Showing 2 results for Anomaly Detection
Mehrdad Kargari, Susan Sahranavard,
Volume 31, Issue 1 (3-2020)
Abstract
Background: The continuous growth of healthcare and medicine costs as a strategic commodity requires tools to identify high cost populations and cost control. After the implementation of the healthcare Reform plan in Iran, a huge share of hospital funding has been spent on undesirable costs due to changes in the use of medicines and instruments.
Objective: The aim of this study was to compare the cost of medicines in both the pre and post period of health plan implementation to detect abnormalities and low frequency patterns in the medical prescriptive that account more than 30% of hospital budget funds.
Method: Therefore a data mining model has been used. First, by forming incidence matrices on the cross-features; categorized prescriptions information. Then using normalized risk function to identify abnormal and high cost cases based on the distance between the input data and the mean of the data. The data used are 15078 records, including information from patients' prescriptions from Shari'ati HIS in Tehran-Iran from 2012 to 2016.
Results: According to the obtained results, the proposed model has a positive Likehood ratio (LR+) of 6.35.
Fatemeh Elhambakhsh, Mohammad Saidi- Mehrabad,
Volume 32, Issue 1 (1-2021)
Abstract
Statistical monitoring of dynamic networks is a major topic of interest in complex social systems. Many researches have been conducted on modeling and monitoring dynamic social networks. This article proposes a new methodology for modeling and monitoring dynamic social networks for quick detection of temporal anomalies in network structures using latent variables. The key idea behind our proposed methodology is to determine the importance of latent variables in creating edges between nodes as well as observed covariates. First, latent space model (LSM) is used to model dynamic networks. Vector of parameters in LSM model are monitored through multivariate control charts in order to detect changes in different network sizes. Experiments on simulated social network monitoring demonstrate that our surveillance monitoring strategy can effectively detect abrupt changes between actors in dynamic networks using latent variables.
