Dynamic price quotation with machine learning in Microsoft Dynamics 365 for Sales
Picture yourself scanning a travel site for the best deal for your long-waited flight to your dream holiday destination. A question pops into your head: Is the price going to be more convenient tomorrow? And here you are, hesitating for a day, and searching again the day after. And yes, the price is different, more likely it has increased!
What is behind flight price fluctuation, which occurs so apparently at random? Can we predict ticket price quotations and find the best deal? And moreover, can we introduce a similar model into our own Dynamics 365 for Sales solution?
My session at Directions EMEA 2017 will introduce the concepts of demand estimation in Machine Learning and generate an experiment for predicting the cost flotation of flight tickets using Azure ML. Demand estimation is a branch of Machine Learning called "regression." Regression is classified as a supervised learning method, meaning it is a learning task of inferring a function from labeled (i.e. known) training data. Microsoft describes supervised and unsupervised machine learning algorithms at this link, and when to choose one over the other.
In layman's terms, this means that we must define our initial dataset as a collection of records with specific columns, or "features," that define a price flotation for an airline ticket. The data model is defined in custom entities in Dynamics 365, and consists of:
- FlightTicket entity, to collect data about the ticket itself, like ticket number, airline, flight number and departure/arrival times.
-
TicketPrice
entitity, to collect data ...
FREE Membership Required to View Full Content:
Joining MSDynamicsWorld.com gives you free, unlimited access to news, analysis, white papers, case studies, product brochures, and more. You can also receive periodic email newsletters with the latest relevant articles and content updates.
Learn more about us hereor login