Scalable online machine learning
for predictive analytics and
real-time interactive visualization

Results

SOFTWARE (Open Source)

PROTEUS is using GitHub as the code repository for the project's results. There are several repositories for the different contributions provided by PROTEUS, open to the community to join, make use, and contribute.

Join us! Explore and contribute at: https://github.com/PROTEUS-H2020

PROTEUS Engine

An overhauled version of Apache Flink supporting hybrid computation on batch datasets and data streams.

Find it at: https://github.com/proteus-h2020/proteus-engine

PROTEUS Language

Declarative language library for Scalable Data Analysis.

Find it at: https://github.com/proteus-h2020/proteus-language

PEACH: Proteus Elastic Cache

The PROTEUS Elastic distributed caché.

Find it at: https://github.com/proteus-h2020/peach

PROTEUS Incremental Analytics

A backend module that implements incremental version (~O(1) computational cost using approximations) of most common analytics operations.

Find it at: https://github.com/proteus-h2020/proteus-incremental-analytics

SOLMA: Scalable Online Machine Learning and Data Mining Algorithms

A scalable library adapted to the data analytics platform, Apache Flink. It consist of efficient distributed online algorithms for basic utilities, sketches as well as advanced online predictive analytics for tasks like classification, clustering, regression, ensemble methods, and novelty and change detection.

Find it at: https://github.com/proteus-h2020/SOLMA

PROTEIC.JS

An HTML5 and CSS3 charts library, ready for batch and streaming data.

Find it at: https://github.com/proteus-h2020/proteic

NEWS & EVENTS

PROTEUS at the EDBT/ICDT 2017

Mon 20 Mar 2017

PROTEUS will be presented at the EuroPro Workshop during the EDBT/ICDT 2017 (20th International Conference on Extending Database Technology; 20th International Conference on Database Theory), taking place in Venice (Italy) from March 21 to 24, 2017.

The paper titled "PROTEUS: Scalable Online Machine Learning for Predictive Analytics and Real-Time Interactive Visualization", by Bonaventura Del Monte, Jeyhun Karimov, Alireza Rezaei Mahdiraji, Tilmann Rabl, and Volker Markl (DFKI) has been recently accepted for presentation.

A full list of accepted papers of the conference can be found at: http://edbticdt2017.unive.it/?accepted-papers

2 Comments

Upcoming consortium meeting in Berlin (26-28 April 2017)

Wed 08 Feb 2017

The next consortium meeting of PROTEUS is approaching. Partners will be meeting in Berlin, hosted by DFKI, next April, 26-28.

In this case, partners are meeting to discuss the progress until the third milestone (closing May 2017), which constitutes the main mid-project achievement. The meeting will also be used for the preparation of the interim review meeting, which will be happening in Luxembourg on July 5th.

3 Comments

PROTEUS presented at the Big Data Spain 2016

Thu 17 Nov 2016

PROTEUS has taken the opportunity to present their results during the Internation conference Big Data Spain 2016. In this case, the talk Advanced data science algorithms applied to scalable stream processing was presented by David Piris and Ignacio García (TREELOGIC).

The public and program of the event brought together an interesting environment for fostering Big Data market opportunities as well as research intiatives. In addition, PROTEUS was also an active presenter in the poster's session organised during the breaks, where partners had the opportunity to network with attendees, including both industry and research representatives.

1 Comments

This project is funded by the European Union (Horizon 2020, Ref: 687691)