This example uses a dictionary table, sentiment_word, instead of a model file. See sentiment_word in SentimentExtractor Example 5: Dictionary Table Instead of Model File.
Input
- Input table: sentiment_extract_input, as in SentimentExtractor Example 1: ModelFile ('dictionary'), AnalysisType ('document')
SQL Call
SELECT * FROM SentimentEvaluator (
ON SentimentExtractor (
ON sentiment_extract_input PARTITION BY ANY
ON sentiment_word AS dict DIMENSION
USING
TextColumn ('review')
Accumulate ('category')
) PARTITION BY 1
USING
ObservationColumn ('category')
SentimentColumn ('out_polarity')
) AS dt;
Output
| evaluation_result |
|---|
positive record (total relevant, relevant, total retrieved): 5 5 6 recall and precision: 1.00 0.83 negative record (total relevant, relevant, total retrieved): 5 3 3 recall and precision: 0.60 1.00 positive and negative record (total relevant, relevant, total retrieved): 10 8 9 recall and precision: 0.80 0.89 |