DataWalk Visual Querying
Unique DataWalk technology enables simple, powerful no-code querying across vast amounts of complex data.
Easily Create And Run No-Code Queries With The DataWalk Universe Viewer
The DataWalk Universe Viewer is an enterprise knowledge graph that provides a visual representation of all data in the system, along with the cross-references between data sets. You can easily generate queries simply by traversing the Universe Viewer, and applying filters to both data sets and the connections between data sets. The Universe Viewer is a “no code” interface, such that no expertise is required in SQL or a programming language in order to create and run even highly complex queries.
Complex Queries Quickly Complete!
A common challenge with traditional systems is that complex queries overwhelm the system and cannot be executed. DataWalk is different! By coupling graph analytics capabilities with patented DataWalk technology, complex queries complete, and complete quickly! As shown in the figure below, DataWalk benchmark results show that unlike traditional relational database systems which fail to generate a result after a relatively small number of joins, DataWalk maintains linear response time through 600 joins.
Figure 1. DataWalk maintains linear performance through 600 joins
Queries You Can Understand And Rely On
Another common issue with traditional relational databases is that complex SQL queries can be many pages long, and be so complicated that even the programmer may not really know what the query is doing. This does not inspire confidence that you can trust the results. DataWalk is different! In DataWalk you can iteratively create your query one step at a time via a visual interface, such that you can be confident that you understand all of the steps of your query, and you can trust the results.
Instantly Create, Save, Re-run, Modify, And Share Queries
In addition to instantly creating queries by traversing and filtering via a visual interface, you can save a query, share it if desired, and you or other users with the appropriate permissions can instantly re-run that query just by clicking on a DataWalk “big button”. This enables you effectively to encode organizational knowledge and empower more users to run more advanced queries, if desired. Saved queries can easily be modified.
Figure 2. Sample DataWalk “Big Buttons.”
Easily Combine Queries To Create Scores
Queries, which can also be considered as analyses or rules, can easily be weighted and combined to create scores, such as a risk score. You can create and modify scores with just a few clicks, to generate and tune risk scores for people, insurance claims, transactions, or anything else.
Use Queries To Generate Training Sets for Machine Learning
DataWalk is a great platform for managing the end-to-end machine learning lifecycle. When saving queries, you can instantly use and mix them as training sets for your own machine learning models, or you can use DataWalk’s AutoML capability to generate the best model automatically. It’s very easy to manage training sets and models, and to share, archive and re-use them while having full lineage of how the highest performance models have been generated.
