein
 
Weakly Supervised Machine Reading
byIsabelle Augenstein
 
Learning with limited labelled data in NLP: multi-task learning and beyond
byIsabelle Augenstein
 
SemEval 2017 Task 10: ScienceIE – Extracting Keyphrases and Relations from Sc...
byIsabelle Augenstein
 
Extracting Relations between Non-Standard Entities using Distant Supervision ...
byIsabelle Augenstein
 
Accountable and Robust Automatic Fact Checking
byIsabelle Augenstein
 
Towards Explainable Fact Checking (DIKU Business Club presentation)
byIsabelle Augenstein
 
Determining the Credibility of Science Communication
byIsabelle Augenstein
 
Machine Reading Using Neural Machines (talk at Microsoft Research Faculty Sum...
byIsabelle Augenstein
 
Multi-task Learning of Pairwise Sequence Classification Tasks Over Disparate ...
byIsabelle Augenstein
 
Tracking False Information Online
byIsabelle Augenstein
 
Beyond Fact Checking — Modelling Information Change in Scientific Communication
byIsabelle Augenstein
 
1st Workshop for Women and Underrepresented Minorities (WiNLP) at ACL 2017 - ...
byIsabelle Augenstein
 
What can typological knowledge bases and language representations tell us abo...
byIsabelle Augenstein
 
USFD at SemEval-2016 - Stance Detection on Twitter with Autoencoders
byIsabelle Augenstein
 

Recently uploaded

PPTX
application security presentation 2 by harman
byharmanideapad
 
PPTX
SNG460-CNG489_Week8_17-21Nov25_Chp8-OdtuClass.pptx
byberayseray382
 
PDF
Kiến trúc hướng dịch vụ UET!!!!!!!!!!!!!!!!!!!!!!!
bydoquangminh962004
 
PPTX
Computer_Virus_Presentation_With_Slide7.pptx
byHEVISLIVE
 
PPTX
Privacy Preserving Detection of Sensitive Data Exposure (1).pptx
byNikitaMandhan4
 
PDF
DSD-INT 2025 Hydrological Modelling in Society Assumptions, Impacts, and Appl...
byDeltares
 
PPTX
Code Assessment Tools .pptx
byCodexpro
 
PPTX
NSF Converter Software to Convert NSF to PST, EML, MSG
byNSF Converter Software
 
PPTX
List on Python Programming Language.pptx
byRICHARDALONSAGAY1
 
PDF
DSD-INT 2025 Introduction to Hydrology Suite - Slootjes
byDeltares
 
PPTX
Lecture 3 - Scheduling - Operating System
bynurulalomador
 
PDF
Jeremy Millul - An NYU Computer Science Graduate
byJeremy Millul
 
PDF
Using Agentic RAG for Research - Demonstration of NotebookLM
byRathish Chandra Gatti,Ph.D
 
PDF
One Agent to Rule Them All - The Fellowship of AI Agents
byIvo Andreev
 
PPTX
OTT Content Analytics Enhanced by ALTBalaji Data Scraping.pptx
byyashpatric
 
PDF
Revolutionising-SQL-Data-Modelling-with-SqlDBM.pdf
bySQL DBM
 
PDF
Database Management Systems(DBMS):UNIT-II Relational Data Model BCA SEP SEM ...
byKuvempu University
 
PDF
DSD-INT 2025 Integrating Nature-Based Solutions into Hydrological Models A Li...
byDeltares
 
PDF
How NetSuite Cloud ERP Helps Businesses Overcome Legacy System Downtime.
byTechWize
 
PDF
Solved First Semester B.C.A. C Programming Question Paper – Jan/Feb 2025
byKuvempu University
 
application security presentation 2 by harman
byharmanideapad
 
SNG460-CNG489_Week8_17-21Nov25_Chp8-OdtuClass.pptx
byberayseray382
 
Kiến trúc hướng dịch vụ UET!!!!!!!!!!!!!!!!!!!!!!!
bydoquangminh962004
 
Computer_Virus_Presentation_With_Slide7.pptx
byHEVISLIVE
 
Privacy Preserving Detection of Sensitive Data Exposure (1).pptx
byNikitaMandhan4
 
