Google has launched a public beta of Cloud Dataprep, an intelligent managed cloud service designed to dig-up, clean and prepare both structured and unstructured data for analysis and the training of machine learning algorithms.
Created in conjunction with data wrangling firm Trifacta, Google Cloud Dataprep runs on serverless infrastructure to dynamically scale with its user performance and data needs.
Sporting an interface that leans heavily on visuals, Google's product manager Eric Anderson claims the Cloud Dataprep makes data preparation "intuitive and approachable" for analysts who want to modify and interact with their databases directly.
When analysts don't want to get their hands digitally dirty, built-in intelligence allows Cloud Dataprep to follow and learn data usage patterns to optimise them in order to make the preparation of data faster and less likely to be affected by user errors.
These capabilities come courtesy of Trifacta's data wrangling tech taken out of its data preparation platform.
Prior to the public beta, Cloud Dataprep has been in a closed testing programme, which saw it put to use by a data-driven marketing agency and an advisory services company that works to provide its clients with ways to identify and use big data to drive changes in their business and monetise their information.
While there are plenty of data visualisation and wrangling tools in the enterprise IT market targeted at making big data analytics a less daunting task, Cloud Dataprep makes uses of Google's large cloud infrastructure footprint. This should put the data wrangling tool in a position to be used by all manner of companies looking to tap into their troves of big data without incurring hefty hardware and infrastructure costs.
