Language Processing at the Core of the Media & Publishing Industries - Daedalus Perspective

743 views

Published on

Presentation delivered by Daedalus at LT-Innovate meeting in Berlin, April 12, 2013

Published in: Technology, Business

Language Processing at the Core of the Media & Publishing Industries - Daedalus Perspective

  1. 1. Language Processingat the core of the Media & Publishing Industries Berlin, April 12th 2013
  2. 2. Language Processing at the core of the Media & Publishing IndustriesMultifacet crisis in the Media & Publishing Industries User generated content Decline in Shift in ways advertising audiences revenues consume contents Business models
  3. 3. Language Processing at the core of the Media & Publishing IndustriesAn old concern: language quality Proofreading with Stilus: Spell, grammar and style checking
  4. 4. Language Processing at the core of the Media & Publishing Industries Daedalus: extracting meaning from multilingual & multimedia contents Semantic Processing: automatic extraction of knowledge items from non-structured content Facts Topics Sentiment Semantic Processing PeopleAnnotation, Enrichment & LinkingNamed entities and concepts extraction,classification, clustering Organizations ConceptsAreas: documentation and advancedcontent search, SEO positioning
  5. 5. Language Processing at the core of the Media & Publishing IndustriesAdvanced Semantic Analysis at Daedalus  People: Ben Bernanke, Mariano Rajoy…  Companies, organizations: BBVA, Bankia, Goldman Sachs, Coca-Cola, Reserva Federal…  Financial named entities: Ibex35, Dax Xetra…  Places: Londres, EE.UU., París…  Concepts: prima de riesgo, presidente del Gobierno, intervención parlamentaria, índice bursátil, situación económica…  Time references: hoy, ayer, sobre las 11 de la mañana…  Money amounts: 104 dólares, 1 euro…  Polarity positive/neutral/negative
  6. 6. Language Processing at the core of the Media & Publishing IndustriesContent aggregation
  7. 7. Language Processing at the core of the Media & Publishing IndustriesUser-generated content: automatic translation
  8. 8. Language Processing at the core of the Media & Publishing IndustriesUser-generated content: automatic moderation Tool for automatic moderation of social media, blogs, fora, etc. Offensive, illegal, inappropriate or objectionable content filtering
  9. 9. Language Processing at the core of the Media & Publishing IndustriesSocial Media Analytics: Sentimentalytics
  10. 10. Language Processing at the core of the Media & Publishing IndustriesVideo/audio indexing & search Transcription Indexing Contents Index Search
  11. 11. Language Processing at the core of the Media & Publishing IndustriesAutomatic Subtitling: transcription, segmentation &synchronization TranscriptionTEXT Processing (checking, proofreading, Storag etc.) e
  12. 12. Language Processing at the core of the Media & Publishing IndustriesData Journalism: exploration & analysis of info sources Look4leaks.net: Wikileaks case • Automatic translation of 251.000 cables (5 languages) • Semantic enrichment: entities, classification • Multifacet search: by embassy, person, country… Trial files: Gürtel case (corruption, >100 Kpages) • OCR, fuzzy recognition and multifacet search Spanish state of the nation address • Semantic analysis and search
  13. 13. Language Processing at the core of the Media & Publishing IndustriesTransmedia Content production for simultaneous and coordinated delivery through different channels Personalized delivery • Content of interest according to user profile • Contextual advertising E.g.: second screen apps
  14. 14. Language Processing at the core of the Media & Publishing IndustriesNew ways for monetizing content Selling content chunks: • chapters, sections of reference books, etc. Selling content aggregates: • full story through news published along the time about one topic
  15. 15. Language Processing at the core of the Media & Publishing Industries DAEDALUS, S.A.Jose C. Gonzalezjgonzalez@daedalus.eshttp://www.daedalus.es

×