How Bayes Analytics can benefit companies in the Travel Industry

The Bayes Analytic engine is nearly ideally suited for use in the travel industry companies like who have a long term relationship with their customers and who need to continually deliver more value and save more time for their customers.    While the engine is optimized to determine the probability a given user or deal will respond favorably or meet a pre-specified goal the same feature can be reversed so it can calculate which offers or goals have the highest probability of success.

Quick Introduction:

As part of my current stock analysis and trading project I have developed a classification based inference engine which is having a high degree of success at selecting offers that can meet specific goals. I believe the same core engine could be applied to predict other things like which customers are most likely to respond favorably to a given offering or which products will sell best in specific cities. I would like to prove or disprove this thesis before offering the technology for outside users.

The engine works by using correlated data similarities between users to identify the probability of success if a similar offer was made to other users. For example: If you have a set of people who consistently rent a rooms at the Marriott who are traveling to a city where there is no Marriott the system can look at past similarities between these customers and determine that of 500 people who visited the same city the 12 people who most closely match the current customer chose 8 selected the Starwood and 4 choose the Hyatt so these two represent the best chance of satisfying that customer.

After training with historical data the system should be able to predict user response to more recent offers. Success would be measured based on comparing the output to offers Expedia actually made. Eg: If Expedia made an offer to 300,000 customers with a 8% success rate and the system was able to statistically select a set a subset of the 300,000 customers where there was a 25% success rate then I will call the experiment a success.

The benefit to Expedia is that if the experiment works they will know it is feasible to select offers for customers with a higher chance of success which could improve customer satisfaction and revenue. If the results are promising I would also give Expedia first chance at exclusive access to the tech for their industry.

Thanks Joe Ellsworth
phone-number-for-bayes-analytic

Leave a Reply