https://cis.temple.edu/~latecki/Papers/mldm07.pdf
Abstract. Outlier detection has recently become an important prob-lem in many industrial and ̄nancial applications. In this paper, a novel unsupervised algorithm for outlier detection with a solid statistical foun-dation is proposed. First we modify a nonparametric density estimate with a variable kernel to yield a robust local density estimation. Out-liers are then detected by comparing the ...