Gene selection is an important issue in analyzing multiclass microarray data. Among many proposed selection methods, the traditional ANOVA F test statistic has been employed to identify informative genes for both class prediction (classification) and discovery problems. However, the F test statistic assumes an equal variance. This assumption may not be realistic for gene expression data. This paper explores other alternative test statistics which can handle heterogeneity of the variances. We study five such test statistics, which include Brown-Forsythe test statistic and Welch test statistic. Their performance is evaluated and compared with that of F statistic over different classification methods applied to publicly available microarray datasets.

Fc d fe E YD2 hg hDg iW 4p q¤ QE F" S¤ @ sr ut £v AD& wE F$ x £y F D h Y f w A F Q ' Y ¥ w d e f g d hx £y w d Y i j k l w m hn qy Fo p d q k r s ut w m wv x y u m z y Y Y {f '

Given a query point and a collection of spatial features, a multi-type nearest neighbor (MTNN) query finds the shortest tour for the query point such that only one instance of each feature is visited during the tour. For example, a tourist may be interested in finding the shortest tour which starts at a hotel and passes through a post office, a gas station, and a grocery store. The MTNN query problem is different from the traditional nearest neighbor query problem in that there are many objects for each feature type and the shortest tour should pass through only one object from each feature type. In this paper, we propose an R-tree based algorithm that exploits a page-level upper bound for efficient computation in clustered data sets and finds optimal query results. We compare our method with a recently proposed method, RLORD, which was developed to solve the optimal sequenced route (OSR) query. In our view, OSR represents a spatially constrained version of MTNN. Experimental results are provided to show the strength of our algorithm and design decisions related to performance tuning.

No abstract