Contains the implementation of algorithms that estimate the geographic location of media content based on their content and metadata. It includes the participation in the MediaEval Placing Task 2014.
Performance benchmark between popular graph databases.
A framework for the analysis of social interaction networks (e.g. induced by Twitter mentions) in time.
Sentiment analysis framework developed by CERTH.
The project contains a set of convenience methods on top of common data repositories.
Monitors a set of social streams (e.g. Twitter status updates) and collects the incoming content.
Collects multimedia content shared through social networks.
Contains data abstractions for mapping a number of different social networks to a single representation.
Implementation of the diverse image search method used in MediaEval Diverse Image Search task.
Creates queries based on DySCOs for (a) collecting content from online sources, (b) retrieving content from SocialSensor store.
This project contains the main class and interface declarations to be used by other projects.
Contains a set of analysis processes on streams of incoming media items.
A set of methods for automatically detecting trending topics in streams of short texts (e.g. tweets).
A framework for "learning" how to classify social content as truthful/reliable or not.
Automatically extracts human readable title from a DySCO.
A collection of utilities that help with the analysis and indexing of geographic data.
Contains implementation of different methods for ranking annotated images based on the public figures (and their co-appearances) appearing in them.
A set of scripts to replicate different multimedia concept detection experiments.
Helps with the distribution of Twitter datasets by downloading sets of tweets (if still available) using their ids as input.
Detecting groups of photos that depict social events (concerts, soccer matches, etc.).