The paper presents an original fuzzy solution to the issue of outliers detection in graph databases. In particular, the following novelties are introduced: redefinition of an outlier in terms of fuzzy logic and methods for finding and marking outliers in graph-organized databases. The former refers to explaining outlying objects via fuzzy rules (IF-THEN rules) when linguistic knowledge rather than crisp data on that are accessible. The latter includes processing NOSQL graph-represented datasets to forms suitable for applying fuzzy rules. An application example on a Customer Relationship Management (CRM) with linguistic knowledge and graph structure is proposed.
Niewiadomski, Adam; Kacprowicz, Marcin; and Bartczak, Monika, "Outliers Detection In Graph-Represented Databases Using Fuzzy Rules" (2021). PACIS 2021 Proceedings. 1.
When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.