“Houston, We’ve Had a Problem”–That’s what Apollo 13 astronaut Jack Swigert famously said to Mission Control on April 13, 1970. It’s dramatic, but as a description of a problem, it’s not very useful. If I had been in Houston, I would have been like “okaaaay, what is the problem exactly?” Without more information, you couldn’t be of much help to the crew.

Figuring out a way to get a specific, useful statement of a problem was one of the central challenges in the early days of NetBase. The original idea behind the company was to scour the web for problems and find technologies to solve them. To do that effectively, you need to define a problem precisely. So I was trying to answer the question, How do you create a structured representation of somebody’s problem? That’s what brought to mind the Apollo 13 quote. What if the transmission had been cut off at that point in the conversation between Apollo 13 and Houston? Engineers on the ground would have had no ability to help the astronauts at all. But we know they did help them, so I researched the incident further and found out that Astronaut James Lovell immediately followed up with a much more useful statement: “It’s a main B bus undervolt.”

Even if we don’t understand what a “main B bus undervolt” is, we can tell that this follow-up has a lot of rich information about the problem; enough, in fact, for Houston to start helping the crew.

Looking at those two statements made me realize that not all sentences about a problem have the same informational value. A market researcher trying to understand needs is going to find a sentence like the second one much more informative than the first. And that’s when I realized that our technology needed to be able to distinguish between these two types of sentences. We couldn’t simply do basic sentiment analysis. Why not? Because sentiment analysis would look at both sentences and conclude that they both say something negative—the first because it has the word “problem” in it, the second because it has the word “undervolt” in it, which you can infer has some kind of negative connotation.

More research brought us to the concept of frame semantics, which showed us how one can distinguish between these two types of sentences. Frame semantics has developed a wide range of frames to represent concepts, such as the concept of a need, which has several components, including who the sufferer of the problem is, when the problem occurred, and how the situation differs from the ideal situation.

That last component really defines the problem. If you can identify how a situation differs from the way it would be ideally, and subtract the ideal state from the actual state, what you have left is the problem. “Undervolt,” for example, identifies the problem by specifying that the voltage ideally would be higher than it was. That term probably gave Houston a better technical understanding of what the crew’s problem was—one they couldn’t get from the first sentence.

The Apollo 13 quote thus triggered for me an understanding of why NetBase would need a technology like frame semantics to specifically define problems—albeit problems with a lot less at stake.

NetBase Product Line

Premier social media analytics platform

Tailored platform for growing businesses

Expand your social platform with LexisNexis news media

Power of social analytics for your entire team

Customer experience analytics platform

AI, Image Analytics, Reporting Tools, APIs & more

Product configurations to meet all needs

Quid Product Line

Media analytics and market intelligence platform

Enrich your media analytics with social data

Media coverage for historical & real-time monitoring

AI algorithms, NLP, data sources, and visualization

Tailored, configurable solutions