DSD-INT 2025 Hydrological Modelling in Society Assumptions, Impacts, and Appl...
byDeltares
 
Code Assessment Tools .pptx
byCodexpro
 
NSF Converter Software to Convert NSF to PST, EML, MSG
byNSF Converter Software
 
List on Python Programming Language.pptx
byRICHARDALONSAGAY1
 
DSD-INT 2025 Introduction to Hydrology Suite - Slootjes
byDeltares
 
Lecture 3 - Scheduling - Operating System
bynurulalomador
 
Jeremy Millul - An NYU Computer Science Graduate
byJeremy Millul
 
Using Agentic RAG for Research - Demonstration of NotebookLM
byRathish Chandra Gatti,Ph.D
 
One Agent to Rule Them All - The Fellowship of AI Agents
byIvo Andreev
 
OTT Content Analytics Enhanced by ALTBalaji Data Scraping.pptx
byyashpatric
 
Revolutionising-SQL-Data-Modelling-with-SqlDBM.pdf
bySQL DBM
 
Database Management Systems(DBMS):UNIT-II Relational Data Model BCA SEP SEM ...
byKuvempu University
 
DSD-INT 2025 Integrating Nature-Based Solutions into Hydrological Models A Li...
byDeltares
 
How NetSuite Cloud ERP Helps Businesses Overcome Legacy System Downtime.
byTechWize
 
Solved First Semester B.C.A. C Programming Question Paper – Jan/Feb 2025
byKuvempu University
 

