Class | Description |
---|---|
AnnotatedTextReader |
CanNOT handle overlapping labeled text (that is one token cannot belong to
multiple labels)! Note that there has to be spaces around the tags for the reader to work correctly!
|
ApplyPatterns | |
ConstantsAndVariables | |
ConstantsAndVariables.DataSentsIterator | |
CreatePatterns | |
Data | |
EditDistanceDamerauLevenshteinLike |
COPIED FROM https://gist.github.com/steveash (public domain license)
Implementation of the OSA (optimal string alignment) which is similar
to the Damerau-Levenshtein in that it allows for transpositions to
count as a single edit distance, but is not a true metric and can
over-estimate the cost because it disallows substrings to edited more than
once.
|
GetPatternsFromDataMultiClass |
Given text and a seed list, this class gives more words like the seed words
by learning surface word patterns.
|
GetPatternsFromDataMultiClass.LabelWithSeedWords | |
InvertedIndexByTokens |
Creates an inverted index of (classkey:value) => {sentid1,sentid2,..
|
LearnImportantFeatures |
The idea is that you can learn features that are important using ML algorithm
and use those features in learning weights for patterns.
|
PatternsAnnotations | |
PatternsAnnotations.Features | |
PatternsAnnotations.LongestMatchedPhraseForEachLabel |
For each label, what was the longest phrase that matched.
|
PatternsAnnotations.MatchedPattern | |
PatternsAnnotations.MatchedPatterns | |
PatternsAnnotations.MatchedPhrases |
All matched phrases - can be from multiple labels
|
PatternsAnnotations.OtherSemanticLabel | |
PatternsAnnotations.PatternLabel1 | |
PatternsAnnotations.PatternLabel10 | |
PatternsAnnotations.PatternLabel2 | |
PatternsAnnotations.PatternLabel3 | |
PatternsAnnotations.PatternLabel4 | |
PatternsAnnotations.PatternLabel5 | |
PatternsAnnotations.PatternLabel6 | |
PatternsAnnotations.PatternLabel7 | |
PatternsAnnotations.PatternLabel8 | |
PatternsAnnotations.PatternLabel9 | |
PatternsAnnotations.ProcessedTextAnnotation | |
PatternsAnnotations.SeedLabeledOrNot | |
PatternsForEachToken |
Created by Sonal Gupta on 10/8/14.
|
PatternsForEachTokenDB |
Created by sonalg on 10/22/14.
|
PatternsForEachTokenInMemory |
Created by sonalg on 10/22/14.
|
PatternToken |
Class to represent a target phrase.
|
PhraseScorer | |
ScorePatterns | |
ScorePatternsF1 |
Used if patternScoring flag is set to F1 with the seed pattern.
|
ScorePatternsFreqBased | |
ScorePatternsRatioModifiedFreq | |
ScorePhrases | |
ScorePhrasesAverageFeatures |
Score phrases by averaging scores of individual features.
|
SentenceIndex |
Created by sonalg on 10/15/14.
|
SQLConnection |
Created by sonalg on 10/8/14.
|
SurfacePattern |
To represent a surface pattern in more detail than TokenSequencePattern (this
class object is eventually compiled as TokenSequencePattern via the toString
method).
|
Token |
Currently can handle only ORs
Created by sonalg on 10/16/14.
|
Enum | Description |
---|---|
ConstantsAndVariables.PatternForEachTokenWay | |
ConstantsAndVariables.ScorePhraseMeasures | |
GetPatternsFromDataMultiClass.PatternScoring |
RlogF is from Riloff 1996, when R's denominator is (pos+neg+unlabeled)
|
PhraseScorer.Normalization | |
SurfacePattern.Genre |