“Insight is the new currency for success,” said Bob Picciano, senior vice president at IBM Analytics. “And Watson is the supercharger for the insight economy.”
Picciano, speaking at the World of Watson conference in Las Vegas on Tuesday, unveiled IBM’s Watson Data Platform, touted as the “world’s fastest data ingestion engine and machine learning as a service.”
The cloud-based Watson Data Platform, will “illuminate dark data,” said Picciano, and will “change everything–absolutely everything–for everyone.”
The Watson Data Platform, available now, “will change how quickly businesses can interact with data,” he said. We are inundated with data, Picciano said, from “Twitter streams to weather data to images to video to text to audio,” and are grappling with how to make intelligent decisions with it.
“This is an on-ramp to building a cognitive business,” said Picciano. “It will take data and enable AI-powered decision-making.”
Back when IBM Watson beat Ken Jennings at Jeopardy, the power processor ran on 100 calculations, a million times per second. Now, Picciano said, it does 1 million calculations, a million times per second.
Many businesses are wondering how to bring machine learning and AI into their enterprise, said Rob Thomas, vice president of product development at IBM.
“You know it’s real,” Thomas said at the World of Watson conference, “but you haven’t made it a reality yet.”
IBM’s Watson Data Platform, he said, can “help you transform, so you can win in the era of cognitive computing.” The platform, built on an open-source foundation, is the first of its kind for enterprise, and will change the way businesses handle data, Thomas said.
The platform is built on Apache Spark, and can “use structured and unstructured data and open machine learning libraries” to build models for businesses. According to the press release, IBM Watson Data Platform will:
- Ingest large volumes of diverse data into the cloud at record speeds–in up to 100 gigabytes per second
- Cleanse, edit and shape data for easier modeling
- Add and remove collaborators as needed while maintaining version control
- Drag and drop services into analytic notebooks for better productivity and time management.
It will also allow businesses to integrate their own data with external data, and will address data governance from the very start of the process.
According to the release, the platform “is building an increasing ecosystem of technology service providers, allowing data professionals to use the language they are comfortable with and the services they prefer.” This includes SQL, Python, R, Java, and Scala, among other services.
“IBM Watson Machine Learning Service will begin with Apache SparkML with additional algorithms included in the future and can be accessed through Watson Data Platform, as an API on IBM Bluemix or on z/OS,” the release stated.
SEE: How Google is getting smarter with artificial intelligence (CBS News)
An important component of IBM’s Watson Data Platform, Ritika Gunnar, vice president of offering management for Watson Data Platform told TechRepublic, is the piece that allows for collaboration.
“The number of people in today’s business who have to be able to leverage data as part of their everyday lives, to make sense of it, to drive intelligent decision-making, has grown rapidly,” she said. Gunnar pointed to the need for businesses to collaborate with data across departments to make decisions. The simple interface, she said, helps give everyone, from those who are data savvy to “citizen analysts,” a chance to work with data.
“The notion of being able to work on data together, to share across the business, is a huge opportunity to accelerate insights and uncover things that weren’t able to because of the silos within the organization that prevented working on common information,” she said.
SEE: Machine learning: The smart person’s guide (TechRepublic)
Gunnar cited a Harvard Business Review study in which 90% of professionals said they wanted digital data to be accessible across their organization.
“Data-driven decision making is not in the hands of the few,” said Gunnar, “it’s in the hands of the many.”
Thomas echoed this point. “We’re making this a team sport,” he said.
IBM’s Watson Data Platform is not the first to make its AI platform available. In January, Microsoft offered open source machine learning on GitHub. Google has its own deep learning platform, TensorFlow. In May, Amazon announced its own Deep Scalable Sparse Tensor Network Engine (DSSTNE) library. And in September, Baidu unveiled a machine learning platform.
Still, IBM sees its service as unique.
“IBM’s Watson Data Platform will bring machine learning to the masses,” said Thomas. “Once data is in the platform, you are ready for AI.”
- Amazon open sources its deep learning software (ZDNet)
- Machine automation policy guidelines template (Tech Pro Research)
- How Amazon wants to bridge the data science gap by bringing machine learning to the cloud (TechRepublic)
- Machine learning face-off: Microsoft uses Band to show what its Watson rival is capable of (ZDNet)
- Google DeepMind: The smart person’s guide (TechRepublic)
- How developers can take advantage of machine learning on Google Cloud Platform (TechRepublic)