Information Extraction with Linked Data

  • 1.
    Information Extraction with LinkedData Isabelle Augenstein Department of Computer Science, University of Sheffield, UK i.augenstein@sheffield.ac.uk 2 September 2015 Information Extraction with Linked Data Tutorial, ESWC Summer School 2015
  • 2.
  • 3.
    3Why Information Extraction? IsabelleAugenstein semi-structured information unstructured information
  • 4.
    4Why Information Extraction? semi-structuredinformation unstructured information How to link this information to a knowledge base automatically?
  • 5.
    5Why Information Extraction? semi-structuredinformation unstructured information How to link this information to a knowledge base automatically? Information Extraction!
  • 6.
    6 Information Extraction Isabelle Augenstein TheArctic Monkeys almost exclusively played songs from their new album AM at Summerfest 2014 at Miller Lite Oasis in Milwaukee on 25 June 2014.
  • 7.
    7 Information Extraction Isabelle Augenstein TheArctic Monkeys almost exclusively played songs from their new album AM at Summerfest 2014 at Miller Lite Oasis in Milwaukee on 25 June 2014. Named Entity Recognition
  • 8.
    8 Information Extraction Isabelle Augenstein TheArctic Monkeys almost exclusively played songs from their new album AM at Summerfest 2014 at Miller Lite Oasis in Milwaukee on 25 June 2014. Named Entity Recognition Named Entity Classification (NEC): Arctic Monkeys: mo:MusicArtist AM: mo:SignalGroup Summerfest 2014: mo:Festival Miller Lite Oases: geo:SpatialThing Milwaukee: geo:SpatialThing
  • 9.
    9 Information Extraction Isabelle Augenstein TheArctic Monkeys almost exclusively played songs from their new album AM at Summerfest 2014 at Miller Lite Oasis in Milwaukee on 25 June 2014. Named Entity Recognition Named Entity Classification (NEC): Named Entity Linking (NEL): Arctic Monkeys: mo:MusicArtist Arctic Monkeys: mo:artist/ada7a83 ... AM: mo:SignalGroup AM: mo:release-group/a348ba2f-f8b3 … Summerfest 2014: mo:Festival Summerfest 2014: mo:event/3fc3 … Miller Lite Oases: geo:SpatialThing Miller Lite Oases: mo:place/3f26acf … Milwaukee: geo:SpatialThing Milwaukee: mo:area/4dc3fa97-cf9b- …
  • 10.
    10 Named Entities: Definition NamedEntities: Proper nouns, which refer to real-life entities Named Entity Recognition: Detecting boundaries of named entities (NEs) Named Entity Classification: Assigning classes to NEs, such as PERSON, LOCATION, ORGANISATION, MISC or fine-grained classes such as SIGNAL GROUP Named Entity Linking / Disambiguation: Linking NEs to concrete entries in knowledge base, example: Milwaukee -> LOC: largest city in the U.S. state of Wisconsin -> LOC: Milwaukee, Oregon, named after the city in Wisconsin -> LOC: Milwaukee County, Wisconsin -> ORG: Milwaukee Tool Corp, a manufacturer of electric power tools -> MISC: early codename for what was to become the Macintosh II -> …
  • 11.
    11 Relations Isabelle Augenstein The ArcticMonkeys almost exclusively played songs from their new album AM at Summerfest 2014 at Miller Lite Oasis in Milwaukee on 25 June 2014. Named Entity Recognition Relation Extraction foaf:made gn:parentFeature mo:Festival
  • 12.
    12 Relations Isabelle Augenstein The ArcticMonkeys almost exclusively played songs from their new album AM at Summerfest 2014 at Miller Lite Oasis in Milwaukee on 25 June 2014. Named Entity Recognition Relation Extraction Temporal Extraction foaf:made gn:parentFeature mo:Festival 2014-06-25
  • 13.
    13 Relations Isabelle Augenstein The ArcticMonkeys almost exclusively played songs from their new album AM at Summerfest 2014 at Miller Lite Oasis in Milwaukee on 25 June 2014. Named Entity Recognition Relation Extraction Temporal Extraction foaf:made gn:parentFeature mo:Festival 2014-06-25 Event Extraction Event: mo:Festival: Summerfest 2014 foaf:Agent: Arctic Monkeys time:TemporalEntity: 2014-06-25 geo:SpatialThing: Miller Lite Oasis
  • 14.
    14 Relations, Time Expressions andEvents: Definition Relations: Two or more entities which relate to one another in real life Relation Extraction: Detecting relations between entities and assigning relation types to them, such as LOCATED-IN Temporal Extraction: Recognising and normalising time expressions: times (e.g. “3 in the afternoon”), dates (“tomorrow”), durations (“since yesterday”), and sets (e.g. “twice a month”) Events: Real-life events that happened at some point in space and time, e.g. music festival, album release Event Extraction: Extracting events consisting of the name and type of event, agent, time and location
