New Features

Platform & Studio

• New Datasets:

• Google Sheets – Ability to create, edit, delete Google Sheets dataset

• Microsoft SQL Server Analysis Services (SSAS) – Ability to use SSAS to create, edit, delete datasets

• MongoDB Connectivity – Ability to create, delete, edit MongoDB dataset

• Algorithms added:

• LSTM Forecasting Algorithm

• Annova Analysis –

  • One-Way Annova

  • Two-Way Annova

  
  

• Linear Discriminant Analysis (LDA)

• Quadratic Discriminant Analysis (QDA)

• Model Comparison – Ability to compare result matrices of multiple algorithms at a time

• Cross-Validation – Available as a sampling method

• AutoML Wizard – Wizard-like interface to use AutoML

• Customized functions for RubiPython –

• getReaderData – Ability to read any reader created in Rubiscape directly in RubiPython

• writeDataToTable – Ability to write data to a table using RubiPython

• writeDataToFile – Ability to write data to a file using RubiPython

• getUserLogsList – Ability to get a list of dates for which user activity logs are available

• retrieveUserLogs – Ability to user activity log for a particular date

• Ability to retrieve user activity logs in the Admin module

• Two Factor Authentication – An extra layer of security has been added by using OTP verification method (through email) in Rubiscape

• Hyperparameters Optimization – Ability to perform hyperparameters tuning for regression and classification algorithms

• Workflow-as-a-Task – Ability to use the workflow entity as a separate task which can also be used recursively

• Task Results as Parameter – Ability to store the results of a task and use them for any further operation

• Conditional Tasks – Ability to execute any task in a workbook/workflow using "If-Else" and "While" conditions

• Ability within the UI to add or remove tenants

• Oracle support for Template Table

• Model Validation

• User-Defined Parameters – Ability for a user to define their parameters/variables at a workbook/workflow level

• Access to Rubipedia provided from within Rubiscape

• Vulnerability and Penetration Testing completed successfully

Rubisight

• Hyperlink options are now supported for multiple columns for the Table, Cross-table and Sparkline charts

• Data Level Security – Filter the dataset columns in dashboards based on user login

• Calculated Fields – Ability to allow the user to add calculated columns in the dashboard based on either existing columns or mathematical expressions

• Conditional Formatting – Charts can be partially highlighted based on specific conditions set by the user on column values or a constant. Currently applicable to Column chart, Bar chart, and Line chart

• Switch Widget – User can switch between the type of widget in a single click based on the source and target widget configuration

• Horizontal and Vertical Tiling – The user can split the charts based on the selected dimension value. This is now applicable to –

  • Column chart

  • Pie chart

  • Area chart

  • Line chart

  • Treemap

  • Donut chart

  • Word cloud

  • Bar chart

  • Histogram

  • Pareto chart

  • Sankey chart

  • Boxplot chart

  • Stacked column chart

  • Stacked bar chart

  • Stacked area chart

  • Bubble chart

  • Scatter chart

  • Waterfall chart

  • Sunburst chart

  • Solid gauge chart

  • Map chart

  • Combination chart

  • Funnel chart

  • Bullet chart

  • Card widget

  • Google map chart

  
  

• New charts added -

  • Google Map Chart

  • Multi-Level Cross Table

  
  

• Refresh Cache for Dataset in Workflow – Ability to add a refresh cache node within a workflow to refresh dataset cache. The workflow can be scheduled to achieve a scheduled dashboard refresh effect

• Copy chart within a dashboard

• Improvements in layouts of Rubisight dashboard

Enhancements

Studio and Rubisight

• AutoML – Top 5 recommended models with parameters are available to the users. These models can now be published as a separate algorithm.

• Lazy loading implemented on exploring datasets

• Data Preparation Enhancements – When a user configures the data preparation node with or without a group-by option and executes any of the available tests, they will now get data for all groups from the group-by dropdown list in the data tab which can be carried forward to the successor nodes. This was a limitation in Rubiscape release 2.0

• Ability to connect multiple predecessors to a RubiPython node

• Ability to add multiple output variables to RubiPython

• Write console logs to files instead of terminal

• Revamp of the existing expression builder

• Support for all types of date formats while creating a dataset

• Improved iconography

• Improved interactivity/filtering performance (bugs resolved)

• Handled large result data scenarios for the following charts –

  • Column chart

  • Bar chart

  • Area chart

  • Line chart

  • Stacked-column chart

  • Stacked-bar chart

  • Stacked-area chart

  • Sunburst chart

  
  

• Improved features for Table, Cross-Table and Sparkline chart by adding lazy loading functionality

• Added feature to merge the same values in the Table chart

• External dashboard creation option is available on the create dashboard form. User can create normal as well as external dashboards from the same dashboard creation form

• Auto-sizing of widget added in "At a Glance" view