A new year has arrived, but our goal remains the same. We’re constantly improving the Re:infer Conversational Data Intelligence platform, extending its capabilities and making it easier to use across the enterprise.
Take a look at some of our biggest launches and improvements from the last quarter…
Dynamic Email Thread View
We’ve redesigned how emails are viewed in Re:infer. Previously, messages were displayed individually in our Explore function. But now users can expand the entire thread of messages that make up an email conversation with the click of a button. You can expand out the thread and see earlier or later messages, and easily label those other emails too when model training.
We’re very proud of our Dynamic Thread View feature. It’s something our customers have been asking for, and it’s giving them a whole new level of context on business communications. They can now quickly and easily analyse entire conversations with full visibility over every message.
Email Attachment Metadata
For email and case management integrations to Re:infer, we now display details of each email or message’s attachments. Users can view the file name as well as the type. When using the Re:infer API, you will also be able to see the attachment file size.
The new feature gives users more context and specificity when deciding how to label verbatims. It also helps our clients with use cases that incorporate some kind of document management or OCR tool.
We’ve changed how Dashboards work on the platform to help make them more useful and shareable.
Datasets now support multiple dashboards to give our customers more granular insight into particular datasets. Each dataset now has a default dashboard and each user can also create an additional one for their own use.
Dashboards can now also be shared via links with other Re:infer users in your organisation. They can view or edit these datasets themselves depending on their permissions.
This change is making it easier than ever to collaborate on Re:infer, helping users share their findings and insights across teams.
We’ve made some changes to our model to increase the consistency and accuracy with which Re:infer predicts entities.
By adding context to trainable entities, the Re:infer platform now uses the previous and next paragraph when learning and predicting the meaning of entities. This has greatly enhanced our model’s understanding of message elements like addresses and tables, and its ability to extract useful insights from them.
Work in progress
We’re working on some big updates for our customers that will really expand the applications of Conversational Data Intelligence and the capabilities of the Re:infer platform.
We’re adding brand new Quality of Service features that will help users gain more insight into the quality of the interactions customers are having with their brand. Messages will be analysed and a Quality of Service score given to each based on the context, intent and sentiment of the message.
Quality of Service will give customers an entirely new metric to analyse, allowing them to measure how the quality of the service they provide to customers has changed over time.
Closely related to this, we are also implementing a pre-trained Tone Analysis feature to the platform. Once enabled, users will be able to see at a glance how positive or negative a message is along a detailed numerical scale - and can choose how this will impact their Quality of Service score.