This course shows you how to build and test models to assess sentiment in electronic text using SAS Sentiment Analysis Studio.
Learn how to
- build simple and advanced statistical models
- write rules and construct rule-based models
- combine statistical and rule-based models in a hybrid model
- upload models to SAS Sentiment Analysis Server.
Who should attend
System designers, developers, analytical consultants, and linguists who want to understand techniques and approaches for identifying sentiment in textual documents
No SAS or programming experience is required. You should be able to log on and off a computer, use a keyboard and mouse, and have a preliminary understanding of the differences between structured (numeric) and unstructured (text) data fields.
This course addresses SAS Sentiment Analysis, SAS Text Analytics software.
Introduction to SAS Sentiment Analysis
Defining the Corpus and Statistical Modeling
- workflow and architecture
Rule-Based and Hybrid Models
- defining a corpus
- building a statistical model
- testing a statistical model
Working with Sentiment Analysis Server
- introduction to rule-based models
- importing learned features and writing CLASSIFIER rules
- building and testing a rule-based model
- writing CONCEPT and REGEX rules
- writing C_CONCEPT, CONCEPT_RULE, and PREDICATE_RULE rules
- building a hybrid model
Sentiment Project Case Study
- uploading models
- configuring Sentiment Analysis Server (self-study)
- case study
- taxonomy structure