information extraction
TRANSCRIPT
![Page 2: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/2.jpg)
Text Mining Course • 1) Introduction to Text Mining • 2) Introduction to NLP • 3) Named Entity Recognition and Disambiguation • 4) Opinion Mining and Sentiment Analysis • 5) Information Extraction
• 6) NewsReader and Visualisation • 7) Guest Lecture and Q&A
![Page 3: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/3.jpg)
Outline 1. What is Information Extraction 2. Main goals of Information Extraction 3. Information Extraction Tasks and Subtasks 4. MUC conferences 5. Main domains of Information Extraction 6. Methods for Information Extraction
o Cascaded finite-state transducers o Regular expressions and patterns o Supervised learning approaches o Weakly supervised and unsupervised approaches
7. How far we are with IE
![Page 4: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/4.jpg)
What is IE? • Late 1970s within NLP field
• Find and extract automatically limited relevant parts of texts
• Merge information from many pieces of text
![Page 5: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/5.jpg)
What is IE? • Quite often in specialized domains
• Move from unstructured/semi-structured data to structured data o Schemas o Relations (as a database) o Knowledge base o RDF triples
![Page 6: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/6.jpg)
What is IE? Unstructured text
• Natural language sentences • Historically NLP system have been designed to process this type of data • The meaning à linguistic analysis and natural language understanding
![Page 7: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/7.jpg)
What is IE? Semi-‐‑structured text
• The physical layout helps to the interpretation • Processing half way linguistic features ßà positional features
![Page 8: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/8.jpg)
What is IE?
![Page 9: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/9.jpg)
Main goals of IE • Fill a predefined “template” from raw text
• Extract who did what to whom and when? o Event extraction
• Organize information so that is useful to people
• Put information in a form that allows further inferences by computers o Big data
![Page 10: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/10.jpg)
IE. Task & Subtasks • Named Entity Recognition
o Detection à Mr. Smith eats bitterballen [Mr. Smith] : ENTITY o Classification à Mr. Smith eats bitterballen [Mr. Smith] : PERSON
• Event extraction o The thief broke the door with a hammer
• CAUSE_HARMà Verb: break Agent: the thief Patient: the door Instrument: a hammer
• Coreference resolution o [Mr. Smith] eats bitterballen. Besides to this, [he] only drinks Belgium beer.
![Page 11: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/11.jpg)
IE. Task & Subtasks • Relationship extraction
o Bill works for IBM PERSON works for ORGANISATION
• Terminology extraction o Finding relevant terms of multi words from a given corpus
• Some concrete examples o Extracting earnings, profits, board members, headquarters from company
reports o Searching on the WWW for e-mails for advertising (spamming) o Learn drug-gene product interactions from biomedical research papers
![Page 12: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/12.jpg)
IE Tasks & Subtasks • Apple mail
![Page 13: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/13.jpg)
MUC conferences • Message Understanding Conference (MUC), held
between 1987 and 1998.
• Domain specific texts + training examples + template definition
• Precision, Recall and F1 as evaluation
• Domains o MUC-1 (1987), MUC-2 (1989): Naval operations messages. o MUC-3 (1991), MUC-4 (1992): Terrorism in Latin American countries. o MUC-5 (1993): Joint ventures and microelectronics domain. o MUC-6 (1995): News articles on management changes. o MUC-7 (1998): Satellite launch reports.
![Page 14: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/14.jpg)
MUC conferences Bridgestone Sports Co. said Friday it has set up a joint venture in Taiwan with a local concern and a Japanese trading house to produce golf clubs to be shipped to Japan. The joint venture, Bridgestone Sports Taiwan Co., capitalized at 20 million new Taiwan dollars, will start production in January 1990 with production of 20,000 iron and “metal wood” clubs a month.
Example from MUC5
![Page 15: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/15.jpg)
Main domains of IE • Terrorist events
• Joint ventures
• Plane crashes
• Disease outbreaks
• Seminar announcements
• Biological and medical domain
![Page 16: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/16.jpg)
Outline 1. What is Information Extraction 2. Main goals of Information Extraction 3. Information Extraction Tasks and Subtasks 4. MUC conferences 5. Main domains of Information Extraction 6. Methods for Information Extraction
o Cascaded finite-state transducers o Regular expressions and patterns o Supervised learning approaches o Weakly supervised and unsupervised approaches
7. How far we are with IE
![Page 17: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/17.jpg)
Methods for IE • Cascaded finite-state transducers
o Rule based o Regular expressions
• Learning based approaches o Traditional classifiers
• Bayes, MME, SVM … o Sequence label models
• HMM, CMM, CRF
• Unsupervised approaches
• Hybrid approaches
![Page 18: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/18.jpg)
Cascaded finite-‐‑state transducers
• Emerging idea from MUC participants and approaches
• Decompose the task into small sub-tasks
• One element is read at a time from a sequence o Depending on the type a certain transition in produced in the automaton
to a new state
o Some states are considered final (the input matches a certain pattern)
• Can be defined as a regular expression
![Page 19: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/19.jpg)
Cascaded finite-‐‑state transducers
Finite Automaton for noun groups
=> John’s interesting book with a nice cover
![Page 20: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/20.jpg)
Cascaded finite-‐‑state transducers
• Earlier stages recognize smaller linguistics objects o Usually domain independent
• Later stages build on top of the previous ones o Usually domain dependent
• Typical IE systems 1. Complex words 2. Basic phrases 3. Complex phrases 4. Domain events 5. Merging structures
![Page 21: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/21.jpg)
Cascaded finite-‐‑state transducers
• Complex words o Multiwords: “set up” “trading house” o NE: “Bridgestone Sports Co”
• Basic Phrases o Syntactic chunking
• Noun groups (head noun + all modifiers) • Verb groups
![Page 22: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/22.jpg)
Cascaded finite-‐‑state transducers
![Page 23: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/23.jpg)
Cascaded finite-‐‑state transducers
• Complex phrases o Complex noun and verb groups on the basis of syntactic information
• The attachment of appositives to their head noun group o “The joint venture, Bridgestone Sports Taiwan Co.,”
• The construction of measure phrases o “20,000 iron and ‘metal wood’ clubs a month”
![Page 24: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/24.jpg)
Cascaded finite-‐‑state transducers
• Domain events o Recognize events and match with “fillers” detected in previous steps
o Requires domain specific patterns • To recognize phrases of interest • To define what are the roles
o Patterns can be defined also as a finite-state machines or regular expressions
• <Company/ies><Set-up><Joint-Venture> with <Company/ies> • <Company><Capitalized> at <Currency>
![Page 25: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/25.jpg)
Cascaded finite-‐‑state transducers
![Page 26: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/26.jpg)
Regular Expressions • 1950’s Stephen Kleene • A string pattern that describes/matches a set of
strings
• A regular expression consists of: o Characters
o Operation symbols • Boolean (and/or) • Grouping (for defining scopes) • Quantification
![Page 27: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/27.jpg)
Regular Expressions Character Description a The character a . Any single character [abc] Any character in the brackets (OR) ‘a’
or ‘b’ or ‘c’ [^abc]
Any character not in the brackets. Any symbol that is not ‘a ‘ or ‘b’ or ‘c’
* Quantifier. Matches the preceding element ZERO or more times
+ Quantifier. Matches the preceding element ONE or more times
? Matches the previous element zero or one time
| Choice (OR) Matches one of the expressions (before of after the |)
![Page 28: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/28.jpg)
Regular Expressions ① .at è ???
