Search Results for Tag: sentiment analysis
Navigating the world of emotions!
MixedEmotions creates large-scale emotion analysis for innovative markets. So far, it has been challenging to fully make sense of big data analytics from different sources, modalities and languages in an industrial context. MixedEmotions aims to change that.
MixedEmotions handles various input channels including multilingual text data sources, A/V signals (speech, audio, video), social media and structured data. Results of these analytics are merged and social context will be exploited by using semantic knowledge graphs to create meaningful applications. Three pilots will be developed addressing broadcasters, call centers and brand reputation management companies.
Deutsche Welle is a user partner and works on two journalistic scenarios: 1. a recommendation system offering customised content for our SmartTV end users, and 2. a smart dashboard for journalists to explore and detect trending topics.
The project runs from April 2015 – March 2017.
DateThursday, June 4th, 2015
TagsEmotion Recognition, Knowledge Graph, Recommender Systems, sentiment analysis, Smart TV, Speech Recognition
Multilingual media monitoring with smart content analysis
MULTISENSOR is developing new multilingual online media analysis technologies. The aim is to enable professional media monitors (e.g. journalists, press agencies, analysts) to attain a comprehensive and exact understanding of topics being discussed on the web.
The MULTISENSOR platform scans multiple heterogeneous sources, then translates and processes the content. It visualizes details about sentiment, entities, social and other important aspects necessary to provide context. Adding automated summarization of each content item MULTISENSOR helps you to better distinguish the relevant signals among the noise.
Deutsche Welle is a user partner, leading the use case development and requirements gathering. We are also involved in dissemination and evaluation of the project’s results.
The project runs from November 2013 – November 2016.
Bring me to the MULTISENSOR website
DateFriday, August 15th, 2014
Tagsmedia monitoring, multilingual, Multisensor, semantic representation, sentiment analysis, social media, summarization
Smart search in the social web
SocialSensor is about multimedia indexing and search in the Social Web. Content from social networks such as Twitter tweets are becoming increasingly important for news and infotainment purposes. For journalists, social media is often the only source of information in situations where other access is limited (e.g. Arab spring, conflict in Syria). But how can a journalist make sure not to miss any relevant information?
With regard to news, SocialSensor aims for developing a reliable news and trend detection algorithm. The system will identify key influentials in social networks and supports the verification of social media content.
Deutsche Welle is a user partner and contributes to news use case, validation and dissemination.
The project runs from Oct 2011 – Sep 2014
Bring me to the SocialSensor website
DateFriday, November 2nd, 2012
Intelligent archive of the social web
ARCOMEM is about memory institutions such as archives, museums, and libraries in the age of the Social Web. Memory institutions are more important now than ever: as we face greater economic and environmental challenges we need our understanding of the past to help us navigate to a sustainable future.
Which contributions of the social web will stay available in the future and what will be preserved? And how will they be accessible and explorable for users? ARCOMEM creates tools which helps archivists to exploit the social web.
Deutsche Welle is a user partner and contributes to the broadcaster use case, dissemination and validation.
The project runs from Jan 2011 – Dec 2013
Bring me to the ARCOMEM website
DateFriday, November 2nd, 2012