DataAI » Blog
Blog

Blog

Our beautiful blog

Presentation at PyData London 2015

This is the abstract of the talk at PyData London 2015 “Jointly embedding knowledge from large graph databases with textual data using deep learning”

pydataI will present recent advances in combining structured graph data with textual data using embedding word representations from a large corpus of unlabelled data. This allows to expand the knowledge base graph and extract complex semantic relationships. Targeting knowledge graphs completion is a recent paradigm that allow extraction of new relations (facts) from existing knowledge graphs like Freebase or GeneOntology.

Word embeddings represents each entity into a low dimensional space and the relationships as transformations which has the advantage of making the search space continuous. This allows to encode the entities and transformations with global information from the entire graph. On the other hand, word embedding approaches, like word2vec, extracted from unlabeled text allows representations of words as vectors, although it doesn’t allow to extract relations. By careful alignment of entities from free text with a knowledge graph it is possible to combine both approaches and jointly extract new knowledge through relationships between entities and words / phrases. We will show results from applying this technology to biomedical data.

EIT Digital involves DataAI to build European Trusted Cloud solutions

dataaieit

DataAI presentation at the EIT Digital/Trusted Cloud initiative in Helsinki

EIT Digital’s Trusted Cloud high impact initiative is an ambitious approach to find a European solution to store digital data and contents, so that consumers and businesses in Europe do not have to worry on where and by whom their valuable digital age assets are handled.

In the Trusted Cloud high impact initiative the small and large companies (British Telecom, Telecom Italia, F-Secure) across Europe are learning from each other and accelerating the building of partnerships and customer-driven businesses.

“This activity is a great example of growing collaboration and innovation when the vision of the big picture is solid. We already see a growing interest to European alternatives for cloud solutions and especially the 8 small companies are finding our support extremely valuable, both in technology and mentorships. EIT Digital creates a beautiful framework for building this ecosystem”, explains Markku Kutvonen, activity leader for Trusted Cloud.

One of the 8 selected SMEs to work on the Trusted Cloud is DataAI. We are a data science consultant firm from UK. DataAi’s projects are focused on data exploration, proof-of-concept development, conceptualization and products. In the Trusted Cloud initiative DataAi works on a data set provided by Thales to detect abnormal maritime traffic behavior with machine learning algorithms.