  • 15.
    15 Summary: Introduction •  Informationextraction (IE) methods such as named entity recognition (NER), named entity classification (NEC), named entity linking, relation extraction (RE), temporal extraction, and event extraction can help to add markup to Web pages •  Information extraction approaches can serve two purposes: •  Annotating every single mention of an entity, relation or event, e.g. to add markup to Web pages •  Aggregating those mentions to populate knowledge bases, e.g. based on confidence values and majority voting Milwaukee LOC 0.9 Milwaukee LOC 0.8 Milwaukee ORG 0.4 à Milwaukee LOC Isabelle Augenstein
  • 16.
    16 NERC: Methods •  Possiblemethodologies •  Rule-based approaches: write manual extraction rules •  Machine learning based approaches •  Supervised learning: manually annotate text, train machine learning model •  Unsupervised learning: extract language patterns, cluster similar ones •  Semi-supervised learning: start with a small number of language patterns, iteratively learn more (bootstrapping) •  Gazetteer-based method: use existing list of named entities •  Combination of the above Isabelle Augenstein
  • 17.
    17 NERC: Methods Developing aNERC involves programming based around APIs.. Isabelle Augenstein
  • 18.
    18 NERC: Methods Developing aNERC involves programming based around APIs.. which can be frustrating at times Isabelle Augenstein
  • 19.
    19 NERC: Methods and (atleast basic) knowledge about linguistics Isabelle Augenstein
  • 20.
    20 Background: Linguistics Sentences Tokens Morphemes Words Sentences Meaning Isabelle Augenstein Lexicon Morphology Thefarmer hit the donkey. Syntax Lexicon Semantics, Discourse The, farmer, hit, the, donkey, . wait + ed -> waited, cat -> cat wait -> V, cat -> N The (D) farmer (N) hit (V) the (D) donkey (N). NP Every farmer who owns a donkey beats it. ∀x∀y (farmer(x) ∧ donkey(y) ∧ own(x, y) → beat(x, y)) NP
  • 21.
    21 Background: NLP Tasks Sentences Tokens Morphemes Words Sentences Meaning IsabelleAugenstein Sentence splitting Tokenisation Lexicon Morphology Syntax Lemmatisation or stemming, part of speech (POS) tagging Chunking, parsing Lexicon Semantics, Discourse Semantic and discourse analysis, anaphora resolution
  • 22.
  • 23.
    23 Background: Linguistics Sentences Tokens Morphemes Words Sentences Meaning Isabelle Augenstein NewYork-based [New, York-based] or [New, York, -, based] Lexicon Morphology Syntax Lexicon Semantics, Discourse
  • 24.
  • 25.
    25 Background: Linguistics Sentences Tokens Morphemes Words Sentences Meaning Isabelle Augenstein Lexicon Morphology Syntax LexiconShe’d -> she would, she had Semantics, Discourse New York-based [New, York-based] or [New, York, -, based]
  • 26.
    26 Background: Linguistics Sentences Tokens Morphemes Words Sentences Meaning Isabelle Augenstein Lexicon Morphology Syntax LexiconShe’d -> she would, she had Semantics, Discourse New York-based [New, York-based] or [New, York, -, based] Time flies like an arrow
  • 27.
    27 Background: Linguistics Sentences Tokens Morphemes Words Sentences Meaning Isabelle Augenstein Lexicon Morphology Syntax LexiconShe’d -> she would, she had Semantics, Discourse New York-based [New, York-based] or [New, York, -, based] Time flies(V/N) like(V/P) an arrow
  • 28.
    28 Background: Linguistics Sentences Tokens Morphemes Words Sentences Meaning Isabelle Augenstein Lexicon Morphology Syntax LexiconShe’d -> she would, she had Semantics, Discourse New York-based [New, York-based] or [New, York, -, based] Time flies(V/N) like(V/P) an arrow The woman saw the man with the binoculars.
  • 29.
    29 Background: Linguistics Sentences Tokens Morphemes Words Sentences Meaning Isabelle Augenstein Lexicon Morphology Syntax LexiconShe’d -> she would, she had Semantics, Discourse New York-based [New, York-based] or [New, York, -, based] Time flies(V/N) like(V/P) an arrow The woman saw the man with the binoculars. -> Who had the binoculars?
  • 30.
    30 Background: Linguistics Sentences Tokens Morphemes Words Sentences Meaning Isabelle Augenstein Lexicon Morphology Syntax LexiconShe’d -> she would, she had Semantics, Discourse New York-based [New, York-based] or [New, York, -, based] Time flies(V/N) like(V/P) an arrow The woman saw the man with the binoculars. -> Who had the binoculars? Somewhere in Britain, some woman has a child every thirty seconds.
  • 31.
