Natural Language Processing: An Approach to Assessing Named Entity Recognition

One of the problems faced in Named Entity Recognition is the extraction of accurate information from texts helping humans in their everyday processes. One such information extraction process is the text revisions that occur in some domains such as education and law enforcement. These revisions take time to do and are prone to errors. This work aims to decide on a final revision using the best classifications of parallel revisions of the same text made by one or more revisers. An important step is the creation of a metric that helps in the choice of ratings and generates a concordance value that can be analyzed. A bibliographic study of existing metrics that can help in the creation of the new metric is also done.