Data Source Selection for Integration in Data Sciences via Complex Hesitant Fuzzy Rough Multi-Attribute Decision-Making Method

Data sources are the raw information that is required for analysis and modeling in data sciences. They assist data scientists in making proper conclusions, proving hypotheses, and making rational decisions. It is always preferable if analytical results can be obtained from multiple reliable sources. Thus, it is essential to assess such data sources in data sciences about various critical characteristics and factors. Nevertheless, the application of all-inclusive multi-attribute decision-making methodologies for the selection of data sources for integration has not received adequate attention in the existing literature and research. Thus, this article explains a new multi-attribute decision-making method through the model of the complex hesitant fuzzy rough set, which is the complex hesitant fuzzy rough multi-attribute decision-making method. This methodology would handle the evaluated values of attributes that have uncertainty, hesitancy, and roughness altogether. Besides, this study introduces several properties of complex hesitant fuzzy rough sets and develops several aggregation operators in the framework of complex hesitant fuzzy rough set and their properties. Subsequently, a case study of data source selection in data science is explained to explain the relevance of the developed multi-attribute decision-making framework in data sciences. Finally, the comparison of the devised theory with prevailing theories is interpreted.