    31 Background: Linguistics Sentences Tokens Morphemes Words Sentences Meaning Isabelle Augenstein Lexicon Morphology Syntax LexiconShe’d -> she would, she had Semantics, Discourse New York-based [New, York-based] or [New, York, -, based] Time flies(V/N) like(V/P) an arrow The woman saw the man with the binoculars. -> Who had the binoculars? Somewhere in Britain, some woman has a child every thirty seconds. -> Same woman or different women?
  • 32.
    32 Background: Linguistics Sentences Tokens Morphemes Words Sentences Meaning Isabelle Augenstein Lexicon Morphology Syntax LexiconShe’d -> she would, she had Semantics, Discourse New York-based [New, York-based] or [New, York, -, based] Time flies(V/N) like(V/P) an arrow The woman saw the man with the binoculars. -> Who had the binoculars? Somewhere in Britain, some woman has a child every thirty seconds. -> Same woman or different women? Ambiguities on every level
  • 33.
    33 Background: Linguistics Sentences Tokens Morphemes Words Sentences Meaning Isabelle Augenstein Lexicon Morphology Syntax LexiconShe’d -> she would, she had Semantics, Discourse New York-based [New, York-based] or [New, York, -, based] Time flies(V/N) like(V/P) an arrow The woman saw the man with the binoculars. -> Who had the binoculars? Somewhere in Britain, some woman has a child every thirty seconds. -> Same woman or different women? Ambiguities on every level Y U SO AMBIGUOUS?
  • 34.
    34 Information Extraction Language isambiguous.. Can we still build named entity extractors that extract all entities from unseen text correctly? Isabelle Augenstein
  • 35.
    35 Information Extraction Language isambiguous.. Can we still build named entity extractors that extract all entities from unseen text correctly? Isabelle Augenstein
  • 36.
    36 Information Extraction Language isambiguous.. Can we still build named entity extractors that extract all entities from unseen text correctly? However, we can try to extract most of them correctly using linguistic cues and background knowledge! Isabelle Augenstein
  • 37.
    37 NERC: Features What canhelp to recognise and/or classify named entities? •  Words: •  Words in window before and after mention •  Sequences •  Bags of words Summerfest 2014 took place at Miller Lite Oasis in Milwaukee on 25 June 2014. w: Milwaukee w-1: in w-2: Oasis w+1: on w+2: 25 seq[-]: Oasis in seq[+]: on 25 bow: Milwaukee bow[-]: in bow[-]: Oasis bow[+]: on bow[+]: 25 Isabelle Augenstein
  • 38.
    38 NERC: Features What canhelp to recognise and/or classify named entities? •  Morphology: •  Capitalisation: is upper case (China), all upper case (IBM), mixed case (eBay) •  Symbols: contains $, £, €, roman symbols (IV), .. •  Contains period (google.com), apostrophe (Mandy’s), hyphen (speed-o- meter), ampersand (Fisher & Sons) •  Stem or Lemma (cats -> cat), prefix (disadvantages -> dis), suffix (cats -> s), interfix (speed-o-meter -> o) Isabelle Augenstein
  • 39.
    39 NERC: Features What canhelp to recognise and/or classify named entities? •  POS (part of speech) tags •  Most named entities are nouns •  Prokofyev (2014) Isabelle Augenstein 4.1 Part-of-Speech Tags Part-Of-Speech (POS) tags have often been considered as an important discriminative feature for term identification. Many works on key term identification apply either fixed or regular expression POS tag patterns to improve their ef- fectiveness. Nonetheless, POS tags alone cannot produce high-quality results. As can be seen from the overall POS tag distribution graph extracted from one of our collections (see Figure 3), many of the most frequent tag patterns (e.g., JJ NN tagging adjectives and nouns6 ) are far from yielding perfect results. Figure 3: Top 6 most frequent part-of-speech tag comma, while the n-gram “au tion”, which is indeed a valid the beginning or at the end The contingency tables giv trate this: The +punctuation respectively, the counts of the punctuation mark in any of of the n-grams that have no occurrences. From the tables of punctuation marks (+pun an n-gram occurs twice as o valid entities compared to th that the absence of punctuat pens less frequently for the va ones. Table 1: Contingency ta appearing immediately be Va +punctuation 16 punctuation 65 Totals 81 Table 2: Contingency ta appearing immediately af Va
  • 40.
  • 41.
    41Morphology: Penn TreebankPOS tags Nouns (all start with N) Verbs (all start with V) Adjectives (all start with J)
  • 42.
    42 NERC: Features What canhelp to recognise and/or classify named entities? •  POS (part of speech) tags •  Most named entities are nouns •  Prokofyev (2014) Isabelle Augenstein 4.1 Part-of-Speech Tags Part-Of-Speech (POS) tags have often been considered as an important discriminative feature for term identification. Many works on key term identification apply either fixed or regular expression POS tag patterns to improve their ef- fectiveness. Nonetheless, POS tags alone cannot produce high-quality results. As can be seen from the overall POS tag distribution graph extracted from one of our collections (see Figure 3), many of the most frequent tag patterns (e.g., JJ NN tagging adjectives and nouns6 ) are far from yielding perfect results. Figure 3: Top 6 most frequent part-of-speech tag comma, while the n-gram “au tion”, which is indeed a valid the beginning or at the end The contingency tables giv trate this: The +punctuation respectively, the counts of the punctuation mark in any of of the n-grams that have no occurrences. From the tables of punctuation marks (+pun an n-gram occurs twice as o valid entities compared to th that the absence of punctuat pens less frequently for the va ones. Table 1: Contingency ta appearing immediately be Va +punctuation 16 punctuation 65 Totals 81 Table 2: Contingency ta appearing immediately af Va
  • 43.
    43 NERC: Features What canhelp to recognise and/or classify named entities? •  Gazetteers •  Retrieved from HTML lists or tables [1] •  Using regular expressions patterns and search engines (e.g. “Popular artists such as * ”) •  Retrieved from knowledge bases [1] https://en.wikipedia.org/wiki/Billboard_200 Isabelle Augenstein
  • 44.
    44 NERC: Training Models Extensivechoice of machine learning algorithms for training NERCs
  • 45.
    45 NERC: Training Models • Unfortunately, there isn’t enough time to explain machine learning algorithms in detail •  CRFs (conditional random fields) are one of the most widely used algorithms for NERC •  Graphical models, view NERC as a sequence labelling task •  Named entities consist of a beginning token (B), inside tokens (I), and outside tokens (O) took(O) place(O) at(O) Miller(B-LOC) Lite(I-LOC) Oasis(I-LOC) in(O) •  For now, we will rule- and gazetteer-based NERC •  It is fairly easy to write manual extraction rules for NEs, can achieve a high performance when combined with gazetteers •  This can be done with the GATE software (general architecture for text engineering) and Jape rules -> Hands-on session Isabelle Augenstein
  • 46.
    46 NLP & MLSoftware Natural Language Processing: -  GATE (general purpose architecture, includes other NLP and ML software as plugins) -  Stanford NLP (Java) -  OpenNLP (Java) -  NLTK (Python) Machine Learning: -  scikit-learn (Python, rich documentation, highly recommended!) -  Mallet (Java) -  WEKA (Java) -  Alchemy (graphical models, Java) -  FACTORIE, wolfe (graphical models, Scala) -  CRFSuite (efficient implementation of CRFs, Python) Isabelle Augenstein
  • 47.
    47 NLP & MLSoftware Ready to use NERC software: -  ANNIE (rule-based, part of GATE) -  Wikifier (based on Wikipedia) -  FIGER (based on Wikipedia, fine-grained Freebase NE classes) Almost ready to use NERC software: -  CRFSuite (already includes Python implementation for feature extraction, you just need to feed it with training data, which you can also download) Ready to use RE software: -  ReVerb (Open IE, extracts patterns for any kind of relation) -  MultiR (Distant supervision, relation extractor trained on Freebase) Web Content Extraction software: -  Boilerpipe (extract main text content from Web pages) -  Jsoup (traverse elements of Web pages individually, also allows to extract text) Isabelle Augenstein
  • 48.
    48 Application: Opinion Mining • Extracting opinions or sentiments in text •  It’s about finding out what people think University of Sheffield, NLP It's about finding out what people think...
  • 49.
    49 Application: Opinion Mining • Opinion Mining is big business •  Someone just bought an album by a music artist •  Writes a review about it •  Someone else wants to buy an album •  Looks up reviews by fans and music critics •  Music artist and music producer •  Get feedback from fans •  Improve their product •  Improve their marketing strategy
  • 50.
    50 Application: Opinion Mining • “Miley Cyrus's attempts to shock would be more effective if she had songs to back up the posturing.” – The Guardian •  “Bangerz is an Amazing album with great lyrics and we can see the Miley Cyrus' musical evolution. Would love to buy it and I already did. ALBUM OF THE YEAR. Peace” – Rodolfoalmeida3
  • 51.
