The Knowledge: January 2017
Here is the first of this year’s “The Knowledge” posts. In the series, we curate links to recent articles that the our team have found interesting. We aim to highlight useful discussions and news for our community of readers. Please let us know in the comments what you think or would like to see more of next month!
GRAKN.AI is the database for AI. It is a database in the form of a knowledge graph that uses machine reasoning to simplify data processing challenges for AI applications. We are calling the posts in this series “The Knowledge” -we figured that the name of the blog series is especially appropriate because “The Knowledge” is also the name for the in-depth study of streets and places that taxicab drivers here in London must complete to obtain a licence to drive an official black cab.
As it is January, it is traditional to prognosticate about the coming year. Here are some of the most notable articles relevant to our field.
Ontotext published a well-researched list of 5 top trends in 2017 for semantic technologies. In the section on Knowledge Representation of Smarter Data, we thought this was good summary:
“Organizations will continue to look for solutions that would enable them to reveal all the relationships between concepts, thus enabling their data to be ‘smart’: classified in a meaningful way and linked to other relevant datasets [...] the technology to do this is not just in sci-fi books. Semantic data integration exists and will be evolving.”
Cambridge Semantics have also been gazing into their crystal ball, and recently published their predictions for the Big Data industry in 2017, which are: 1. Democratisation of data analysis, 2. The evolution of relational databases, 3. Graph-based databases for emerging technology, 4. Cloud-enabled platforms, and 5. Uptake by small and medium businesses.
Finally, on the theme of trends, at the end of November last year, Dataversity published a review of trends and predictions for Semantic Web and Semantic Technologies. The article predicts that there will be increasing emphasis on developing and investing in knowledge graphs. An interesting point they make is that:
“the pace of progress will be moderate […] one reason is the lack of skilled personnel in Semantic Web and knowledge-enhanced computing topics.”
This is something we at GRAKN.AI have considered key for some time, and we are seeking to eliminate the dependency on such personnel. Large-scale data in a traditional graph database results in an exponential growth in the permutation of paths to get between data. It is true that queries become complex, an engineer must form them and they are not reusable between engineers. However, with a knowledge graph the responsibility moves to the ontology engineer, and the ontology is reusable across the data, so it can be transferred. Additionally, the use of a knowledge graph allows you to reduce the complexity of the query. GRAKN.AI stores data in a form that allows machines to understand the meaning of information in the complete context of their relationships. Consequently, GRAKN.AI allows computers to process complex information more intelligently with less human intervention.
Setting aside the tea leaves for now, we have also enjoyed reading a well-crafted introduction to graph databases, published recently by Bloor research, that covers what graph databases are, what they do, and why we need them. It also discusses emerging trends and the current landscape, and provides a link to download an excellent primer on graph databases for further information.
We particularly like the point made in the article that there is
“ […] a distinction between vendors targeting known-known problems as opposed to those that also cover known-unknowns and those tackling unknown-unknowns: the most intractable of all”.
We are firmly in the latter camp!
News from GRAKN.AI
We’ve been busy here at GRAKN.AI over the last few months. Earlier this month, we were proud to announce that we were part of a collaboration with a community of individuals and companies, including Expero, Google, Hortonworks and IBM, to launch JanusGraph, a new fork of Titan under The Linux Foundation. To find out more, please take a look at our blog post, or a summary of the story on Datanami.
GraphDay Texas 2017
The videos are a great way to find out more about us and what we are building. Boris also wrote a summary of the presentations he attended.
We have just announced two meetups for early March: on March 6th in London, we have the Inaugural Opensource Graph Technologies meetup. This meetup will be the first in a series where we introduce different opensource graph technologies and their applications. In this meeting, we will focus on three technologies: TinkerPop, Giraph, and GRAKN.AI. The session will be followed with some time for participants to chat and share their experiences.
In addition, the first JanusGraph meetup takes place on March 1st in NYC.
Introducing Graql Bot
You may have noticed that our blog has had a makeover in the last few weeks — we have a new logo!
The posts we’ve published previously are still all available, although we’ve reorganised the blog to make it more easy to find things, by subdividing our posts into those that teach you about Grakn, Graph News, Startup Life and Technology. In recent weeks, on our blog, we’ve published articles about:
- Building Chatbots
- Working with Grakn using Haskell
- A Comparison of GRAKN.AI and OWL
- A Review of 2016 for our Startup
We also ran a highly-successful series of posts about popular technology during Advent 2016, which is still available to read. As it is a virtual, chocolate-free advent calendar, you can dive in without feeling guilty this January. Enjoy, and see you next time!