A Multi Disciplinary Data Quality Curation Framework for Discrete and Continuous IoT Data

The linked devices create enormous amounts of raw data every day. Every day, the wirelessly linked gadgets produce over 2.5 quintillion bytes of data. Assuring clean, high-quality original information before analysis is a difficult undertaking. A little amount of inaccurate data during the examination of medical data may cause the decision-making process to be misled. It is challenging to associate data items during a quality review when the data is unstructured and incorrectly formatted. The current paradigm for data curation uses the three-phase quality assessment to transform a collection of unclean data into a set of standard data. In an existing method, the dependence assessment is not taken into account for quality analysis. The 4-tier feature matching technique is used in the proposed study for information extraction and evaluation. In our work, the continuous template-matching & dependency checking are carried out concurrently. To verify the strong map with the specified allusion model, perform template matching. Similarly, the unfair relation finding column will be removed by the dependency verification. Trifacta and OpenRefine 2.0 are used for the experiment's quality analysis.