Behavioral and exogenous factor analytics based user clustering and migration

Patent metadata

Field	Value
Publication number	US-2016379323-A1
Application number	US-201514751210-A
Country	US
Kind code	A1
Filing date	Jun 26, 2015
Priority date	Jun 26, 2015
Publication date	Dec 29, 2016
Grant date	—

Abstract

A present risk aversion of a customer is determined. A temporal preference is detected using historical data related to the customer, the temporal preference showing a preference of current utility over a future utility of a product. Using the temporal preference and a negative transaction risk, a future risk aversion of the customer is projected at a future time. A pattern of offer acceptance by the customer is identified in the historical data. A value of an exogenous factor is determined on which a buying ability of the customer depends. The customer is classified in a cluster, where all customers in the cluster have the present risk aversion, the temporal preference, the negative transaction risk, the future risk aversion, and the value. An offer for a product is presented to the cluster, which satisfies the present risk aversion, and where a probability of acceptance of the offer exceeds a threshold.

First claim

What is claimed is:

1 . A method for behavioral and exogenous factor analytics based user clustering and migration, the method comprising:
determining, using a processor and a memory, a present risk aversion of a customer; detecting a temporal preference using historical data related to the customer, the temporal preference showing a preference of current utility over a future utility of a product; projecting, using the temporal preference and a negative transaction risk, a future risk aversion of the customer at a future time; identifying a pattern of offer acceptance by the customer in the historical data; determining a value of an exogenous factor on which a buying ability of the customer depends; classifying the customer in a cluster, wherein all customers in the cluster have the present risk aversion, the temporal preference, the negative transaction risk, the future risk aversion, and the value of the exogenous factor; and presenting an offer for a product to the cluster, wherein the product satisfies the present risk aversion, and wherein a probability of acceptance of the offer exceeds an offer probability threshold.

2 . The method of claim 1 , further comprising:
assigning, according to a probability distribution model, a corresponding initial probability of acceptance to each offer in a set of offers; presenting a first offer from the set of offers to a first portion of the cluster and a second offer from the set of offers to a second portion of the cluster; measuring a first rate of acceptance of the first offer by the first portion and a second rate of acceptance of the second offer by the second portion; adjusting a first initial probability of acceptance of the first offer to a first adjusted probability of acceptance using the first rate of acceptance of the first offer by the first portion; adjusting a second initial probability of acceptance of the second offer to a second adjusted probability of acceptance using the second rate of acceptance of the second offer by the second portion; determining that the first adjusted probability of acceptance exceeds the offer probability threshold; determining that the second adjusted probability of acceptance does not exceeds the offer probability threshold; and selecting the first offer as the offer for the product.

3 . The method of claim 2 , further comprising:
selecting a subset of offers wherein each offer in the subset has a corresponding initial probability of acceptance exceeding a cutoff threshold probability, the subset of offers includes the first offer and the second offer.

4 . The method of claim 2 , further comprising:
selecting, to explore acceptability in a cluster, a subset of offers wherein each offer in the subset has a corresponding initial probability of acceptance below a cutoff threshold probability, the subset of offers includes the first offer and the second offer.

5 . The method of claim 2 , further comprising:
selecting a sampling of offers wherein an offer in the sampling has an initial probability of acceptance, and the initial probability of acceptance is representative of a sample probability of acceptance.

6 . The method of claim 1 , further comprising:
determining, as a part of determining the present risk aversion, using the historical data, a product type to which the customer is averse, wherein the product is of a type other than the product type to which the customer is averse.

7 . The method of claim 1 , wherein the historical data is data of a second customer, the second customer having a second present risk aversion that is similar to the present risk aversion of the customer.

8 . The method of claim 1 , further comprising:
identifying, using the future risk aversion of the customer, a future product to offer the customer at the future time; and selecting the offer for the product by determining that a sale of the product and a sale of the future product together satisfy a profitability threshold.

9 . The method of claim 1 , wherein the exogenous factor is a social factor, wherein the social factor is indicative of a preference of a social group in which the customer participates.

10 . The method of claim 1 , wherein the exogenous factor is a professional factor, wherein the professional factor is indicative of an economic stability of a profession in which the customer participates.

11 . The method of claim 1 , wherein the exogenous factor is indicative of a risk in selling a future product to the customer.

12 . The method of claim 1 , further comprising:
evaluating a response of the customer to the offer for the product; determining that the response is different from a response expected from the customers in the cluster; recomputing (i) the present risk aversion, (ii) the future risk aversion, (iii) the pattern of offer acceptance, and (iv) the value of the exogenous factor to at least one of (i) a revised present risk aversion, (ii) a revised future risk aversion, (iii) a revised pattern of offer acceptance, and (iv) a revised value of the exogenous factor, respectively; migrating the customer to a second cluster, wherein all customers in the cluster have the revised present risk aversion, the revised future risk aversion, the revised pattern of offer acceptance, and the revised value of the exogenous factor.

13 . The method of claim 1 , wherein the method is embodied in a computer program product comprising one or more computer-readable storage devices and computer-readable program instructions which are stored on the one or more computer-readable tangible storage devices and executed by one or more processors.

14 . The method of claim 1 , wherein the method is embodied in a computer system comprising one or more processors, one or more computer-readable memories, one or more computer-readable storage devices and program instructions which are stored on the one or more computer-readable storage devices for execution by the one or more processors via the one or more memories and executed by the one or more processors.

15 . A computer program product for behavioral and exogenous factor analytics based user clustering and migration, the computer program product comprising one or more computer-readable storage devices, and program instructions stored on at least one of the one or more storage devices, the stored program instructions comprising:
program instructions to determine, using a processor and a memory, a present risk aversion of a customer; program instructions to detect a temporal preference using historical data related to the customer, the temporal preference showing a preference of current utility over a future utility of a product; program instructions to project, using the temporal preference and a negative transaction risk, a future risk aversion of the customer at a future time; program instructions to identify a pattern of offer acceptance by the customer in the historical data; program instructions to determine a value of an exogenous factor on which a buying ability of the customer depends; program instructions to classify the customer in a cluster, wherein all customers in the cluster have the present risk aversion, the temporal preference, the negative transaction risk, the future risk aversion, and the value of the exogenous factor; and program instructions to present an offer for a product to the cluster, wherein the product satisfies the present risk aversion, and wherein a probability of acceptance of the offer exceeds an offer probability threshold.

16 . The computer program product of claim 15 , furthe

Assignees

• IBM

Inventors

• Ezry Raphael
• Goyal Munish
• Mitchell-Jones Gareth J
• Pinchuk Steven G

Classifications

• G06Q30/0201

  Market modelling; Market analysis; Collecting market data · CPC title

• G06Q50/188 Primary

  Electronic negotiation · CPC title

• G06Q40/04

  Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange · CPC title