Abstract: Embodiments of the present disclosure relate to a database system for dynamically and automatically accessing and storing data items from multiple data sources. The system may, for example, determine data sources to access, and ways of accessing data items from those data sources, based on an indication of an analysis level and/or other analysis criteria. Further, the system may selectively and efficiently integrate data items from the multiple data sources. Selective integration of data items may be based, for example, on the indication of the analysis level and/or other analysis criteria. The system may further generate outputs of the selective integration of data items. These outputs may, for example, include specialized reports and/or user interfaces. The outputs of the system, in some implementations, may be interactive and dynamically updated in response to user inputs, for example.

Abstract: The present invention is in the field of Healthcare Claims Fraud Detection. Fraud is perpetrated across multiple healthcare payers. There are few labeled or “tagged” historical fraud examples needed to build “supervised”, traditional fraud models using multiple regression, logistic regression or neural networks. Current technology is to build “Unsupervised Fraud Outlier Detection Models”. Current techniques rely on parametric statistics that are based on assumptions such as outlier free and “normally distributed” data. Even some non-parametric statistics are adversely influenced by non-normality and the presence of outliers. Current technology cannot represent the combined variable values into one meaningful value that reflects the overall risk that this observation is an outlier. The single value, the “score”, must be capable of being measured on the same scale across different segments, such as geographies and specialty groups.