SAS® Visual Text Analytics in SAS® Viya® Flashcards
What is Natural language?
verbal language spoken by human beings
What is Formal language?
languages for any purpose other than verbal communication such as computer languages
What is Natural Language Processing (NLP)?
a sub-field of Artificial Intelligence that uses computers to process natural language for specific purposes
What is the primary role of NLP in SAS Visual Text Analytics?
to efficiently parse the document collection
What is a major challenge to NLP?
word sense disambiguation, i.e. when context is necessary to identify whether or not a word is a noun, verb, or adjective
What is cognitive bias?
what you think you know, when what you think you know might be different from reality
What is SAS Visual Text Analytics?
a modern, flexible, and end-to-end text analytics framework that combines text mining, contextual extraction, categorization, sentiment analysis, and search
What is contextual extraction?
Enables non-ambiguous coding of subject matter expertise to extract specific information from within documents
What is LITI?
Language Interpretation Text Interpretation
What do the acronyms CAS and CASL represent?
Cloud Analytics Services Language
What are the benefits of NLP?
eliminates manual reviews/bias, regulates classifications, and removes requirement of training corpus
Why are iterative search algorithms used for solving non-linear programming (NLP) problems?
A unique, optimal solution is not guaranteed.
Name 5 elements that separate deep learning from traditional neural networks
- hidden unit saturation activation functions that are more resistant to saturation than conventional activation functions
- fast-moving gradient-based optimizations
- weight initializations that consider the amount of incoming information
- new regularization techniques
- innovations in distributed processing
When would you recommend a convolutional neural network (CNN) model?
advanced neural networks for traditional or computer vision problems
When would you recommend a recurrent neural network (RNN) model?
sequential data problems such as time series or NLP