![Page 29: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/29.jpg)
Regular Expressions ① .at è hat cat bat xat … ② [hc]at è hat cat ③ [^b]at è all matched by .at but “bat” ④ [^hc]at è all match by .at but “hat” and
“cat” ⑤ s.* è s sssss ssbsd2ck3e
![Page 30: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/30.jpg)
Regular Expressions ① .at è hat cat bat xat … ② [hc]at è hat cat ③ [^b]at è all matched by .at but “bat” ④ [^hc]at è all match by .at but “hat” and
“cat” ⑤ s.* è s sssss ssbsd2ck3e ⑥ [hc]*at è hat cat hhat chat cchhat at … ⑦ cat|dog è cat dog ⑧ …. ⑨ ….
![Page 31: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/31.jpg)
Using Regular Expressions
• Typically extracting information from automatic generated webpages is easy
o Wikipedia
• To know the country for a given city o Amazon webpage
• From a list of hits
o Weather forecast webpages
o DBpedia
![Page 32: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/32.jpg)
![Page 33: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/33.jpg)
![Page 34: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/34.jpg)
Using Regular Expressions
![Page 35: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/35.jpg)
Using Regular Expressions
• Some “unstructured” pieces of information keep some structure and are easy to capture by means of regular expressions o Phone numbers
o What else?
o …
o ...
![Page 36: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/36.jpg)
Using Regular Expressions
• Some “unstructured” pieces of information keep some structure and are easy to capture by means of regular expressions o Phone numbers
o E-mails
o URL Websites
![Page 37: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/37.jpg)
Using Regular Expressions
• Also to detect relations and fill events
• Higher level regular expressions make use of “objects” detected by lower level patterns
• Some NLP information may help (pos tags, phrases, semantic word categories) o Crime-Victim can use things matched by “noun-group”
• Prefiller: [pos: V, type-of-verb: KILL] WordNet MCR • Filler: [phrase: NOUN-GROUP]
![Page 38: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/38.jpg)
Using Regular Expressions
• Extraction relations between entities o Which PERSON holds what POSITION in what ORGANIZATION
• [PER], [POSITION] of [ORG]
Entities: PER: Jose Mourinho POSITION: trainer ORG: Chelsea
Relation
Jose Mourinho Trainer
Chelsea
![Page 39: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/39.jpg)
Using Regular Expressions
• Extraction relations between entities o Which PERSON holds what POSITION in what ORGANIZATION
• [PER], [POSITION ] of [ORG] • [ORG] (named, appointed,…) [PER] Prep [POSITION]
o Nokia has appointed Rajeev Suri as President
o Where a ORGANIZATION is located
• [ORG] headquarters in [LOC] o NATO headquarters in Brussels
• [ORG][LOC] (division, branch, headquarters…) o KFOR Kosovo headquarters
![Page 40: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/40.jpg)
Extracting relations with palerns
• Hearst 1992
• What does Gelidium mean?
• “Αγαρ ισ α συβστανχε πρεπαρεδ φροµ α µιξτυρε οφ ρεδ αλγαε, συχη ασ
Gelidium, φορ λαβορατορψ ορ ινδυστριαλ υσε”
![Page 41: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/41.jpg)
Extracting relations with palerns
• Hearst 1992
• What does Gelidium mean?
• “Agar is a substance prepared from a mixture of red algae, such as Gelidium, for laboratory or industrial use”
• How do you know?
![Page 42: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/42.jpg)
Extracting relations with palerns
• Hearst 1992: Automatic Acquisition of Hyponyms (IS-A)
X à Gelidium (sub-type) Y à red algae (super-type) X à IS-A à Y
• “Y such as X” • “Y, such as X” • “X or other Y” • “X and other Y” • “Y including X” • ….
![Page 43: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/43.jpg)
Extracting relations with palerns
![Page 44: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/44.jpg)
Hand-‐‑built palerns • Positive
o Tend to be high-precision o Can be adapted to specific domains
• Negative o Human patterns are usually low-recall o A lot of work to think all possible patterns o Need to create a lot of patterns for every relation
![Page 45: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/45.jpg)
Learning-‐‑based Approaches
• Statistical techniques and machine learning algorithms o Automatically learn patterns and models for new domains
• Some types o Supervised learning of patterns and rules o Supervised Learning for relation extraction o Supervised learning of Sequential Classifier Methods o Weakly supervised and supervised
![Page 46: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/46.jpg)
Supervised Learning of Palerns and Rules
• Aiming to reduce the knowledge engineering bottleneck to create an IE in a new domain
• AutoSlog and PALKA à first IE pattern learning systems o AutoSlog: syntactic templates, lexico-syntactic patterns and manual
review
• Learning Algorithms à generate rules from annotated text o LIEP (Huffman 1996) : syntactic paths, role fillers. Patterns that work ok in
training are kept o (LP)2 uses tagging rules and correction rules
![Page 47: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/47.jpg)
Supervised Learning of Palerns and Rules
• Relational learning methods o RAPIER: rules for pre-filler, filler, and post-filler component. Each
component is a pattern that consists of words, POS tags, and semantic classes.
![Page 48: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/48.jpg)
Supervised Learning for relation extraction (I)
• Design a supervised machine learning framework
• Decide what relations we are interested in
• Choose what entities are relevant
• Find (or create) labeled data o Representative corpus o Label the entities in the corpus (Automatic NER) o Hand label relation between these entities o Split into train + dev + test
• Train, improve and evaluate
![Page 49: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/49.jpg)
Supervised Learning for relation extraction (II)
• Relation extraction as a classification problem • 2 classifiers
o To decide if two entities are related o To decide the class for a pair or related entities
• Why 2? o Faster training by eliminating most pairs
o Appropriate feature sets for each task
• Find all pairs of NE (restricted to the sentence) o For every pair
1. Are the entities related (classifier 1) 1. no à END 2. Yes à guess the class (classifier 2)
![Page 50: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/50.jpg)
Supervised Learning for relation extraction (III)
• Are the two entities related? • What is the type of relation?
![Page 51: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/51.jpg)
Supervised Learning for relation extraction (IV)
“[American Airlines], a unit of AMR, immediately matched the move, spokesman [Tim Wagner] said” • What features?
o Head words of entity mentions and combination • Airlines Wagner Airlines-Wagner
o Bag-of-words in the two entity mentions • American, Airlines, Tim, Wagner, American Airlines, Tim Wagner
o Words/bigrams in particular positions to the left and right • M2#-1: spokesman M1#+1: said
o Bag-of-words (or bigrams) between the 2 mentions • a, AMR, of, immediately, matched, move, spokesman, the, unit
![Page 52: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/52.jpg)
Supervised Learning for relation extraction (V)
“[American Airlines], a unit of AMR, immediately matched the move, spokesman [Tim Wagner] said” • What features?
o Named entity types • M1: ORG M2: PERSON
o Entity level (Name, Nominal (NP), Pronoun) • M1: NAME (“it” or “he” would be PRONOUN) • M2: NAME (“the company” would be NOMINAL)
o Basic chunk sequence from one entity to the other • NP NP PP VP NP NP
o Constituency path on the parse tree • NP é NP é S é S ê NP
![Page 53: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/53.jpg)
Supervised Learning for relation extraction (VI)
“[American Airlines], a unit of AMR, immediately matched the move, spokesman [Tim Wagner] said” • What features?
• Trigger lists o For family à parent, wife, husband… (WordNet)
• Gazetteers o List of countries…
• …. • …. • …
![Page 54: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/54.jpg)
Supervised Learning for relation extraction (VII)
• Decide your algorithm o MaxEnt, Naïve Bayes, SVM
• Train the system on the training data
• Tune it on the dev set
• Test on the evaluation test o Traditional Precision, Recall and F-score
![Page 55: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/55.jpg)
Sequential Classifier Methods
• IE as a classification problem using sequential learning models.
