Checklist Screening Similarity Matching Algorithm Based on Decision Fusion

In this paper, we propose the algorithm of checklist screening based on decision fusion. First, it Converts unstructured checklist data to structured, regularizes structured checklist data such as extracting index words and removing modifiers, and creates Solr-index based on the customer entity name. Then we calculate the adjustment Levenshtein-Distance(LD) similarity matching score and Longest-Common-Subsequence(LCS) similarity matching score between the customer to be screened and the checklist data respectively. Finally, decision fusion results can be gained by fusing the two scores. The experimental results show that hit-rate and accuracy of the decision fusion algorithm which uses LD algorithm and LCS algorithm to calculates similarity is higher than single algorithm, and the Solr-index checklist data is effectively reduced, also the checklist screening performance is improved.