The function outputs a message and a table. The schema of the table depends on the values of the arguments SummaryOutput and OutputByWord.
Output Message Schema
| Column |
Data Type |
Description |
| message |
VARCHAR |
Reports this model information:
- Number of topics
- Number of unique words (vocabulary size)
- Hyperparameter values (specified by LDA arguments Alpha and Eta)
- Number of training documents and words
- Perplexity (defined in LDA Output)
|
Output Table, SummaryOutput ('false') and OutputByWord ('true')
| Column |
Data Type |
Description |
| topicid |
INTEGER |
Topic identifier. |
| word |
VARCHAR |
Unique word that appears in topic. |
| wordweight |
DOUBLE PRECISION |
[Column appears only with WordWeight argument.] Weight (probability of occurrence) of word in topic. Weights for unique words in each topic are normalized to 1. |
| wordcount |
DOUBLE PRECISION |
[Column appears only with WordCount argument.] Count (number of occurrences) of word in topic. Topic distribution is factored into word count. |
Output Table, SummaryOutput ('false') and OutputByWord ('false')
| Column |
Data Type |
Description |
| topicid |
INTEGER |
Topic identifier. |
| wordsequence |
VARCHAR |
Unique words that appear in topic, separated by commas. |