Wednesday, October 28, 2009

Book of Odds: A Semantic Knowledgebase of Probability

When I think about quotes related to probability, I can't decide which one of these is more profound or I like more:
1) "Probability is the very guide of life" by Marcus Tullius Cicero
2) “Once you eliminate the impossible, whatever remains, no matter how improbable, must be the truth.” by Sherlock  Holmes
3) "Million to one odds happen eight times a day in New York" by Penn Jillette

I am always tempted to choose number three because it is more funny and  if you have lived, worked or visited NewYork city then you can relate to it. But this blog is not about quotes related to probability! It is about Book of Odds: World's first reference on the odds of everyday life or a Knowledgebase of Probability built using Semantic Technologies. It is not another search engine or "yet another knowledgbase" as I explained in one of my earlier blogs. It tells you odds about things which you care about like "odds of surviving in a plane crash" or  "odds of having a heart attack if you are of certain age and ethnicity". Odds, Probability and Chances are used interchangeably in this context -- it is basically the ratio of favourable outcomes to total outcomes. The world had  references like google, wikipedia, Webster Dictionary, encyclopedias and so many other for  almost everyhting but didn't have "Book Of Odds" to understand the probability of an event or an occurence - so the founders had a good business case to start this initiative.

It seems that the project started almost three years back and millions of data points were collected by the team of very talented researchers. Once data have been collected, analyzed, and the odds have been computed, every Odds Statement then undergoes a rigorous quality review, is modeled ontologically (or in simple words by using semantic technologies explained throughout this blog), and finds its place in underlying semantic database of probabilities. The Semantic middleware is provided by Cambridge Semantics, a Semantic Technology company, mentioned in one of my earlier blogs, which structures the data, adds metadata and other information that helps users find the relevant odds statements using simple search. I consider it a very creative use of applying semantic technologies to the problems which could not have been solved using traditional database management systems. This proves again that a typical DBMS is not the answer to all the storage and query related problems! The semantic technology also allows it to find interesting and unexpected connections. It is not only a fun site but aims to be a respected new reference source to increase the tolerance for uncertainity and help people to deal with tough decisions.

To start using Book of Odds, you can browse by topics like health and illness; accidents and death; relationship and society; and daily life and activities. It is understood that it will always remain work in progress as this planet is becoming more complex and eventful everyday. There will always be  new areas to add and grow this reference.

My limited intearction with the website tells me that content is of high quality. I will recommend readers to at least give it a try and encourage this team. The most important thing for "Book of Odds Inc." is to have more people registered and start using this knowledgebase. I will "not" keep on pressurizing the founders at this stage by asking about business models and revenue stream.What business model did google have for many years? I am sure that they know that these questions are going to come up repeatedly after an year of their launch.  If  they will keep on producing high quality content or probabilistic ratios in this case and keep it meaningful for people then good things will happen. They do need to think creatively about advertising and marketing campaigns. They shouldn't be perceived as another google killer! Atleast they were smart enough to not use "engine" in their name. A business model will follow if they can attract more users and keep them interested.  My only recommendation to the founders will be to start thinking about combining their work or knowledgebase with various forms of predictive modeling or business analytics in the business world. Who knows better than companies in this economy that  "By looking at the past we can come to a greater understanding of the world and what is forever an uncertain future!"

No comments:

Post a Comment