How to create models to use them in Text Analyzer ?
How to create Taxonomy model for Text Analyzer ? Do we have to enter hundreds of records in excel files manually ?
***Edited by Moderator Marissa to update platform capability tags; update SR details****
Pegasystems Inc.
US
If you are running Pega 8.1x (Pega Infinity), there is more efficient way of building Taxonomy models: see the link: https://community.pega.com/knowledgebase/articles/create-taxonomies-more-efficiently
Perficient
US
I have following questions about Text Analyzer.
1) How does training and test data affect the model.
2) How to update the model.
3) How to see contents of JAR files of already built model.
Pegasystems Inc.
NL
1) The training data is used for extracting features and building a machine learning model. It represents the vocabulary which the model learns from. The test data is used as unseen data for evaluating the learning of the model.
2) The model can be updated by adding feedback data using an activity.
3) The jar consists of binary files, properties files and encoders. The binary file consists weights, vocabulary and feature set. It is not human readable.