Text analytics is a genre of analytic capabilities intended to function across the typed/written word. This area of analytics seeks to learn from huge quantities of text data to expose human intent, sentiment, and behaviors. Examples include doctors/nurses notes, call center notes, tweets, product/content reviews, survey text response, and much more. There are varying types of text analytics such as text parsing, Levenshtein distance, entity extraction, tagging/classification, and chunking just to name a few. Many areas of machine learning use text analytics as data preparation steps in order to develop models. Text analytics is a very fascinating part of the big data eco-system and is vital in understanding behavior analytics. Teradata Aster offers wide variety of text analytics capability such as sentiment analysis, TF_IDF and Levenshtein Distance.
Interview with Mark Turner, Principal Data Scientist, Teradata Aster
“In user organizations, the basics are a big help: having good data dictionaries and other documentation on the data makes analytics easier, more productive, and more reliable for user organizations,” says Mark Turner.
Expertise of Mark Turner: Conducting analytics on a wide range of large data sets, both structured and unstructured (text), in major firms across multiple industries. Applications include predictive analytics, identity matching, concept association, sentiment analysis, document clustering, and others. Interview with Mark Turner
Blog: New Aster Text Capabilities
By Omri Shriv Our world is changing, but our reliance on documents seems to only be increasing. While the medium may have changed, as a technology, the document has only created problems. Read blog
ExtractSentiment is a map function that extracts the opinion or sentiment from input text. Much of user generated content includes the author’s feelings and opinions (happy, angry etc.). This function helps to extract the polarity of the content as positive, negative or neutral.
TextChunker divides text into phrases in such a way that syntactically-related words become members of the same phrase. The TextChunker function divides text into phrases and assigns each phrases a tag identifying its type.
The Aster Community is now available for all users to learn, discover and share their Aster experience. The portal features a public area where anyone can learn about Aster and download Aster Express. It also features a private Aster Customer Zone where customers can discuss, ask questions and share insights with peers and Aster experts.