    51 Application: Opinion Mining Whyis opinion mining and sentiment analysis challenging? •  Relatively easy to find sentiment words in sentences, difficult to identify which topic they are about
  • 52.
    52 Application: Opinion Mining Whyis opinion mining and sentiment analysis challenging? •  Relatively easy to find sentiment words in sentences, difficult to identify which topic they are about •  “The album comes with a free bonus CD but I don't like the cover art much.” Does this refer to the cover art of the bonus CD or the album?
  • 53.
    53 Application: Opinion Mining Whyis opinion mining and sentiment analysis challenging? •  Relatively easy to find sentiment words in sentences, difficult to identify which topic they are about
  • 54.
    54 Application: Opinion Mining Whyis opinion mining and sentiment analysis challenging? •  Relatively easy to find sentiment words in sentences, difficult to identify which topic they are about •  Whitney Houston was quite unpopular… University of Sheffield, NLP Whitney Houston wasn't very popular...
  • 55.
    55 Application: Opinion Mining Whyis opinion mining and sentiment analysis challenging? •  Relatively easy to find sentiment words in sentences, difficult to identify which topic they are about •  Whitney Houston was quite unpopular… or was she? •  Death confuses opinion mining tools University of Sheffield, NLP Or was she?
  • 56.
    56 Application: Opinion Mining Whyis opinion mining and sentiment analysis challenging? •  It’s not just about finding sentiment words, context is important too •  “It's a great movie if you have the taste and sensibilities of a 5- year-old boy.” •  “It's terrible Candidate X did so well in the debate last night.” •  “I'd have liked the track a lot more if it had been a bit shorter.” •  If sentiment words are neutral, negative or positive depends on domain •  “a long track” vs “a long walk” vs “a long battery life”
  • 57.
    57 Application: Opinion Mining Whyis opinion mining and sentiment analysis challenging? •  How much should every single opinion be worth? •  experts vs non-experts •  relationship trust •  reputation trust •  spammers •  frequent vs infrequent posters •  “experts” in one area may not be expert in another •  how frequently do other people agree?
  • 58.
    58 Application: Opinion Mining Subtopics • Opinion extraction: extract the piece of text which represents the opinion •  Cyrus has made a 23-song, purposely strange psych-rock record. Make no mistake, some of this album is unlistenable. But Cyrus is also too skilled of an artist to not place some beauty inside this madness, and Miley Cyrus and Her Dead Petz swerves into thoughtful territory when it’s least expected. •  Sentiment classification/orientation: extract the polarity of the opinion (e.g. positive, negative, neutral, or classify on a numerical scale) •  negative: purposely strange, some is unlistenable •  positive: skilled artist, beauty inside madness, thoughful •  Opinion summarisation: summarise the overall opinion about something •  Strange, some unlistenable: negative, skilled artist, beauty, thoughful: positive, Overall 6/10
  • 59.
    59 Application: Opinion Mining Subtopics • Feature-opinion association: given a text with target features and opinions extracted, decide which opinions comment on which features. •  “The tracks are good but not so keen on the cover art” •  Target identification: which thing is the opinion referring to? •  Source identification: who is holding the opinion?
  • 60.
    60 Application: Opinion Mining OpinionMining Resources Bing Liu’s English Sentiment Lexicon •  2006 pos words, 4783 neg words •  Useful properties: includes misspellings, morphological variants, slang •  Available from: http://www.cs.uic.edu/~liub/FBS/opinion-lexicon-English.rar The MPQA Subjectivity Lexicon •  Polarities: positive, negative, both, neutral •  Subjectivity: strongsubj or weaksubj •  Download from: http://mpqa.cs.pitt.edu/lexicons/subj_lexicon/
  • 61.
    61 Application: Opinion Mining OpinionMining Resources WordNet Affect •  Extension of WordNet with affect words •  Useful properties: includes POS categories •  Available from: http://wndomains.fbk.eu/wnaffect.html Hands-on session: Applying standard opinion mining lexicons with GATE •  Spoiler: general purpose lexicons do not always perform well, for better performance, domain- or context-specific lexicons are necessary
  • 62.
    62Information Extraction withLinked Data Thank you for your attention!(And thank you to Diana Maynard for allowing me to adapt and reuse her Opinion Mining slides!) Questions? Isabelle Augenstein