A picture can paint a thousand words, the saying goes. But unless we know how to decode its message, a picture can also be difficult to read.Â Words, on the other hand, usually have a very direct meaning and are simple to understand. But critical messages can be lost if they are not used in an efficient way, and structuring them efficiently takes time and effort.
Business Management Dashboards And Business Analytics
In Big Data analytics, reporting the insights weâ€™ve gleaned from analyzing large amounts of messy data sets is the crucial â€œlast stepâ€ of the process â€“ and itâ€™s often a step which causes us to stumble. We may have crunched terabytes of data in real time to come up with our world changing revelations. But unless we can communicate them convincingly to those who need to take action, they are useless, and worse than that, a waste of valuable time and money.
This is why data analysts have come to rely increasingly on graphics and visualizations combined with text â€“ such as the now ubiquitous â€œInfographicsâ€ â€“ to get a message across. But infographics rarely tell the whole story, and are still generally issued alongside written reports or summaries, particularly if they have a corporate purpose and detail is required.
This is the reason that I am particularly excited about a new partnership announced yesterday between Qlik and Narrative Science. A free trial plugin now allows users of the Qlik Sense business intelligence platform to take advantage of the powerful natural language generation tool Quill to create automated, intelligent narratives from data and visualizations.
Iâ€™ve written about Quill, developed by Narrative Science before â€“ the tool is used across industries from finance to retail and even journalism to create detailed natural language reports from data. Quillâ€™s technology is ground breaking because it is able to structure written words in a manner that is â€“ very nearly â€“ indistinguishable from how it could be done by a human writer.
In a world where â€œdrowning in dataâ€ is a real danger that I see many companies falling foul of, the emphasis needs to move away from measuring and collecting everything we possibly can, to identifying what is truly important. As Narrative Science CEO Stuart Frankel tells me, â€œWhat weâ€™ve seen over the last 12 months is significant demand for embedding natural language capabilities into existing BI and analytic applications.
â€œIn part this demand is being driven by the need to make these applications easier for business users.
Visualizations are usually used as a layer of the top of data, designed to make the data more digestible. With Narrative Scienceâ€™s tools embedded into it, Qlik Sense now effectively goes a step further by adding another, textual layer, which breaks the visualizations down into natural language.
â€œWhen youâ€™re looking at a picture,â€ says Frankel, â€œthereâ€™s often a subjective determination based on the way that the visualization is built. So it still requires a skillset for the user to really understand and interpret and find out whatâ€™s important. And most importantly â€“ find out whatâ€™s actionable.â€
The procedure is remarkably straightforward and actually very impressive to see in operation. A user simply interacts with visualized data through the Qlik Sense dashboard and while they work will see, in real-time, a narrative write itself alongside the data under interrogation. For example, the narrative will change depending on the information you highlight or the type of data you visualize in different graphs.
NLP in dashboards and analytics
And it writes well, in simple, easily digested language, prioritizing what its algorithms determine to be the most essential, insightful information. While perhaps its prose is lacking in what real (human) writers like to think of as artistic flair or creative, individualistic style, it makes up for this with conciseness and clarity. And, to be fair â€“ who really needs to read poetry in their corporate dashboards?
The automated textual narrative will respond in real time as the user defines and refines their analysis, and will continue to relay new relationships and connections, as they are discovered. These are likely to be relationships which would not be apparent from the data alone.
â€œThese narratives are dynamic,â€ explains Frankel, â€œItâ€™s not just a static template experience â€“ pre-canned documents with different numbers filled inâ€¦ they are intelligent, analytics-driven insights that can really help someone go from a gut assumption, or a hypothesis, to a discovery.
â€œIt isnâ€™t always easy to identify relationships in data and it isnâ€™t always obvious what the trends are. Sometimes they are very subtle but important. Quill and Qlik addresses that.â€
With the ever increasing amount of information that businesses are finding themselves expected to deal with in the Big Data age, thereâ€™s never been a bigger need for tools to break down the communication barrier between humans and machines. I confidently expect that this sort of functionality will become a standard feature of all forms of BI, analytics and visualization platforms in the near future. Natural Language generation and processing is one of the key machine learning technologies powering the move to a data-driven society, and Qlik, with the help of Narrative Science, have just made it available to a whole new audience.