Abstract: Systems, methods and software products identify emerging issues from textual customer feedback. A message stream of customer feedback is received. The message stream includes a plurality of unstructured text messages from at least one homogeneous source. A time interval is established. The volume of text messages for the time interval is determined to establish a reference volume. The volume of text messages in subsequent time intervals is measured to establish a trend volume. The trend volume is compared to the reference volume to determine a volumetric change. At least one action is initiated in response to a volumetric change above a pre-determined threshold. At least one action is initiated in response to a volumetric change below a pre-determined threshold.
Type:
Application
Filed:
January 16, 2007
Publication date:
September 27, 2007
Applicant:
ISLAND DATA CORPORATION
Inventors:
Guy Jones, Scott Austin, Grant Foster, Eric Scott
Abstract: A concept recognition system includes a concept recognition training system and a real-time system. The concept recognition training system processes a training set and produces a lexical profile keyed to a target category. The lexical profile comprises a set of lexical cues, which are words and phrases associated with the target category. A trainer starts with an initial lexical profile that comprises a small set of seed cues. The training system retrieves samples from the training set that match lexical cues in the lexical profile. The trainer determines which of the retrieved samples are positive instances of the target category. The training system extracts lexical cues from the positive instances and adds new lexical cues to the lexical profile. The real-time system uses the lexical profile as the basis for making confidence judgments for each new incoming message from the same input stream with respect to whether the message is an instance of the target category.