Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

antonio-lingmotif - Sentiment Analysis for the Masses

42 views

Published on

Sentiment Analysis for the Masses

Published in: Technology
  • Be the first to comment

  • Be the first to like this

antonio-lingmotif - Sentiment Analysis for the Masses

  1. 1. @DataBeersMLG 15-Sept-2016 Antonio Moreno-Ortiz Sentiment Analysis for the Masses
  2. 2. Sentiment Analysis in a nutshell WHAT A Natural Language Processing task. Also called opinion mining. Is this text positive or negative? WHY Because companies, organizations, and governments are interested in what people think about a particular product, service, topic... HOW By analyzing user-generated text, especially on social networking sites.
  3. 3. I’m interested. What’s available? Most available software is in the form of: •NLP software toolkits • NLTK, Freeling, Stanford CoreNLP... •Web services (via an API) • IBM Watson, Microsoft Azure, TweetSentiment…
  4. 4. But I’m not a programmer!!! We got you covered! 😏 LINGMOTIF A multilingual, multiplatform desktop application Developed by the Tecnolengua Group @ UMA http://tecnolengua.uma.es
  5. 5. How does it work? • Identifies sentiment- related words and expressions • Large lexical database • Derived from multiple sources, including large multi-million-word corpora • Takes context into account English Spanish Single Words 28,000 201,000 Multiword Expressions 38,500 130,000 Context Rules 750 700
  6. 6. What do I get?
  7. 7. Loads of data
  8. 8. Classification
  9. 9. Sentiment Profiles
  10. 10. Detailed Sentiment Analysis
  11. 11. So what you up to now? • Currently working on version 2… • More accurate lexical data for English and Spanish • Include support for French and Italian (maybe Catalan too!) • More user-friendly functionalities • Sentiment Score based on Machine Learning algorithms
  12. 12. Shut up and take my money! http://tecnolengua.uma.es
  13. 13. Thanks!! Antonio Moreno-Ortiz @tecnolengua amo@uma.es

×