Version 1.4
Introduction
Othot is pleased to announce release 1.4 of the Othot platform. The release primarily focuses on the new Othot Sensitivity module and it also adds several new usability features and data science innovations.
If you have any questions regarding the contents of this document or require additional information on any new feature or change to the platform, please contact Othot Support at support@othot.com.
Sensitivity
A new module named Sensitivity has been added to the platform. This module provides customers with a very powerful analysis tool to determine likelihood changes based on variations of What-if variables. Results can be displayed at the individual level or can be aggregated based on predefined ranges of features. Results can be exported to Excel in pivot format.
More details and training on how to use the Sensitivity module will be provided to customers who have access to it.
Repredict/Retrain Option
Data modelers can now choose to re-train or re-predict any HIQ with the click of a button. This enables customers to take advantage of new platform additions, HIQ configurations or alternate algorithms without having to download and re-upload data files.
Filter Tool Tips
The platform now supports tool tips that are displayed when a mouse hovers over any green filter box in the application.
Forgot Password and Secret Question Admin Option
Othot customer administrators can now send their users a reset password and secret question email in cases where users cannot recall their secret question answer.
Improved Most Impactful Feature
The most impactful feature display has been improved to show more accurate identification of the maximum feature that has impact for the individual likelihood. The features displayed are set to show maximum positive or negative impact based on the HIQ configuration.
Extended Correction Factor Features
Othot has expanded the flexibility in applying its proprietary method for correcting predictions to give more accurate aggregate total predictions. Othot can work with customers to apply the correction factors for aggregate visual views only. Corrected likelihood scores can be shown in the grid along with the uncorrected likelihood scores.
What-if Warnings
Mass What-if results can now be configured to show warnings down to the individual when the What-if condition results in a business condition outlier. Customers can work with Othot to define the business conditions applicable for each What-if feature.
Feature Impact Improvements
The platform has improved the generation of individual feature impacts in a way that is consistent across all algorithms. The presentation of the impacts is now done with icons. Icon counts convey magnitude and icon colors (green or red) convey direction. Counts are determined using quartile logic.
User Activity Reports
Othot customer administrators can export to Excel their user login activity within each division.
Improved Data Cleansing
The Othot platform and data science engine have implemented additional data cleansing steps to account for data that may have unexpected formatting, including more extensive special characters and line formatting.
Improved and Expanded Integration
Release 1.4 has added more APIs to extract task data from the platform, including task APIs to the Othot scheduler. Using these APIs, customers can now identify task owners as systems and assign to the task owners results of Optimization or What-if operations. The integration supports task completion. External systems can tag tasks that have been consumed so they are removed from the task queue.
Data Science Engine
Underlying the Othot solution is the powerful Othot Data Science Engine. The latest version of the Othot Data Science Engine is part of release 1.4. This engine expands the algorithms available for use in the platform. These additional algorithms provide even better performing predictions and prescriptions for customer lifecycles. This engine also delivers a new, more consistent method for calculating feature impacts for individuals. This will give customers better insights in to predictions and prescriptions. Many additional refinements have been added for quality improvements, performance improvements and issue identification.
Quality
Release 1.4 improves the overall quality of the Othot platform by addressing several functional and usability issues reported by our customers and the Othot team. One major aspect of these improvements is the continued expansion of the Othot platform on the Google Cloud infrastructure. Othot continues to leverage the power and scalability of Google Cloud to provide the most scalable, secure and reliable solution on the market.
Next Release
The Othot development team has already begun working on the next release, 2.0. This release is scheduled to go into production at the end of Q1 2018. For more details about the features planned for the 2.0 release, contact Othot Customer Success or Othot Sales.