Megaputer Intelligence Weekly Meeting 4/8/2016

Working with the New PolyAnalyst Dictionaries -- Ankit

Morphology, Spellcheck and Phrase dictionaries will be covered.

14 types of dics in PA:

Entity extraction and sentiment analysis: Companies, Geo, Human, Org, WordClasses

Keyword extraction: Statistics

And one more category:

Support 14 languages currently, with multiple dics in each. Language detection node is helpful here.

Index works on morphology dic, after that can add nodes such as SpellCheck, KeywordExtraction, SentimentA, etc.

Default morphology dic: 86289 words, including words that belong to different parts of speech.

Different attributes of verb:

Example: “The patient receives 100mg of medicine X each day. Last week she received 80mg per day due to reduced tolerance.”

Query: lemma(verb_personthird_singular_present,“receive”)

Can do keyword extraction, then generate a dataset, then import this dataset as a new dictionary.

New task: decide order of nodes, entity extraction before keyword extraction, name of a person?

xEntity extraction. Link analysis can be directional.

Reach out for webinars: text analytics and pharmaceuticals. Seminar for pharma companies in Bloomington, how to use text analytics in pharma.

PA User Conference 2016: Oct.

Raffles: number of bugs reported each month

cgss15/students/jiayil/polyanalyst/weeklymeetings/4_8_2016.txt · Last modified: 2016/04/08 19:26 by jiayil
Except where otherwise noted, content on this wiki is licensed under the following license: CC Attribution-Share Alike 4.0 International
Recent changes RSS feed Donate Powered by PHP Valid XHTML 1.0 Valid CSS Driven by DokuWiki