Browse Prior Art Database

Action response evaluation over time

IP.com Disclosure Number: IPCOM000242220D
Publication Date: 2015-Jun-26
Document File: 2 page(s) / 58K

Publishing Venue

The IP.com Prior Art Database

Abstract

In this invention, we first propose a technique for a baseline discovery based approach for action response evaluation, which further includes: a) integrating time series data before action and other profile information to generate most similar baseline samples, b) A technique for generating the action response score and confidence score based on 1) the pre-action time series similarity, and 2) the post-action time series similarity; We also propose a control group generation method (not completely all groups) by top-down aggregation samples. This component constructs the time series tag by clustering pre-action time series data.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 54% of the total text.

Page 01 of 2

Action response evaluation over time

Our invention is based on the following facts and business need examplified as follows : 1) Facts •Business analytics is valuable, especially when they can generate actionable advise •Actions based on analytics need a second-round analytics in a loop •This problem is especially pronounced in these examples as our projects •HR retention bonus based on staff/skill categorization -> retention bonus •Sales branch expansion profiling -> investment and budget •Sales pipeline lead quality categorization -> salesforce allocation rebalance 2) Business need
•Quantitatively evaluate the action's performance is important
•What is the return of interest (ROI) in a retrospective perspective?

•Shall I continue my action and trust the business analytics service from the vendor company?

•If the action does not generate 'good' results, what's the problem and other external facts not accounted in previous analytics?
•In summary, we define the problem as

  •Evaluate the performance of actions by comparing the time series pattern before and after the action period or time point The proposed invention is described as follows: •Advantage
•Compute a response score for each job title whose staff group received retention bonus
•The method is assumed to counter-seasonality and also counter the common external factors over 'similar' job title
•A confidence score can be also generated to indicate the reliability of our estimation...