project presentation
play

Project Presentation eur euros osent entiment ent.eu eu - PowerPoint PPT Presentation

Project Presentation eur euros osent entiment ent.eu eu EUROSENTIMENT FP7-ICT-2011-SME-DCL GA 296277 Language Resource Pool for Sentiment Analysis in European Languages At a glance EUROSENTIMENT Language Resource Pool for Sentiment


  1. Project Presentation eur euros osent entiment ent.eu eu EUROSENTIMENT FP7-ICT-2011-SME-DCL GA 296277 Language Resource Pool for Sentiment Analysis in European Languages

  2. At a glance ● EUROSENTIMENT Language Resource Pool for Sentiment Analysis in European Languages (GA 296277) ● STREP Project co-funded by the European Union FP7-ICT 2011 SME Digital Contents and Languages ● Project start: September 2012 (24 months) ● EC Contribution: 1.930.000 € ● Project Coordinator: Paradigma Tecnológico ● Partners: Fondazione Bruno Kessler, Expert System S.P.A., Sindice LTD., Universidad Politécnica de Madrid, National University of Ireland EUROSENTIMENT FP7-ICT-2011-SME-DCL GA 296277 Language Resource Pool for Sentiment Analysis in European Languages

  3. Vision and Goal EUROSENTIMENT interoperable pool of shared language resources , accessible by means of well-defined models and frameworks that leverage the promotion of SMEs in the emerging market of Sentiment Analysis products and services. The project participants contribute with high-quality language resources in 6 languages: English , Italian , Spanish , Catalan , German , Portuguese EUROSENTIMENT FP7-ICT-2011-SME-DCL GA 296277 Language Resource Pool for Sentiment Analysis in European Languages

  4. Value Proposition EUROSENTIMENT proposes extending the market / business opportunities of the participants providing... ● A platform for cooperating, being able to interoperate their resources / services in a win-win manner ● A platform for gaining visibility at European level, being a reference for potential customers ● A platform that allows to scale their business model. Shift from case-by-case integration to a SAAS business model EUROSENTIMENT FP7-ICT-2011-SME-DCL GA 296277 Language Resource Pool for Sentiment Analysis in European Languages

  5. How? Our Language Resource Pool Vision ● Interoperability between resources and services ● Follow a domain oriented approach. We need to combining semantic and linguistic analysis ● Be able to define different levels of accuracy/throughput. Document-level: easy, Paragraph-level: average, Entity-level: hard. ● Based on well-known resources , facilitating its adoption and extension. EUROSENTIMENT FP7-ICT-2011-SME-DCL GA 296277 Language Resource Pool for Sentiment Analysis in European Languages

  6. Project Overview: Barriers & Solutions ● Fear of competitive losses: ● Language resource sharing based on a community governance model ● Business manager for managing partners ● Security / Restriction in access to resources ● No standards or normalisation: semantic interoperatibility (Wordnet + Linked Data) ● Multilingual issues: bunch of specialists (community) and interoperability ● Atomized and scattered projects, with reduced visibility: Shared pool with QA procedures and project dissemination strategies EUROSENTIMENT FP7-ICT-2011-SME-DCL GA 296277 Language Resource Pool for Sentiment Analysis in European Languages

  7. Objectives O1 . Provide semantic interoperability and connectivity between several multilingual sentiment analysis resources available online for the first time. O2 . Reduce the cost of aggregating new language resources to the shared resource pool by providing best-practice guidelines and QA procedures. O3 . Provide a self-sustainable and profitable framework for language resource sharing. O4 . Demonstrate the impact of the developed pool by providing public access to a multilingual demonstrator in the media domain EUROSENTIMENT FP7-ICT-2011-SME-DCL GA 296277 Language Resource Pool for Sentiment Analysis in European Languages

  8. Target Groups ● Service developers : able to use dataset in order to develop new services by integrating the provided interfaces, based on a commercial license. ● Content providers : can use dataset in order to increase the value of their contents by aggregating consumers' trends and customers’ perceptions. ● Language resource owners : benefit from having a shared pool where there language resources are valorised thanks to the integration with the rest of resources, resulting in higher visibility and return of investment. EUROSENTIMENT FP7-ICT-2011-SME-DCL GA 296277 Language Resource Pool for Sentiment Analysis in European Languages

  9. Research Areas ● Domain-oriented shared language resource based on WordNetDomains and aligned with WordNet Affect ● Shared language resource based on Linked Data ● Multi-lingual shared language data pool ● Self-sustainable and profitable framework for language resource sharing EUROSENTIMENT FP7-ICT-2011-SME-DCL GA 296277 Language Resource Pool for Sentiment Analysis in European Languages

  10. Strategic Impact ● Improved European competitive position in a multilingual digital market through the provision of better products and services to citizens and businesses . ● Novel forms of partnership between new programme entrants and established players, reduced development costs and shorter time-to-market, thus stimulating innovation and expanding markets . ● Result-driven knowledge transfer between research centres (and their spin-offs) and progressive technology providers (especially SMEs), data brokers/aggregators and content providers. EUROSENTIMENT FP7-ICT-2011-SME-DCL GA 296277 Language Resource Pool for Sentiment Analysis in European Languages

  11. Consortium ● SME's ● Paradigma Tecnológico ● Expert System ● Sindice ● Universities and Research Centres ● Universidad Politécnica de Madrid ● National University of Ireland, Galway ● Fondazione Bruno Kessler EUROSENTIMENT FP7-ICT-2011-SME-DCL GA 296277 Language Resource Pool for Sentiment Analysis in European Languages

  12. Thank you for your attention EUROSENTIMENT FP7-ICT-2011-SME-DCL GA 296277 Language Resource Pool for Sentiment Analysis in European Languages

Recommend


More recommend