How to optimize an academic team when the outlier member is leaving?
- Authors: Yu, Shuo , Liu, Jiaying , Wei, Haoran , Xia, Feng , Tong, Hanghang
- Date: 2021
- Type: Text , Journal article
- Relation: IEEE Intelligent Systems Vol. 36, no. 3 (May-Jun 2021), p. 23-30
- Full Text:
- Reviewed:
- Description: An academic team is a highly cohesive collaboration group of scholars, which has been recognized as an effective way to improve scientific output in terms of both quality and quantity. However, the high staff turnover brings about a series of problems that may have negative influences on team performance. To address this challenge, we first detect the tendency of the member who may potentially leave. Here, the outlierness is defined with respect to familiarity, which is quantified by using collaboration intensity. It is assumed that if a team member has a higher familiarity with scholars outside the team, then this member might probably leave the team. To minimize the influence caused by the leaving of such an outlier member, we propose an optimization solution to find a proper candidate who can replace the outlier member. Based on random walk with graph kernel, our solution involves familiarity matching, skill matching, as well as structure matching. The proposed approach proves to be effective and outperforms existing methods when applied to computer science academic teams.
Isolation-based anomaly detection using nearest-neighbor ensembles
- Authors: Bandaragoda, Tharindu , Ting, Kaiming , Albrecht, David , Liu, Fei , Zhu, Ye , Wells, Jonathan
- Date: 2018
- Type: Text , Journal article
- Relation: Computational Intelligence Vol. 34, no. 4 (2018), p. 968-998
- Full Text: false
- Reviewed:
- Description: The first successful isolation-based anomaly detector, ie, iForest, uses trees as a means to perform isolation. Although it has been shown to have advantages over existing anomaly detectors, we have identified 4 weaknesses, ie, its inability to detect local anomalies, anomalies with a high percentage of irrelevant attributes, anomalies that are masked by axis-parallel clusters, and anomalies in multimodal data sets. To overcome these weaknesses, this paper shows that an alternative isolation mechanism is required and thus presents iNNE or isolation using Nearest Neighbor Ensemble. Although relying on nearest neighbors, iNNE runs significantly faster than the existing nearest neighbor–based methods such as the local outlier factor, especially in data sets having thousands of dimensions or millions of instances. This is because the proposed method has linear time complexity and constant space complexity. © 2018 Wiley Periodicals, Inc.
Defying the gravity of learning curve : A characteristic of nearest neighbour anomaly detectors
- Authors: Ting, Kaiming , Washio, Takashi , Wells, Jonathan , Aryal, Sunil
- Date: 2017
- Type: Text , Journal article
- Relation: Machine Learning Vol. 106, no. 1 (2017), p. 55-91
- Full Text: false
- Reviewed:
- Description: Conventional wisdom in machine learning says that all algorithms are expected to follow the trajectory of a learning curve which is often colloquially referred to as ‘more data the better’. We call this ‘the gravity of learning curve’, and it is assumed that no learning algorithms are ‘gravity-defiant’. Contrary to the conventional wisdom, this paper provides the theoretical analysis and the empirical evidence that nearest neighbour anomaly detectors are gravity-defiant algorithms.
Commentary : A decomposition of the outlier detection problem into a set of supervised learning problems
- Authors: Zhu, Ye , Ting, Kaiming
- Date: 2016
- Type: Text , Journal article
- Relation: Machine Learning Vol. 105, no. 2 (2016), p. 301-304
- Full Text: false
- Reviewed:
- Description: This article discusses the material in relation to iForest (Liu et al. in ACM Trans Knowl Discov Data 6(1):3, 2012) reported in a recent Machine Learning Journal paper by Paulheim and Meusel (Mach Learn 100(2–3):509–531, 2015). It presents an empirical comparison result of iForest using the default parameter settings suggested by its creator (Liu et al. 2012) and iForest using the settings employed by Paulheim and Meusel (2015). This comparison has an impact on the conclusion made by Paulheim and Meusel (2015). © 2016, The Author(s).
