Version 1.2
Introduction
Othot is pleased to announce the release 1.2 of the Othot platform. This release adds significant new functionality to the platform as well as improved quality, performance, and security. This document details the changes, additions, and unseen improvements to the platform.
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.
Optimization
A new module named Optimization has been added to the solution. This module provides a very powerful genetic algorithm-based optimization engine and interface. Using the optimization module, our higher education customers will be able to develop objective scenarios such as maximizing enrollment or maximizing class quality while selecting decisions such as financial aid awards or marketing program inclusion. In addition to the objectives and decisions, customers can establish constraints such as a total amount of aid range to allocate or allocation of enrollment to specific states or ethnicities.
For example, an institution with a prospect pool of 75,000 names who wants to Optimize the ROI on their marketing budget could build the following scenario: set the goal to “maximize enrollment,” build in restraints such as, “I’m planning to print 15,000 viewbooks, 15,000 visit pieces, and 5000 each of a 6-postcard series. Which prospects should I send any or all of these marketing pieces to in order to achieve maximum impact, and what’s the total enrollment that will yield?”
Optimization outcomes are presented in the insights grid and customer can export the outcomes to share with teams or systems to execute the optimization decisions.
Customers interested in utilizing the Othot Optimization solution are encouraged to reach out to their customer care representative or sales contact to schedule a session to learn how to effectively use the Optimization module.
Data Management
Othot customer instances now have a secure location to share data files with their internal data team and with the Othot Data Science team. Customers can upload files, send notices of these uploads to the Othot team, and subsequently remove these files once they are no longer needed. The Othot Data Science team has access to these files and can use them to enhance or create customer models and prediction data sets. These files are uploaded securely and stored at rest securely so customers can be confident that their data is always protected as they more openly share powerful predictive data with the Othot team.
Grid Aggregation
Customers can now view aggregated data in their Insights grid. The grid now enables average, sum and count aggregation tools on a filtered set of data --- for example, a user may filter down to all the prospect students in New Jersey, and see values such as average GPA, average SAT, etc. displayed in a new data bar at the bottom of the grid, along with a total of all the financial aid already allocated to those prospects. The aggregation view can be enabled through the grid feature selection and a single feature can be displayed as a count, average and sum simultaneously. Customers can work with their Othot customer care or sales team to enable aggregation for the specific features in their HIQ solution.
What-If Action Column
An additional column is now displayed as part of the “What-if?” and Optimization output that shows the new action appropriate to the prospect’s updated “What-if?” likelihood and status. This column can be exported and shared with a customer’s team or systems with the resulting “What-If?” or optimization action word.
Groupings
Customers can now add a new “Grouping” column to their insights grid. The Grouping column enables customers to find similar students based on their predictive outcomes. Students in the same group have very similar impacts and are likely to respond the same way to any “What-If?” outcome.
Using the Groupings function, an institution can identify a single student that would respond very positively to an action such as a grant award or marketing campaign, and then use that student’s grouping to find all other students in the grid that are likely to react in a similar fashion.
Extended Performance Metrics
The Othot Data Science engine accuracy and accuracy objectives are made up of three critical performance measurements. These measurements are the Individual Accuracy, Aggregate Accuracy, and the Rank Accuracy. High performance in each of these areas is critical in achieving predictions and prescriptions upon which customers can base business decisions. With release 1.2, Othot now displays each of these individual performance metrics to modelers in the Model page for each model.
Other Enhancements
This release adds several enhancements to the Othot platform functionality. The following is a list of the most significant enhancements included in this release:
- Feature-Impact Values
The individual display of the feature impacts now displays the value for that feature – e.g., if SAT score was a feature impact, the student’s SAT score is displayed. Customers can now see the most impactful features that illustrate the why behind the prediction and the value for that feature. - Visual Filters
When a user applies a filter to the prospects in the grid, visual filter boxes are now displayed above the grid interactively to indicate which filters were applied. Customers can easily remove filters by clicking the ’x’ on the filter display box. - Lifecycle Individual Display
The individual data shown in a student’s individual record screen can now be modified by predictive lifecycle. This means the individual record pages for prospect-stage students whose data is more limited can be supplemented with external data supplied by Othot, such as the census data, while an admitted student’s record page can show the detailed data that is typically available to the institution at this stage such as SAT, class rank or visitation status. - Additional Build Display
Customers that have the role as modeler will now see additional details when a build operation is submitted. This data includes the file name and the date and time the build operation was submitted. - Enhanced Configuration Features
Many additional features were added to the administration portal for Othot administrators to more effectively and efficiently manage customer instances. Although these features are not readily visible to customers, they do enable the Othot team to be more responsive and effective in assuring customers are utilizing the best and most impactful aspects of the Othot solution.
Data Science Engine
Underlying the Othot solution is the powerful Othot Data Science engine. With release 1.2 Othot is releasing the next version of this engine in the platform. This version improves performance and speed, and enables additional predictive algorithms that expand the options that can be employed by Othot to solve customer High Impact Questions (HIQs). This version of the engine also introduces the Optimization engine described in the Optimization section of this document.
Performance and Quality
This Othot platform release improves the overall quality of the platform by addressing several functional and usability issues reported by the Othot team and customers. One major aspect of these improvements is the migration of the Othot solution to the Google Cloud infrastructure. Othot now has the power and scalability of Google behind the scenes to assure that Othot provides the most scalable, secure and reliable solution in the marketplace. There are also several areas of performance improvements in the application core that will result in an overall enhanced user experience.
Next Release
The Othot development team has already begun working on the next release, 1.3. This release is scheduled to go into production in late July 2017. More details about the features planed for the 1.3 release can be provided by your Othot Customer Success or Othot Sales contact.