• A classifier is induced from annotated data to sequentially scan a text from left to right and decide what piece of text must be extracted or not
• Decide what you want to extract
• Represent the annotated data in a proper way
![Page 56: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/56.jpg)
Sequential Classifier Methods
![Page 57: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/57.jpg)
Sequential Classifier Methods
• Typical steps for training o Get the annotated training data o Represent the data in IOB o Design feature extractors o Decide the algorithm to use o Train the models
• Testing steps o Get the test documents o Extract features o Run the sequence models o Extract the recognized entities
![Page 58: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/58.jpg)
Sequential Classifier Methods
• Algorithms o HMM o CMM o CRF
• Features o Words (current, previous, next) o Other linguistic information (PoS, chunks…) o Task specific features (NER…)
• Word shapes: abstract representation for words
![Page 59: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/59.jpg)
Sequential Classifier Methods
• Algorithms o HMM o SVM o CRF
• Features o Words (current, previous, next) o Other linguistic information (PoS, chunks…) o Task specific features (NER…)
• Word shapes: abstract representation for words
![Page 60: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/60.jpg)
Weakly supervised and unsupervised
• Manual annotation is also “expensive” o IE is quite domain specific à not reuse
• AutoSlog-Ts: o Just needs 2 sets of documents: relevant/irrelevant o Syntactic templates + relevance according to relevant set
• Ex-Disco (Yangarber et al. 2000) o No need preclassified corpus o They use a small set of patterns to decide relevant/irrelevant
![Page 61: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/61.jpg)
Weakly supervised and unsupervised
• OpeNER: • European project dealing with entity recognition,
sentiment analysis and opinion mining mainly in hotel reviews (also restaurants, attractions, news)
• Double propagation o Method to automatically gather opinion words and targets
• From a large raw hotel corpus • Providing a set of seeds and patterns
![Page 62: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/62.jpg)
Weakly supervised and unsupervised
• Seed list • + à good, nice • - à bad, ugly
• Patterns • a [EXP] [TAR] • the [EXP] [TAR]
• Polarity patterns • = [EXP] and [EXP] [EXP], [EXP] • ! [EXP] but [EXP]
![Page 63: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/63.jpg)
Weakly supervised and unsupervised
• Propagation method o 1) Get new targets using the seed expressions and the
patterns • a nice [TAR] a bad [TAR] the ugly [TAR] • Output à new targets (hotel, room, location)
o 2) Get new expression using the previous targets and the patterns • a [EXP] hotel the [EXP] location • Output à new expressions (expensive, cozy, perfect…)
o Keep running 1 and 2 to get new EXP and TAR
![Page 64: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/64.jpg)
Weakly supervised and unsupervised
• Polarity guessing o Apply the polarity patters to guess the polarity
• = a nice(+) and cozy(?) à cozy(+) • ! Clean(+) but expensive(?) à expensive (-)
hlps://github.com/opener-‐‑project/opinion-‐‑domain-‐‑lexicon-‐‑acquisition
![Page 65: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/65.jpg)
Outline 1. What is Information Extraction 2. Main goals of Information Extraction 3. Information Extraction Tasks and Subtasks 4. MUC conferences 5. Main domains of Information Extraction 6. Methods for Information Extraction
o Cascaded finite-state transducers o Regular expressions and patterns o Supervised learning approaches o Weakly supervised and unsupervised approaches
7. How far we are with IE
![Page 66: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/66.jpg)
How good is IE
![Page 67: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/67.jpg)
How good is IE • Some progress has been done • Still the barrier of 60% seems difficult to outperform • Most errors on entities and event coreference • Propagation errors
o Entity recognition à 90% o One event -> 4 entities o 0.9 x 4 à 60%
• A lot of knowledge is implicit or “common world knowledge”
![Page 68: Information Extraction](https://reader033.vdocument.in/reader033/viewer/2022042716/55a525971a28abe50e8b46b8/html5/thumbnails/68.jpg)
How good is IE Information Type Accuracy Entities 90 – 98% Alributes 80% Relations 60 – 70% Events 50 – 60%
• Very optimistic numbers for well-established tasks • The numbers go down for specific/new tasks