Released on October 9, 2014.

New Features

Enrich

The new Enrich menu can be used to configure Search tagging and pipelets.

Any pipelet which is available to a user can be configured as an enrichment. The main pipelets that we currently have readily available:

• Unknown Entity Extraction (TextRazor): Uses the TextRazor API to do natural language parsing of the text and extract keywords.
• Known Entity Extraction: This uses Squirro's own technology to extract keywords from text. Those keywords are imported from the customer's data set, for example a list of suppliers or customers. Talk to your contact at Squirro to make use of this new feature. This will become a self-service offering at a future point in time.

Web Site Crawler

The Web site crawler data source is now readily available in the Add Data menu. Setting up such a crawl can take a few tries to get the white- and blacklists right. That's why there's a Preview Log available, which shows in detail what's happening behind the scenes based on the configuration.

Improvements

• Improve quality of PDF and Office documents text extraction
• Improve display of duplicate items (and renamed from "similar stories" to clarify the scope)
• Some improvements on the embedded PDF viewer

Bug Fixes

• Fix positioning and handling of search results in the result list
• Fix add topic action in the context menu
• Fix handling of new sources in the embedded iFrame
• Fix line charts on the "source" keyword
• Fix reset password action in an embedded dashboard