At the Google Cloud Next conference on Wednesday, Google is rolling out a slew of AI and smart analytics tools. The tools are focused on applying AI to common business challenges such as structuring data from documents or forecasting inventory.
First, Google announced AI Platform in beta -- an end-to-end development platform that helps teams collaborate on machine learning projects. It's built for developers, data scientists and data engineers, enabling them to share models, train and scale workloads from the same dashboard within Cloud Console.
Next, Google is rolling out new versions of Cloud AutoML, the software that automates the creation of machine learning models that Google announced last year. Google initially rolled out AutoML Vision, which effectively extended Google's Cloud Vision API to recognize entirely new, customized categories of images.
Now, Google has made available in beta AutoML Tables, which lets you build and deploy machine learning models on structured tabular datasets. Users can ingest data from BigQuery and other GCP storage services into AutoML Tables.
The new AutoML Video is also in beta, allowing developers to create custom models that automatically classify video content Some clear use cases would be in the media and entertainment industry, where businesses could simplify tasks like automatically removing commercials or creating highlight reels.
Additionally, Google is rolling out AutoML Vision Edge to simplify training and deployment of high-accuracy, low-latency custom ML models for edge devices.
Google on Wednesday also announced Document Understanding AI in beta. The serverless platform automatically classifies, extracts, and enriches data from scanned or digital documents. It turns unstructured document data into structured data, and it integrates with integrates with technology stacks from Google partners like Iron Mountain, Box, DocuSign, Egnyte, Taulia, UiPath and Accenture.
Additionally, Google announced that its Contact Center AI service is now in beta. It New partners to the Contact Center AI program include 8x8, Avaya, Salesforce and Accenture.
On the data analytics front, Google is introducing new ways to move data into Google Cloud, as well as ways to clean, categorize and interpret data.
Cloud Data Fusion is a new, fully-managed service that lets users integrate data from various sources and join it with other data sources. It lets organizations take siloed data and prepare it for analysis in BigQuery.
Customers can also now get more data into BigQuery with the expanded BigQuery Data Transfer Service. BigQuery DTS automates data movement from SaaS applications to Google BigQuery on a scheduled, managed basis. In addition to Gogole's own apps, it now supports more than 100 popular SaaS applications, including Salesforce, Marketo, Workday and Stripe.
While Google is making it easier to get data to BigQuery, the exabyte-scale, serverless data warehouse is already growing quickly. The volume of data analyzed has grown by over 300 percent in just the last year, Google said.
Meanwhile, data analysts will be able to build their own data pipelines with Cloud Dataflow SQL, coming soon in public alpha. Using SQL, they can build Dataflow pipelines that automatically detect the need for batch or stream data processing. Dataflow SQL uses the same SQL dialect used in BigQuery, which allows data analysts to, for instance, use Dataflow SQL from within the BigQuery UI.
For analyzing data, Google is introducing BigQuery BI Engine in beta. It allows BigQuery users to analyze complex data sets interactively with sub-second query response time and with high concurrency. The tool is currently available through Google Data Studio. In the coming months, third-party business intelligence providers like Looker and Tableau will be able to leverage BI Engine as well.
Given how much business users rely on spreadsheets for analysis, Google is also introducing connected sheets, a new type of spreadsheet that works with a full dataset from BigQuery.
Additionally, Google is expanding BigQuery ML, a tool that lets data analysts build and deploy machine learning models on massive datasets directly inside BigQuery using SQL. Now, BigQuery ML includes new models like k-means clustering (in beta) and matrix factorization (in alpha) to build customer segmentations and product recommendations. Customers can also now also build and directly import TensorFlow Deep Neural Network models (in alpha) through BigQuery ML.
More from Google Cloud Next: