Contextual Outlier Interpretation

نویسندگان

Ninghao Liu

Donghwa Shin

Xia Hu

چکیده

Outlier detection plays an essential role in many data-driven applications to identify isolated instances that are dierent from the majority. While many statistical learning and data mining techniques have been used for developing more eective outlier detection algorithms, the interpretation of detected outliers does not receive much aention. Interpretation is becoming increasingly important to help people trust and evaluate the developed models through providing intrinsic reasons why the certain outliers are chosen. It is dicult, if not impossible, to simply apply feature selection for explaining outliers due to the distinct characteristics of various detection models, complicated structures of data in certain applications, and imbalanced distribution of outliers and normal instances. In addition, the role of contrastive contexts where outliers locate, as well as the relation between outliers and contexts, are usually overlooked in interpretation. To tackle the issues above, in this paper, we propose a novel Contextual Outlier INterpretation (COIN) method to explain the abnormality of existing outliers spoed by detectors. e interpretability for an outlier is achieved from three aspects: outlierness score, aributes that contribute to the abnormality, and contextual description of its neighborhoods. Experimental results on various types of datasets demonstrate the exibility and eectiveness of the proposed framework compared with existing interpretation approaches.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Identification and assessment of potentially high-mortality intensive care units using the ANZICS Centre for Outcome and Resource Evaluation clinical registry.

PURPOSE A hospital's highest-risk patients are managed in the intensive care unit. Outcomes are determined by patients' severity of illness, existing comorbidities and by processes of care delivered. The Australian and New Zealand Intensive Care Society (ANZICS) Centre for Outcome and Resource Evaluation (CORE) manages a binational clinical registry to benchmark performance, and report and asse...

متن کامل

Context Aware Anomalous Behaviour Detection in Crowded Surveillance

This work addresses the detection of human behavioural anomalies in surveillance. We address in particular the problem of detecting subtle behaviour in a crowded behaviourally heterogeneous surveillance scene. We novel methods of extracting scene context and social context to improve the detection of behavioural anomalies, and in particular permit the detection of subtle behavioural anomalies. ...

متن کامل

A Supervised Approach for Multiple Sclerosis Lesion Segmentation Using Context Features and an Outlier Map

Automatic multiple sclerosis (MS) lesion segmentation in magnetic resonance imaging (MRI) is a challenging task due to the small size of the lesions, its heterogeneous shape and distribution, overlapping tissue intensity distributions, and the inherent artifacts of MRI. In this paper we propose a pipeline for MS lesion segmentation that combines prior knowledge and contextual information into a...

متن کامل

Local Outlier Detection with Interpretation

Outlier detection aims at searching for a small set of objects that are inconsistent or considerably deviating from other objects in a dataset. Existing research focuses on outlier identification while omitting the equally important problem of outlier interpretation. This paper presents a novel method named LODI to address both problems at the same time. In LODI, we develop an approach that exp...

متن کامل