Over the past 6 months and 50+ webinars, we’ve realised that while we want to continue to build out the content that goes into specific areas and topics around Grakn’s capabilities; we also want to give the community a platform.
Our newest community event brings members of the Grakn Community to the forefront. Through tech panels, presentations, use case deep dives and more, our community shares the latest and greatest from the Grakn Cosmos.
Looking forward to seeing you there and hopefully participating as a panelist, presenter, round table guest, or running demo – this is the Grakn Orbit.
Join as an attendee and hear from those building and solving the world’s most complex problems.
Check out past sessions at the bottom of this topic…
Digging into Public Cancer Literature Using Python with Gunnar Kleeman
Data originating from deep-tech domains such as biotechnology, agriculture, and chemistry often require significant study to understand. At the same time, deep-tech data is becoming more relevant and available in the form of publicly funded data repositories. Cancer research is an example of a deep-tech treasure that can be accessed with the right tools. In this talk, I will demonstrate how I used Biopython for the collection and compilation of cancer data and then the Grakn python client to analyze it with a knowledge graph. This pipeline allowed me to get a birds-eye view of multiple cancer types and the research groups working on them.
Exploring Drink Recipes Biochemical Balance with Justine Chia
Cocktailgraph is a platform to facilitate the exploration of ingredients and recipes from a biochemical perspective. We built a graph containing drink and cocktail recipes from around the world; this data space is enriched with biochemical data sourced from scientific literature.
Meet GraMi - the Grakn migrator from Henning Kuich
Do you have a lot of data for your Grakn use case and want to focus on modeling, inference, and queries? Use GraMi (GraknMigrator) to read your entity and relation data from tabular files and migrate them into Grakn at scale.
gTime - how time works in Grakn from Brett Forbes
Depending on the application, various complexities can arise in modelling time, including:
- Storage and Indexing: How to store time data fields so they can be accessed quickly
- Division and Aggregation: The smallest sensible time increment, and aggregation methods
- Multiple Timelines: Dealing with changing data
- Slowly Changing Dimensions: Classifying responses to temporal data changes:
- Temporally Variable Schema and Data: Capturing schema and data changes
- Mining Temporal Relations: Methods of relating times to other times (e.g. before, after etc.)
These are not small issues, as entire databases have been devised and market shares carved out
around the idea of being good at Problem 1 and 2 generally (e.g. MapR, TimeDB), or for specific use
cases such as IoT. Much time and effort has been put into Problems 3, 4 and 5 particularly with
RDBMS and Dimensional Warehouses (e.g. DataVault), and even SQL extension recommendations.
OWL-Time is the dominant solution for Problem 6 and has notorious ambiguity problems.
Finally, there is an unspoken desire to get a more functional description of time in knowledge graphs, calculable and unambiguous, as well as to connect causality into our models. A new way is needed.
Applications, business impact and getting it done; how knowledge graphs are changing the game
Konrad Mysliwiec - GSK
Michael Seidel - Meta.link
Andrew Jillions - Adjective Ventures
Julien Richard - OpenCTI
Aun Hussain - AppSavy.Its
OnDemand Videos: https://www.youtube.com/watch?v=So_j87HxJ1c
Text Mined Knowledge Graphs with Grakn Tech Panel | Grakn Warriors Grakn Community | Chris from accha Introduction to Knowledge Graphs with Grakn and Graql Tech Panel | BioGrakn COVID group Grakn Community | Natalie from AstraZeneca Grakn Academy short | Knowledge Modelling Principles