“We wanted to democratise data preparation tasks”
Tableau has added natural language processing to its platform with its release this morning of Ask Data, which allows users to search for – and visualise – data using natural language, rather than keywords. The release is one of two major upgrades bundled into Tableau 2019.1, which is generally available today.
The other has been dubbed Data Prep Conductor. A native integration with Tableau Server, it allows organisations to schedule and manage self-service data preparation at scale across the 65 different data sources Tableau can connect with, from Aurora to MemSQL, via Apache Drill, SAP, MongoDB, Oracle and more.
Natural language recognition is something the Seattle-based data visualisation specialist has been working on for some time. It turbocharged the plans with its 2017 acquisition of natural language query specialist startup ClearGraph.
Tableau Upgrades: Ask Data
Demonstrated to Computer Business Review by Tableau’s Chief Product Officer Francois Ajenstat this week, the tool was fast, effective and user-friendly
Ask Data uses algorithms to automatically profile and index data sources/queries; for example if someone searches their sales figures for “British furniture”it automatically recognises that they need “product category” filtered to “furniture,” and “Country” set to “United Kingdom.” Terms can then by refined as desired.
Ajenstat describes Ask Data as “providing a more intuitive and natural way to interact with data, lowering the barrier to entry for analytics and allowing people to ask questions in plain language and get highly relevant insights faster.”
It is available as part of Tableau Server and Tableau Online for no extra cost.
Tableau Data Management
Also launching alongside Tableau 2019.1 is Tableau Data Management. Being sold as a new subscription offering for $5.50/user per month, it includes Tableau Prep Builder for building data flows, and Tableau Prep Conductor for sharing flows and managing them across the organisation.
A demonstration by CPO Francois Ajenstat showcased the tool. As he put it: “Data is growing at velocity. We wanted to democratise data preparation tasks. This is an easy to use, point and prep data clean tool that has been designed to be fast, visual and approachable for business users, not just specialists.”
Demonstrated on car sales figures, the tool suggested options for users including removing spaces (a common error in data formatting), or numbers, or merging terms with similar spelling and/or different uses of uppercase and lowercase. (e.g. Merging Honda and hondaa; march2019 and Mar19)
He added: “Most of the tools in this space were designed for specialists; designed for people who can code. Only once the data is prepared can analysts and business users start performing their analysis. This allows business users as well to clean their data so they can get insight fast that helps inform business decisions.”
Prep Conductor also provides increased visibility into the health of data sources, allowing users to set status reports and even proactive alerts, e.g. if sales or returns go above a certain threshold in a certain geography.