We are pleased to announce the release of an update for Derwent Data Analyzer. Version 12.1 adds several performance enhancements and stability improvements to the DDA suite. The highlights are outlined below.
Downloading and installing DDAv12
Details on upgrading from an earlier version can be found on our website in the upgrade FAQ.
Version 12.1 Release Notes (released December 2022)
- Added Dataset Identifier field to default import filters to differentiate between “old” and “new” records when combining datasets.
- Fixed issue where process fields were being added to the record view.
- Fixed bug in Remove Duplicates where selecting records by date wasn’t ordered properly.
- Fixed bug in Remove Duplicates where matching criteria was dependent on the order they were selected.
- Fixed issues with child fields and further processing commands breaking child history or improperly assigning processed fields as active children.
- Fixed bug in Cleanup Confirm window improperly highlighting terms as matched by thesaurus.
- Fixed bug resuming saved list cleanup when using a thesaurus during cleanup.
- Improvements to Refine NLP to remove more “junk” terms.
- Added field selection dialog for Concept Grid to limit to only fields that make sense for the algorithm.
- Fixed bug in Pivot Tool to properly count record counts for multi-valued fields when filtering.
- When importing categories in the Record Classification dialog, users will now be prompted to automatically assign records to those categories if they come from a field within that dataset.
- Allow user to bias terms as “most important” for the auto-classifier by assigning a keyword to a classification in the auto-classifier settings dialog.
- In Smart Trainer, prioritize English Only records first for training.
- Add Record Classification and Record notes to fields shared during “Record Share”.
- Fixed bugs and stability issues.
- Updated Context Sensitive Help (F1) for many dialogs and visualizations.
- Fixed alpha sorting for field names on some dialogs.
- After resizing, dialogs reposition to the center of the screen so buttons are visible.
- Updates to Chromium and Stingray Libraries.
- Force DDA window to the front after completing a script.
Version 12 Release Notes (released September 2022)
New capabilities for Find Records
- Search for records based on group membership
- Search for records based on record note text
Add To Group (new Dialog)
Accessed via Right click-> Add to Group in List/Detail/Matrix views and in Find -> Add to Group. This change adds the ability to type to filter groups, which is especially useful when looking for a specific group out of 50+ groups.
Combine Duplicate Records (new look)
More intuitive interface allows users to quickly identify which fields to match on.
Remove Duplicate Records (new look – new capability)
The new interface allows users to select multiple criteria for choosing which record to keep when removing duplicates. These criteria, which can be layered as a filter, include record length, dates, values, group membership, language, as well as specific terms. This flexible, user-defined filter enables customers to precisely specify which record to keep when removing duplicates
Split items into multiple items in List Cleanup Confirm
In List Cleanup Confirm an item can now be split into two. This is useful when an item contains two different items in one. E.g. “Georgia Tech and Univ of Georgia” can be cloned and added to both the “Georgia Tech” and “Univ of Georgia” sets separately, preserving record counts.
IPC-WIPO Green Codes
Adds a definition file for converting IPC Codes to WIPO's environmentally friendly technology categories.
Users can analyze a set of individuals (as well as other fields) using Network Metrics. The feature allows the user to select which field to use for the network calculation. The feature then calculates Betweenness Centrality, Closeness Centrality, and Degree Centrality. The feature also calculates a composite Keystone Indicator consisting of all three measures which ranges from 0 to 1. The higher the Keystone Indicator, the more networked an individual is within the search space.
- Betweenness Centrality – how often a node lies along the path between other nodes. (i.e. a lot of information flows through that node)
- Closeness Centrality – total distance to all other nodes
- Degree Centrality – how many connections does a node in the network have?
- Keystone - Harmonic Mean of the three centrality measures and the number of records.
Smart Trainer – add fields to Smart Trainer View
Enables user to select and display non-training fields in the Smart Trainer view to help provide context during training
Create a flow chart with up to 5 fields showing the cooccurrence between the fields. Records will cascade down from left to right. In other words, data in the columns will not contain any records that were not in set of records in the column to the left of it.
Limited to a max of 5 fields.
Clicking the link between two nodes in the chart will show the shared records between those two items in the data.
Create Concept Tree
This algorithmic approach to clustering concepts uses NLP phrases as an input. The algorithm then conducts a series of optimized clustering calculations to divide the topic into major areas. The algorithm then divides the major areas into minor areas. Finally, the algorithm visualizes the output, names the areas, and color codes overlapping minor areas.
Add Third Field to Bubble Chart (Bubble Pie)
Users can add additional data to DDA Bubble Chart by including a Pie Chart within each bubble
Export to Tableau and PowerBI
- Users can export content out of DDA to import into PowerBI.
- Users can export content out of DDA to import into Tableau.
Share Record and Email
Customers can use the flexible Record View layout to customize how records are displayed including which fields are shown (including images) and what order they are shown in, as well as highlighting. A record can then be shared with others as a PDF.