TigerGraph, provider of an advanced analytics and ML platform for connected data, is unveiling updates to TigerGraph Cloud, the native parallel graph database-as-a-service. With the update comes two new tools to improve the existing platform; TigerGraph Insights, a visual graph analytics tool for searching and exploring business insights, and ML Workbench, a Python-based framework built to expedite graph-enhanced machine learning application development.
“We believe that every data problem—underneath it—is a graph problem,” said Dr. Jay Yu, VP of product and innovation at TigerGraph. “By connecting data together, you can reveal really deep, elevated relationships.”
TigerGraph Cloud aims to improve graph technology adoption through both its streamlined design and real-time analytics and transactional workloads processing. Emphasizing its ease of use and accessibility, TigerGraph Cloud is widely available on AWS, Google Cloud Platform (GCP), and Microsoft Azure.
The platform simplifies not only deployment, but maintenance; offering users predictions, no-code what-if analyses, out-of-the-box support, and a variety of intuitive tools, TigerGraph Cloud helps both developers and data scientists alike unlock the potential of their data, according to the company.
TigerGraph Insights, the new no-code and low-code visual graph analytics tool, accommodates both technical and non-technical users in their development of multidimensional and interactive graphics for BI applications. Operating on top of TigerGraph’s parallel graph database platform, TigerGraph Insights connects intuitive graph data to generate graphics that can be linked together to create tables, charts, and maps for enhanced graph visualization. These graphics can further be integrated with an interactive dashboard application, which explains, in-depth, the connected graph data for its users’ conveniences.
“This particular insights tool is designed to expand our coverage from developers to non-technical users, like business or data analysts. They don’t want to write code,” said Dr. Yu. “In the past, these analysts had to rely on other engineers to create a middle tier via a visualization app to see a representation of the insights. We have a built-in capability so non-technical users can benefit from TigerGraph right away.”
The latter tool featured in this announcement, TigerGraph’s ML Workbench, is a graph machine learning toolkit that optimizes ML model accuracy, development cycle times, and business value. With a plug-and-play approach, ML Workbench works seamlessly with existing data pipelines and ML infrastructure, allowing its users to utilize the tools, workflows, and libraries best suited for them, according to the company. The tool deploys built-in, high-performance graph feature generation, sampling, and training, which is then extracted and converted into data formats compatible with downstream graph neural network modeling.
“With TigerGraph, we push machine learning into the database. We’re executing the machine learning algorithm as just a query inside the database engine,” said Dr. Yu. “People realized that graph data connections can introduce context to the data, allowing graph-based features to be brought as a feature into machine learning. As a result, we were able to improve precision recall by 50%—that’s game changing.”
“Graph is the future,” concluded Dr. Yu. “What’s next for big data are graphs, and TigerGraph is at the forefront for that as the most scalable system.”
To learn more about TigerGraph’s latest tool offerings, please visit https://www.tigergraph.com/.