Browse Prior Art Database

System and method to calculate multi-dimensional state based recommendations for targeted customers

IP.com Disclosure Number: IPCOM000250179D
Publication Date: 2017-Jun-08
Document File: 4 page(s) / 97K

Publishing Venue

The IP.com Prior Art Database

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 28% of the total text.

TITLE: System and method to calculate multi-dimensional state based recommendations for targeted customers ABSTRACT: We are proposing a multi-dimensional state based recommender system based on the optimal user state for providing the recommendations. We are suggesting calculation of differential pricing discount for targeted customers on run time. There are a lot user state dimensions which can be utilized to optimize upon the concept of optimal state recommender solutions, where state can be a multi-dimensional concept ranging from time, location, biometric, acceleration and other attributes readily available from IoT devices and other real time sources. DETAILS: The Recommender systems have made good progress driven by e-Commerce providers. It brings significant benefits in terms of the sale conversion, revenues, customer experience, loyalty and lifetime value achieved by using advance analytics and recommender systems. There are different types of recommender systems ranging from memory based, collaborative filter based, association mining, SVD and others machine learning algorithm based recommender filters are present in practice and does a fairly good job in ascertaining which item may be preferred by a given user, or otherwise which user may be most likely to purchase a given item. But does a customer's propensity to buy a recommended item remain same at every moment (time dimension) and every state (multi-dimension user state including time)? Probably not. Can we increase the probability of a customer accepting a particular offer by sending a recommendation when he is most likely to- 1) Share his focus of attention (share of mind) to the provided recommendation (through mobile or other push enabled recommendation medium), and 2) To desire or have an urge to purchase the underlying item at the provided offer (stateful impulse for the product- offer combination) for the underlying recommendation? Probably Yes. For an example assume an e-Commerce grocery player A want to increase his market share by means of offering relevant products at a good discount to customers, and has been sending mail/ SMS/ app based notification of discounts to the customers, which the customer after some days has stopped even bothering. Whereas a player B, in the same business, and market issuing our suggested multi-dimensional state recommendations and sends a discount on a desired health drink just after the customer finishes an extraneous workout, and is able to communicate and relate and the discount offered with the quantum of work out. Player B, in this case enjoys the opportunity to change the habits positively of his customer, and get sustainable share of mind of the consumer as his notifications are tied with user’s behavior. Similar concepts may be applied to multiple other businesses, for example a travel insurance provider may instead of sending the recommendations naively, can optimize the purchase probability by sending the recommendations when th...