Emotional sentiments expressed in employee electronic communications could provide warnings of an imminent threat from an insider to the organization's critical data and systems.

New "sentiment and linguistic" systems use algorithms to analyze the content of email, text messages and other communications to determine if users pose an insider threat to their organizations. The systems count the number and types of words used in an electronic message that could help pinpoint an individual who's likely to conduct malicious activity against an organization.

"The way people write email messages ... might indicate someone has the propensity to be a spy against an organization or commit some type of fraudulent activity motivated by financial gain," Randy Trzeciak, technical manager of the Software Engineering Institute's CERT Insider Threat Center at Carnegie Mellon University, says in an in-depth audio report (click on player above to listen).

In this audio report, you'll hear:

Trzeciak discuss the need for organizations to identify critical assets and who has access to them; Scott Weber, managing director at the risk management firm Stroz Friedberg, and Michael Bruemmer, vice president at Experian Data Breach Resolution, address the sharp increase of insider threats; Chris Bowen chief privacy and security officer of security firm ClearDATA, explain that insiders aren't just employees on an enterprise's payroll; and Robert Leithiser, chief software architect at artificial intelligence software firm Authentic Intelligence, describe the role machine learning and behavioral analysis systems perform in identifying insiders who present a threat to the enterprise.

Analyzing emails and text messages for sentiment, language and behavior is critical because that's how most workers communicate, Weber says. "Companies' employees are spread out or work remotely, and we've all become addicted to our iPhones and our Samsung's. It's much quicker to fire off an electronic communication than it is to set up a meeting and sit down and talk face to face."