1-3hit |
Recently, multivariate time-series data has been generated in various environments, such as sensor networks and IoT, making anomaly detection in time-series data an essential research topic. Unsupervised learning anomaly detectors identify anomalies by training a model on normal data and producing high residuals for abnormal observations. However, a fundamental issue arises as anomalies do not consistently result in high residuals, necessitating a focus on the time-series patterns of residuals rather than individual residual sizes. In this paper, we present a novel framework comprising two serialized anomaly detectors: the first model calculates residuals as usual, while the second one evaluates the time-series pattern of the computed residuals to determine whether they are normal or abnormal. Experiments conducted on real-world time-series data demonstrate the effectiveness of our proposed framework.
Sang-Chul LEE Christos FALOUTSOS Dong-Kyu CHAE Sang-Wook KIM
This paper deals with a novel, interesting problem of detecting frauds in comparison-shopping services (CSS). In CSS, there exist frauds who perform excessive clicks on a target item. They aim at making the item look very popular and subsequently ranked high in the search and recommendation results. As a result, frauds may distort the quality of recommendations and searches. We propose an approach of detecting such frauds by analyzing click behaviors of users in CSS. We evaluate the effectiveness of the proposed approach on a real-world clickstream dataset.
Sang-Chul LEE Sang-Wook KIM Sunju PARK Dong-Kyu CHAE
This paper addresses recommendation diversification. Existing diversification methods have difficulty in dealing with the tradeoff between accuracy and diversity. We point out the root of the problem in diversification methods and propose a novel method that can avoid the problem. Our method aims to find an optimal solution of the objective function that is carefully designed to consider user preference and the diversity among recommended items simultaneously. In addition, we propose an item clustering and a greedy approximation to achieve efficiency in recommendation.