Guest Blog by Jon Woodward, CEO of NeutrinoBI
I’d like to start this post by thanking Michael Hiskey for his excellent article on Big Data a couple of weeks ago. If you haven’t read his take on why the Era of Big Data equates to storming the castle with data freedom for all then I recommend you check it out.
In the article, Michael discusses the importance of making it easy for the masses to get at their data in the way that it’s needed – a point that I’d like to explore a little further…
A recent Analyst Insight paper by the Aberdeen Group opens with the statement: “Imagine the World Wide Web without search. A mind-boggling, almost laughable proposition. And yet for generations that is how business intelligence has worked.” The paper goes on to provide evidence that “just as search made a rapidly growing unwieldy public internet useful for millions of people, so it promises to do the same for analytics.” And the tool that instantly springs to mind when we think about Internet searching is a good place to start when examining how to make it as easy as possible for people to discover data and information.
A question that I like to ask people is ‘have you ever had to explain to someone how to use Google?’ Since 1998, Google has been revolutionizing the way we search the Internet for answers. The inviting Search Bar and the ranked set of relevant results is a best-practice lesson in interactive simplicity.
It’s a lesson that can also be applied to searching structured data stored in warehouses, databases, line of business applications, and spreadsheets in the business context.
Over the years, multi-dimensional and faceted search have provided users with an increasingly sophisticated way of combining text search with a progressive narrowing of choices along predetermined paths to generate answers to queries: Useful, but not exactly as easy or as fast as a Google search.
At NeutrinoBI, our algorithmic breakthrough – Freeform Search – provides freedom from these facets, enabling users to question all their data sources in the same way as searching the Internet. Free text search strings typed in natural language can now be intelligently parsed and processed to return ranked, graphical results. By creating an ‘information context’ at the data mapping stage of solution implementation, any question typed by users can be understood – and temporal meaning, synonyms, and pseudonyms, interpreted. Autocomplete and autosuggest functions have made it even quicker and easier to type questions that find relevant answers.
Future developments in the area of ‘search’ will focus on smarter metadata analysis, including the wider context of questions and the location data from which they are asked, to provide even smarter insights. Ultimately, any question will be asked of a universe of data by harnessing all sources, where unstructured big data is just another one of those sources – and answers will be returned as quickly as they can be typed.
Whilst ‘google-like search’ is key to the ease of use – ease of analysis and data visualization is just as important. Which is why NeutrinoBI has partnered with Kognitio, to support intelligent search bar analytics with an in-memory analytical processing layer, which enables a zero disruption approach to existing systems.
It’s also why we’ve combined our search technology with all the precision and engagement of a gaming engine to deliver seriously smart data visualization capability with fully interactive gesture-control. Our Technical Director Eugene Ang has written a great article on this next chapter in the story of ‘The Beautiful Interface’.