On the Advanced tab, you can configure language detection settings, enable spell checking, and control how the text is categorized based on various criteria.
Language settings
Use these settings to control and modify the language detection features. You can configure the system to:
Sentiment settings
You can control the score range of the positive, negative, and neutral sentiment to make the sentiment analysis suit your business requirements.
Check spelling
By enabling the spell checking option, you can eliminate spelling mistakes from the analyzed text. Without spelling mistakes, the system can assign the analyzed text to a specific category more accurately and with a higher confidence score.
The Pega Platform uses a special spell checker decision data rule for checking spelling. A single spell checker rule can contain multiple entries. Each entry in the spell checker rule has a Spell Checker Property containing a language dictionary (in the form of a .CLX extension dictionary file). The Pega Platform comes with the default pySpellChecker decision data rule containing English (U.K. English and U.S. English), French, German, and Spanish dictionaries.
Classification settings
In the case of the model-based classification, the system identifies categories for a specific sample of text. You can use this option to segregate text samples according to the categories.
When you analyze larger amounts of text (for example, lengthy Facebook comments or emails), the text can have many associated categories. In such situation, you can take the following actions: