1、JavaNLP time annotations,SUTime,What does SUTime do?,Similar to GUTime Recognizes time expressions using patterns Deterministic, based on regular expression patterns Greedy (picks longest sequence of tokens that may represent a time expression) Normalizes time expressions Annotations follow TimeML T
2、IMEX3 standard http:/www.timeml.org/site/publications/timeMLdocs/timeml_1.2.1.html#timex3 XSD: http:/www.timeml.org/timeMLdocs/TimeML.xsd Extensions for time expressions that are not supported by TIMEX3 standard Resolves relative times with respect to reference date,SUTime Time Representation,Main T
3、emporal types Time A instance in time (2011-08-11), can be partially specified (Friday), with limited granularity Duration - A length of time (3 days) Range Time interval with start and end points Set A set of temporals Periodic sets: Every Friday,SUTime Representation,Time Standard date and times (
4、in years, months, days, day of week, hours, minutes, seconds, milliseconds) Common times: Seasons (e.g. winter), Time of day (e.g. morning), Weekend Partial Times (June = XXXX-06) Relative Time (last week) Duration Exact durations (specified in milliseconds or in fields) Inexact durations (a few yea
5、rs = PXY) Duration ranges (2 to 3 months = P2M/P3M),SUTime Limitations,Holidays are not supported Support for ranges is poor from 3 to 4 p.m is identified as 15:57:00 12-13 March 2011 (12-13 is ignored) Resolving relative expressions with respect to the given reference date can be problematic Handli
6、ng of ambiguous phrases is poor Some common words (e.g. spring/fall) are always identified as a temporal expression Patterns are language (English) specific ,SUTime Usage,TimeAnnotator TimeAnnotator timeAnnotator = new TimeAnnotator(“sutime”, properties); Properties: Specifies SUTime options (prefix
7、ed by “sutime.”) Pipeline TimeAnnotator should come after the tokenizer, sentence splitter, and pos tagger Optional (also before): NER or NumberAnnotator/QuantifiableEntityNormalizingAnnotator,SUTime Options,SUTime input annotations,DocDateAnnotation (String) If present, then the string is interpret
8、ed as a date/time and used as the reference document date with respect to which other temporal expressions are resolved SentencesAnnotation (List) If present, time expressions will be extracted from each sentence and each sentence will be annotated individually. TokensAnnotations (List) Required eit
9、her at the entire annotation level or per sentence level.,SUTime output annotations,Timex.Annotations (List) List of time expressions (each a CoreMap) On the entire annotation and also for each sentence Time annotations (for each time expression/CoreMap),SUTime output annotations,Standard annotation
10、s (for each time expression),Note: Indices are 0-based, and always relative to the original annotation. Begin indices are inclusive, end indices are exclusive.,Comparison with GUTime,SUTime and GUTime examples,Examples (GUTime unsupported),Reference Date is 2011-08-01,Examples (SUTime unsupported),Reference Date is 2011-08-01,
