Methodology for Preprocessing Semi-Structured Data for Making Managerial Decisions in the Healthcare

Abstract

This paper describes the process of supporting management decisionmaking in healthcare based on data mining. The authors described various problems and specifics of data in medical information systems, leading to the complexity of their analysis and integration, such as: the presence of a large number of specific abbreviations, errors in the data and their poor structure. The paper demonstrates an approach to the search and further disclosure of abbreviations in texts, built on a combination of machine and human processing. A method for extracting features from semi-structured fields using an expert in the subject area and using various visualizations is proposed. The proposed abbreviation search and disclosure methods, based on a hybrid approach combining the strengths of processing with the help of a machine and an expert, can increase the number of abbreviations found automatically and significantly reduce the time spent by experts on processing the remaining reductions. In addition, the method for automated feature extraction during integration can significantly increase the amount of useful input data, while reducing the time of the expert.

Review of Time Domain Electronic Medical Record Taxonomies in the Application of Machine Learning

“…Ref. [78] is a method to allocate resources from the knowledge of semi-structured healthcare data. Ref.…”