|Session Chair:||Guy Djokam|
Speaker 1: Katelynn Huneycutt
- Computing Rankability using Evolutionary Optimization
- While there are many algorithms available to rank data, it can be difficult to determine the validity of the resulting ranking. In their 2019 paper, Anderson, Chartier, and Langville define rankability to be a dataset's inherent ability to produce a meaningful ranking of its items. In this talk, we present an evolutionary optimization algorithm, a heuristic method, to calculate the rankability for large data sets using the rankability measure presented in their paper.
Speaker 2: Michael Lucagbo
- Rectangular Reference Regions for Multivariate Normal Measurements in Laboratory Medicine
- Reference regions are invaluable in the interpretation of results of biochemical and physiological tests of patients. Moreover, when there are multiple biochemical analytes measured from each subject, a multivariate reference region is called for. Such reference regions are more desirable than multiple univariate reference regions because of their greater specificity against false positives. Traditionally, multivariate reference regions have been constructed as ellipsoidal regions. This approach suffers from a major drawback: it cannot detect if a specific measurement is an outlier. For this reason, we develop a procedure to construct a rectangular reference region in a multivariate normal setup. We focus on prediction regions, a specific type of reference region. The approach uses parametric bootstrap estimation to estimate that prediction factor and hence construct the prediction region. The results show good coverage that is fairly robust to small sample sizes and high dimensions.