TopicModelDiagnostics.TopicScores |
TopicModelDiagnostics.getCoherence() |
|
TopicModelDiagnostics.TopicScores |
TopicModelDiagnostics.getDistanceFromCorpus() |
Low-quality topics may be very similar to the global distribution.
|
TopicModelDiagnostics.TopicScores |
TopicModelDiagnostics.getDistanceFromUniform() |
|
TopicModelDiagnostics.TopicScores |
TopicModelDiagnostics.getDocumentEntropy(int[] tokensPerTopic) |
|
TopicModelDiagnostics.TopicScores |
TopicModelDiagnostics.getDocumentPercent(int i) |
|
TopicModelDiagnostics.TopicScores |
TopicModelDiagnostics.getDocumentPercentRatio(int numeratorIndex,
int denominatorIndex) |
|
TopicModelDiagnostics.TopicScores |
TopicModelDiagnostics.getEffectiveNumberOfWords() |
|
TopicModelDiagnostics.TopicScores |
TopicModelDiagnostics.getExclusivity() |
Low-quality topics may have words that are also prominent in other topics.
|
TopicModelDiagnostics.TopicScores |
TopicModelDiagnostics.getRank1Percent() |
|
TopicModelDiagnostics.TopicScores |
TopicModelDiagnostics.getTokenDocumentDiscrepancies() |
|
TopicModelDiagnostics.TopicScores |
TopicModelDiagnostics.getTokensPerTopic(int[] tokensPerTopic) |
|
TopicModelDiagnostics.TopicScores |
TopicModelDiagnostics.getWordLengthScores() |
Low-quality topics often have lots of unusually short words.
|
TopicModelDiagnostics.TopicScores |
TopicModelDiagnostics.getWordLengthStandardDeviation() |
Low-quality topics often have lots of unusually short words.
|