Optimizing Selection of Assessment Solutions for Completing Information Extraction Results

ncomplete information has serious consequences in information extraction: it increases the costs on the one hand, and leads on the other to problems in downstream processing. This research work focuses on improving the completeness of extraction results by applying judiciously selected assessment methods to information extraction within the principle of complementarity. A recommendation model simplifies the selection of assessment methods that can overcome a specific incompleteness problem. This paper focuses on (i) the proposed approach to selecting appropriate assessment methods for the complementarity approach; (ii) the characterization of information extraction and assessment methods; (iii) a rule base that allows general processability, profitability in the complementarity approach, and performance of an assessment method to be assessed.