Clarabridge Blog

Last week we watched Watson, an IBM super-computer, beat not one but two Jeopardy champions simultaneously at their own game.

What especially intrigued me most is the way Watson does NLP, as it is similar in philosophy to how we do it here at Clarabridge. You see, Watson doesn’t just use one approach. The boffins behind the machine could have fought religious NLP geek wars and said “we will only use statistical NLP” or “we will only use machine learning” or “we will only use full rules-based, triples/case-frame NLP” or “we will only use search” or “we will only use neural networks.”

We have gathered here today to remember our old friend, the customer data warehouse. The customer data warehouse has served us well over the years, consistently bringing us consolidated analytics across all the data we had about our customers. It toiled day and night to ETL our data, and to respond to our increasingly lengthy multi-pass SQL queries trying to tease out insights.

Today is a very exciting day for us at Clarabridge, we have just announced the launch of the beta of our Self Service text analytics offering! Self Service is a great way to get started with text analytics.  Although we have some of the best sales and professional services people in the world, sometimes you just want to try something out on your own, or perhaps your project or company is too small to justify a standard software evaluation process.

A week ago I was in the MidWest demoing some of the newest capabilities from Enterprise 4; next week, I will be starting a journey that when complete will span 4 continents. Sometimes it feels like I’m in the George Clooney movie “Up in the Air,” well at least without the part where he fires people all day. But, THIS week shows the result of an even more remarkable journey –I’m here at our user conference, Clarabridge Customer Connections (C3) link in Orlando, FL.

Trade-offs are a part of life. I, for instance, used to only eat half of the Oreos® in the cookie packs in our break room. I’d rather do that than put extra time on a treadmill. In classification, there is an inherent trade-off between Recall and Precision, and I’d like to provide some background on those terms and on why the trade-off occurs.

Today we announced the GA release of our 4.0 version. We’re excited about the new functionality in this release, but even more excited about the great collaboration with our customers and partners that occurred to bring it about. Almost every new feature in 4.0 was the result from a customer at some point saying “wouldn’t it be nice if I could do X...” Working with our customer steering committee of 10 of our largest and most vocal (in a good way) customers, we worked to decide which of these ideas would provide the biggest additional business value to the users of Clarabridge. 

One of the highlights in this new version is the ability for authorized end-users of Clarabridge to directly upload their own data into their system straight through their web browser. What we found working with many of our large enterprise customers is that while the bulk of their verbatim came from their CRM and survey systems, that are typically directly connected to the Clarabridge server, they would often have users who would...