Half-space mass : a maximally robust and efficient data depth method
- Authors: Chen, Bo , Ting, Kaiming , Washio, Takashi , Haffari, Gholamreza
- Date: 2015
- Type: Text , Journal article
- Relation: Machine Learning Vol. 100, no. 2-3 (2015), p. 677-699
- Full Text: false
- Reviewed:
- Description: Data depth is a statistical method which models data distribution in terms of center-outward ranking rather than density or linear ranking. While there are a lot of academic interests, its applications are hampered by the lack of a method which is both robust and efficient. This paper introduces Half-Space Mass which is a significantly improved version of half-space data depth. Half-Space Mass is the only data depth method which is both robust and efficient, as far as we know. We also reveal four theoretical properties of Half-Space Mass: (i) its resultant mass distribution is concave regardless of the underlying density distribution, (ii) its maximum point is unique which can be considered as median, (iii) the median is maximally robust, and (iv) its estimation extends to a higher dimensional space in which the convex hull of the dataset occupies zero volume. We demonstrate the power of Half-Space Mass through its applications in two tasks. In anomaly detection, being a maximally robust location estimator leads directly to a robust anomaly detector that yields a better detection accuracy than half-space depth; and it runs orders of magnitude faster than L2 depth, an existing maximally robust location estimator. In clustering, the Half-Space Mass version of K-means overcomes three weaknesses of K-means.
- Description: Data depth is a statistical method which models data distribution in terms of center-outward ranking rather than density or linear ranking. While there are a lot of academic interests, its applications are hampered by the lack of a method which is both robust and efficient. This paper introduces
LiNearN : A new approach to nearest neighbour density estimator
- Authors: Wells, Jonathan , Ting, Kaiming , Washio, Takashi
- Date: 2014
- Type: Text , Journal article
- Relation: Pattern Recognition Vol. 47, no. 8 (2014), p. 2702-2720
- Full Text: false
- Reviewed:
- Description: Despite their wide spread use, nearest neighbour density estimators have two fundamental limitations: O(n2) time complexity and O(n) space complexity. Both limitations constrain nearest neighbour density estimators to small data sets only. Recent progress using indexing schemes has improved to near linear time complexity only.We propose a new approach, called LiNearN for Linear time Nearest Neighbour algorithm, that yields the first nearest neighbour density estimator having O(n) time complexity and constant space complexity, as far as we know. This is achieved without using any indexing scheme because LiNearN uses a subsampling approach for which the subsample values are significantly less than the data size. Like existing density estimators, our asymptotic analysis reveals that the new density estimator has a parameter to trade off between bias and variance. We show that algorithms based on the new nearest neighbour density estimator can easily scale up to data sets with millions of instances in anomaly detection and clustering tasks. Highlights•Reject the premise that a NN algorithm must find the NN for every instance.•The first NN density estimator that has O(n) time complexity and O(1) space complexity.•These complexities are achieved without using any indexing scheme.•Our asymptotic analysis reveals that it trades off between bias and variance.•Easily scales up to large data sets in anomaly detection and clustering tasks.
Mass estimation
- Authors: Ting, Kaiming , Zhou, Guang , Liu, Fei , Tan, Swee
- Date: 2013
- Type: Text , Journal article
- Relation: Machine Learning Vol. 90, no. 1 (2013), p. 127-160
- Full Text: false
- Reviewed:
- Description: This paper introduces mass estimation—a base modelling mechanism that can be employed to solve various tasks in machine learning. We present the theoretical basis of mass and efficient methods to estimate mass. We show that mass estimation solves problems effectively in tasks such as information retrieval, regression and anomaly detection. The models, which use mass in these three tasks, perform at least as well as and often better than eight state-of-the-art methods in terms of task-specific performance measures. In addition, mass estimation has constant time and space complexities.
Unitary anomaly detection for ubiquitous safety in machine health monitoring
- Authors: Amar, Muhammad , Gondal, Iqbal , Wilson, Campbell
- Date: 2012
- Type: Text , Conference paper
- Relation: 19th International Conference on Neural Information Processing (INCONIP) p. 361-368
- Full Text: false
- Reviewed:
- Description: Safety has always been of vital concern in both industrial and home applications. Ensuring safety often requires certain quantifications regarding the inclusive behavior of the system under observation in order to determine deviations from normal behavior. In machine health monitoring, the vibration signal is of great importance for such measurements because it includes abundant information from several machine parts and surroundings that can influence machine behavior. This paper proposes a unitary anomaly detection technique (UAD) that, upon observation of abnormal behavior in the vibration signal, can trigger an alarm with an adjustable threshold in order to meet different safety requirements. The normalized amplitude of spectral contents of the quasi stationary time vibration signal are divided into frequency bins, and the summed amplitudes frequencies over bin are used as features. From a training set consisting of normal vibration signals, Gaussian distribution models are obtained for each feature, which are then used for anomaly detection.