You are here

Advancing Natural Language: Semantic Analysis using Semantic Parsing

Submitted by admin on Mon, 04/02/2012 - 16:00

Semantic parsing is the process of extracting meaning and intent from natural language text, which can often come from voice interactions. Well performed semantic parsing can also enable machines and software to learn by reading.

Traditional approaches to semantic parsing have been supervised: that is, humans provide a set of well-known input and output to the problem, and the machine "learns by example". But this has several problems, most obviously the way the machine will only learn as well as the examples teach--if you give the same software that performs great on the example set some natural language about a totally different topic, it wil often fail to perform adequately.

Unsupervised semantic parsing is relatively new but has been shown to outperform supervised approaches by three times on a regular basis. It relies on the key idea that the names of actions and object can be learned. This involves some self-learning--by providing a wide variety of examples about the same topic, the system is able to find similarities and infer that they're talking about the same subject, and learn based on it.

Newer semantic parsing systems are inherently probability-based. Meaning they understand what you're saying within a range of confidence. This isn't the same as hearing you wrong, as when you tell your phone to find the nearest steak house and it finds the nearest take out. That is bad voice recognition. This is where you tell Siri to find the nearest steak house and it thinks you means the nearest steak that can be bought near your house.

Department Name: 
Cognitive Data Visualization and Comprehension

Data Stories

How to increase sales by looking at your customer data

Are you a B2C company interested in increasing sales? If you have some form of customer data, you may be in luck.

IBM's Watson and the future of Healthcare Analytics

What would it be like to have a doctor who’s always up on the latest research and has learned about treatments from over 1.5 million previous cases? It would look alot like Watson, IBM’s Jeopardy! playing supercomputer that’s getting ready to roll out with an all new look and a Memorial Sloan Kettering Cancer Center education in oncology.

How Facebook's Graph Search will affect Google, Technology, and Privacy

What has been both feared and expected is finally on its way: Facebook is building a better search; they're opening their vast stores of user data and giving us the ability to discover what’s inside. Lars Rasmussen, the mind that brought us Google Maps, is now hard at work creating Facebook’s new Graph Search, and from the looks of it, it’s going to put unprecedented power in the hands of its users.

How Big Data and Analytics will Change Society.

The mission statement of most police departments includes something like this, “our goal is to increase public safety, prevent crime, and protect human life.” With sufficient records of criminal behavior and analytics tools like Crimespotting, Cities can have the ability to predict when and where crimes are likely to take place and dispatch accordingly.

The Big Data Revolution

Want to learn more about how big data is changing business and how you can take advantage? Pick up our latest book: The Big Data Revolution.

"With everyone talking about Big Data and Data Science, its tough to know whom to listen to and how to make sense of it all. This book cuts through the noise and presents the reader with a roadmap for success. Whether you've been in this space for a while or are just coming up to speed, Secrets of the Big Data Revolution is a must read."
-Chris Crosby, CEO of Inflection Point Global


Trend Watch

What Higher Education Teaches us about Data-Driven Customer Retention

Rio Salado Community College is currently optimizing their student retention through focused testing and they are finding some truly telling results. We can learn a lot about customer retention and segmenting by studying what they've done.

The Growing Collaborative Consumption Market

Remember the Big Data mantra? how Big Data will enable us to better understand everything, reduce waste, and improve efficiency? Well honestly, without concrete examples it fades into the mass of voices shouting about how great the world is going to be. So lets take a moment to talk about collaborative consumption and it’s implications.

Andreas Weigend on the Future of Social Data

Dr. Weigend, Stanford Professor and former Chief Data Scientist at Amazon, tells us about social data and it's implications.

Big Data and Government Transparency

A wealth of government data is available to us today on .gov sites and private sites across the web. If we analyzed this data properly, we could build a rich understanding of how our government works and how it could be improved. But as the big data challenge dictates, the chokepoint is consumption.

Commentary

Gartner splits the 2014 Business Intelligence Magic Quadrant in two.

In an interesting turn, Gartner decided this year to split the annual Magic Quadrant for Business Intelligence and Analytics Platforms.

New Gartner Magic Quadrant: Advanced Analytics Platforms

The big story this year is how Gartner split the Business

How RoboCharm is using data to optimize customer interactions

Let’s face it, the ultimate goal for any use of data is to drive profits, and more often than not that comes back to learning how to enga