White Papers

Accuracy Metrics for Entity Extraction

Kelly Enochson, PhD; Gregory Roberts
Rosoka Software, Inc. 950 Herndon Parkway, Suite 370, Herndon, VA 20170

Published January 2017

Abstract

Entity extraction software is typically evaluated based on the widely-accepted accuracy metrics of precision, recall, and F-measure. These metrics are certainly useful but limited in their scope. Additional factors including the types of errors, the cost of different error types, the facility of making changes to the system, and the efficiency of the system compared to human tagging should also be incorporated when evaluating entity extraction software. This paper illustrates the need for these additional factors and demonstrates how they can be implemented in evaluation.

Read Full Text

 

Analyzing Newswire and Social Media Data Using Multi-Vector Sentiment Analysis

Kelly Enochson, PhD; Gregory Roberts; Michael Sorah; Jamie Thompson
Rosoka Software, Inc., 950 Herndon Parkway, Suite 280, Herndon, VA 20170

Published September 2016

Abstract

The vast amount of written text available on the Internet provides a treasure trove of information for intelligence and security analysts, but only if the useful data can be quickly identified among all the irrelevant information. Many analytical tools available provide information about the sentiment of written texts; however, these tools typically utilize only one measure of sentiment in their metrics. Rosoka Software leverages psycholinguistic research across multiple sentiment vectors to provide precise information about an author’s language and pinpoint documents that do not follow predicted patterns. Rosoka’s multi-vector sentiment analysis uses four metrics to help analysts identify outliers in their data, recognize heightened emotional language, sort data by media type, and subset large data sets into only the documents that require further assessment.

Read Full Text

 

Parserless Extraction; Using a Multidimensional Transient State Vector Machine

Michael Sorah
Rosoka Software, Inc., 950 Herndon Parkway, Suite 280, Herndon, VA 20170

Published March 2016

Read Full Text