Jeffrey Schwartz of Redmond Magazine recently wrote, “Nearly a year after launching its Hadoop-based Azure HDInsight cloud analytics service, Microsoft believes it’s a better and broader solution for real-time analytics and predictive analysis than IBM’s widely touted Watson. Big Blue this year has begun commercializing its Watson technology, made famous in 2011 when it came out of the research labs to appear and win on the television game show Jeopardy. Both companies had a large presence at this year’s Strata + Hadoop World Conference in New York, attended by 5,000 Big Data geeks. At the Microsoft booth, Eron Kelly, general manager for SQL Server product marketing, highlighted some key improvements to Microsoft’s overall Big Data portfolio since last year’s release of Azure HDInsight including SQL Server 2014 with support for in-memory processing, PowerBI and the launch in June of Azure Machine Learning.” Read more
The Ventana Research summit took place this week, and semantic and related technologies had a place at the table.
Among the keynoters discussing the topic of Inspiring Business Technology Innovation to Change Business and IT Forever, for example, was Nedshad Bardoliwalla, co-founder and vp of products at data prep vendor Paxata. He discussed the need to rethink how to innovate with data, as that will “drive the biggest increases in value for your organization for the foreseeable future.”
As part of that, he explained that in a world where everything physical on the planet will have a digital representation, businesses should pay attention to factors including the “massive and interesting algorithms around recognition systems, around deep learning, around semantic models that let us understand images and text in ways we never could….Take advantage of those if you are to innovate and bring capabilities to market that change way people think of data.”
Manju Bansal of the MIT Technology Review recently wrote, “The year was 1961. Computers were still in their infancy, and the race to the moon was just beginning. Edward Lorenz, an MIT meteorologist, was developing a weather- prediction model. Lorenz theorized that a miniscule occurrence, such as a tiny butterfly flapping its wings in the Amazon, could hypothetically set in motion a chain of events that could cause tornadoes to touch down in Texas a few days later. That model (illustrations of which visually resembled a butterfly) eventually came to be known as the butterfly effect.” Read more
Research firm Forrester at the end of September issued its Forrester Wave: NoSQL Key-Value Databases, Q3 2014 report. The report looked at seven enterprise-class vendors in the space: Amazon Web Services, Aerospike, Basho Technologies, Couchbase, DataStax, MapR Technologies, and Oracle.
Noting that the current adoption of NoSQL is at 20 percent and is likely to double by 2017, Forrester principal analyst and report author Noel Yuhanna and his co-authors explain that top use cases for key-value database include social and mobile apps, scale-out apps, Web 2.0, line-of-business apps, big data apps, and operational and analytical apps.
That said, he also notes that the lines between key-value store, document database and graph database NoSQL solutions are blurring, as vendors look to satisfy broader enterprise needs and better appeal to app developers. “Relational database management system vendors, such as Oracle, IBM, Microsoft and SAP, will broaden their current relational database products to include key-value, graph and document features and functionality to deliver more comprehensive data management platforms in the coming years,” the report states.
Alok Prasad and Lee Feigenbaum of Cambridge Semantics recently wrote for CMS Wire, “Over the past few years, major enterprises have shown interest in combining semantic web technology with big data for added value. Let’s take a look at what enterprises are seeking and why they think semantic web can make big data smarter… In traditional big data IT solutions, the data model and the IT solutions are designed to address specific business needs and to handle specific data types and data sources. As the business needs and data sources change, the IT solutions no longer work and new data marts and new solutions must be built.” Read more
Bernard Marr recently wrote, “It’s been estimated that by 2015, almost two million people will be employed in big data jobs in the US. Hal Varian, Google’s chief economist, is quoted as saying “…the sexy job in the next 10 years will be statisticians” and Tom Davenport, Distinguished Professor at Babson College, believes that a data scientist has the sexiest job of the 21st century. So what are these sexy jobs? Here’s a quick look at some of the positions available today that might allow you to break into the glamorous and exciting world of the big data professionals.” Read more
Ron Miller of TechCrunch reports, “IBM today announced a new product called Watson Analytics, one they claim will bring sophisticated big data analysis to the average business user. Watson Analytics is a cloud application that does all of the the heavy lifting related to big data processing by retrieving the data, analyzing it, cleaning it, building sophisticated visualizations and offering an environment for communicating and collaborating around the data. And lest you think that IBM is just slapping on the Watson label because it’s a well known brand (as I did), Eric Sall, vp of worldwide marketing for business analytics at IBM says that’s the not the case. The technology underlying the product including the ability to process natural language queries is built on Watson technology.” Read more
Big Data has been getting its fair share of commentary over the last couple months. Surveys from multiple sources have commented on trends and expectations. The Semantic Web Blog provides some highlights here:
- From Accenture Anayltics’s new Big Success With Big Data report: There remain some gaps in what constitutes Big Data for respondents to its survey: Just 43 percent, for instance, classify unstructured data as part of the package. That option included open text, video and voice. Those are gaps that could be filled leveraging technologies such as machine learning, speech recognition and natural language understanding, but they won’t be unless executives make these sources a focus of Big Data initiatives to start with.
- From Teradata’s new survey on Big Data Analytics in the UK, France and Germany: Close to 50 percent of respondents in the latter two countries are using three or more data types (from sources ranging from social media, to video, to web blogs, to call center notes, to audio files and the Internet of Things) in their efforts, compared to just 20 percent in the UK. A much higher percentage of UK businesses (51 percent) are currently using just a single type of new data, such as video data, compared with France and Germany, where only 21 percent are limiting themselves to one type of new data, it notes. Forty-four percent of execs in Germany and 35 percent in France point social media as the source of the new data. About one-third of respondents in each of those countries are investigating video, as well.
What best practices should inform your company’s text analytics initiatives? Executive Lessons on Modern Text Analytics, a new white paper prepared by: Geoff Whiting, principal at GWhiting.com and Alesia Siuchykava, project director at Data Driven Business provides some insight. Contributors to the lessons shared in the report include Ramkumar Ravichandran, Director, Analytics, at Visa and Matthew P.T. Ruttley, Manager of Data Science at Mozilla Corp
One of the interesting points made in the paper is that text analytics can be applied to many use cases: customer satisfaction and management effectiveness, product design insights, and enhancing predictive data modeling as well as other data processes. But at the same time, a takeaway is that it is better for text analytics teams to follow a narrow path than to try to accommodate a wide-ranging deployment. “All big data initiatives, and especially initial text analytics, need a specific strategy,” the writers note, preferable focusing on “low-hanging fruit through simple business problems and use cases where text analytics can provide a small but fast ROI.
NEXT PAGE >>