3.2 Evaluating Temporal Taggers
3.3 Research Competitions
3.4 Annotated News-style Corpora
3.5 Summary of the Chapter
4 Domain-sensitive Temporal Tagging
4.1 Temporal Tagging of News-style Documents
4.2 The Concept of a Domain
4.3 Annotated Non-news-style Corpora
4.4 Characteristics of Different Domains
4.4.1 News-style Documents
4.4.2 Narrative-style Documents
4.4.3 Colloquial-style Documents
4.4.4 Autonomic-style Documents
4.4.5 Further Domains and Summary
4.5 Comparative Corpus Analysis—Challenges for Domain-sensitive Temporal Tagging
4.6 Strategies for Domain-sensitive Temporal Tagging
4.7 Summary of the Chapter
5.1 Overview of Approaches to Temporal Tagging
5.2 Overview of Existing Temporal Taggers
5.2.1 TIPSem—Exploiting Semantic Information
5.2.2 HeidelTime—Rule-based, Multilingual, Domain-sensitive
5.2.3 SUTime—Rule-based, Stanford CoreNLP Component
5.2.4 UWTime—Semantic Parsing of Time Expressions
5.2.5 Comparison and Evaluation Results
5.3 The Value of Domain-sensitive Temporal Tagging
5.4 Approaches to Subtasks and Related Tasks
5.5 Highly Multilingual Temporal Tagging
5.6 Summary of the Chapter
6 Summary and Future Research Directions
6.2 Future Directions
List of Figures
1.1 The two tasks of temporal tagging: extraction and normalization
1.3 Excerpts of the CNNMoney article of Figure 1.1
1.4 Temporal information retrieval example
2.1 Temporal information is well-defined
2.2 Temporal information can be normalized
2.3 Temporal information can be organized hierarchically
2.4 Different realization types of date expressions in documents
4.1 Example of a news document with arrows indicting required information to interpret the temporal expressions
4.2 Excerpts of documents of the news domain
4.3 Tense information may be misleading to normalize underspecified expressions
4.4 Excerpts of a document of the narrative domain
4.5 Excerpts of a French narrative document of the AncientTimes corpus (i.e., a document about history)
4.6 Excerpts of a narrative document of the English AncientTimes corpus
4.7 Excerpts of colloquial-style documents of the Time4SMS corpus containing short messages with relative and underspecified expressions
4.8 Excerpts of an autonomic-style document of the Time4SCI corpus
4.9 An autonomic timeline for the document shown in Figure 4.8
4.10 Excerpts of autonomic- and narrative-style literary texts
4.11 Distribution of date, time, duration, and set expressions in the corpora of the four different domains
4.12 Distribution of the occurrence types of date and time expressions in the four corpora
4.13 Domain-dependent strategies to detect the reference times for relative and underspecified expressions
4.14 Domain-dependent strategies to identify the temporal relation between an underspecified expression and its reference time
List of Tables
2.1 The four categories how temporal expressions can be realized
