Lecture 13 Information Extraction CSCE 771 Natural Language Processing Lecture 13 Information Extraction Topics Name Entity Recognition Relation detection Temporal and Event Processing Template Filling Readings: Chapter 22 February 27, 2013

Overview Last Time Today Readings Dialogues Human conversations Slides from Lecture24 Dialogue systems Dialogue Manager Design Finite State, Frame-based, Initiative: User, System, Mixed VoiceXML Information Extraction Readings Chapter 24, Chapter 22

Information extraction Information extraction – turns unstructured information buried in texts into structured data Extract proper nouns – “named entity recognition” Reference resolution – \ named entity mentions Pronoun references Relation Detection and classification Event detection and classification Temporal analysis Template filling

Template Filling Example template for “airfare raise”

Figure 22.1 List of Named Entity Types

Figure 22.2 Examples of Named Entity Types

Figure 22.3 Categorical Ambiguities

Figure 22.4 Categorical Ambiguity

Figure 22.5 Chunk Parser for Named Entities

Figure 22.6 Features used in Training NER Gazetteers – lists of place names www.geonames.com www.census.gov

Figure 22.7 Selected Shape Features

Figure 22.8 Feature encoding for NER

Figure 22.9 NER as sequence labeling

Figure 22.10 Statistical Seq. Labeling

Evaluation of Named Entity Rec. Sys. Recall terms from Information retreival Recall = #correctly labeled / total # that should be labeled Precision = # correctly labeled / total # labeled F- measure where β weights preferences β=1 balanced β>1 favors recall β<1 favors precision

NER Performance revisited Recall, Precision, F High performance systems F ~ .92 for PERSONS and LOCATIONS and ~.84 for ORG Practical NER Make several passes on text Start by using highest precision rules (maybe at expense of recall) make sure what you get is right Search for substring matches or previously detected names using probabilistic searches string matching metrics(Chap 19) Name lists focused on domain Probabilistic sequence labeling techniques using previous tags

Relation Detection and classification Consider Sample text: Citing high fuel prices, [ORG United Airlines] said [TIME Friday] it has increased fares by [MONEY $6] per round trip on flights to some cities also served by lower-cost carriers. [ORG American Airlines], a unit of [ORG AMR Corp.], immediately matched the move, spokesman [PERSON Tim Wagner] said. [ORG United Airlines] an unit of [ORG UAL Corp.], said the increase took effect [TIME Thursday] and applies to most routes where it competes against discount carriers, such as [LOC Chicago] to [LOC Dallas] and [LOC Denver] to [LOC San Francisco]. After identifying named entities what else can we extract? Relations

Fig 22.11 Example semantic relations

Figure 22.12 Example Extraction

Figure 22.13 Supervised Learning Approaches to Relation Analysis Algorithm two step process Identify whether pair of named entities are related Classifier is trained to label relations

Factors used in Classifying Features of the named entities Named entity types of the two arguments Concatenation of the two entity types Headwords of the arguments Bag-of-words from each of the arguments Words in text Bag-of-words and Bag-of-digrams Stemmed versions Distance between named entities (words / named entities) Syntactic structure Parse related structures

Figure 22.14 a-part-of relation

Figure 22.15 Sample features Extracted

Bootstrapping Example “Has a hub at” Consider the pattern / * has a hub at * / Google search 22.4 Milwaukee-based Midwest has a hub at KCI 22.5 Delta has a hub at LaGuardia … Two ways to fail False positive: e.g. a star topology has a hub at its center False negative? Just miss 22.11 No frill rival easyJet, which has established a hub at Liverpool

Figure 22.16 Bootstrapping Relation Extraction

Using Features to restrict patterns 22.13 Budget airline Ryanair, which uses Charleroi as a hub, scrapped all weekend flights / [ORG] , which uses a hub at [LOC] /

Semantic Drift Note it will be difficult (impossible) to get annotated materials for training Accuracy of process is heavily dependant on initial sees Semantic Drift – Occurs when erroneous patterns(seeds) leads to the introduction of erroneous tuples

Fig 22.17 Temporal and Durational Expressions Absolute temporal expressions Relative temporal expressions

Fig 22.18 Temporal lexical triggers

Fig 22.19 MITRE’s tempEx tagger-perl

Fig 22.20 Features used to train IOB

Figure 22.21 TimeML temporal markup

Temporal Normalization iSO 8601 - standard for encoding temporal values YYYY-MM-DD

Figure 22.22 Sample ISO Patterns

Event Detection and Analysis Event Detection and classification

Fig 22.23 Features for Event Detection Features used in rule-based and statistical techniques

Fig 22.24 Allen’s 13 temporal Relations

Figure 22.24 continued

Figure 22.25 Example from Timebank Corpus

Template Filling

Figure 22.26 Templates produced by Faustus 1997

Figure 22.27 Levels of processing in Faustus

Figure 22.28 Faustus Stage 2

Figure 22.29 The 5 Partial Templates of Faustus

Figure 22.30 Articles in PubMed

Figure 22.31 biomedical classes of named entities