Summary
The invention relates to a computer-implemented method for calculating the value of individual passengers to a passenger transport company such as an airline. A database of feature parameters and corresponding algorithms is used in combination with historical passenger information to train a machine learning model. Passenger information is subsequently input to the model to generate passenger values which can be used to inform financial and marketing decisions. Examples of the feature parameters are a variation tendency parameter, a bad experience parameter, a travel parameter and a booking parameter; the parameters may also include the number of passengers travelling together, social influence information and interest-related information. The invention is claimed as a method and as a machine-learning system.
The Hearing Officer agreed with the examiner-™s reasoning which followed the four stepAeroteltest to determine whether there was a technical contribution and applied theAT&Tsignposts as modified inHTC. The contribution was found not to provide the required technical effect and hence the claimed invention related solely to a method for doing business and a program for a computer as such. Thus, the application did not meet the requirements of section 1(2)(c) and was refused under section 18(3).
Full decisionO/191/21 211Kb