Set Of Data Instances Research Articles

Effectiveness of Tree-based Ensembles for Anomaly Discovery: Insights, Batch and Streaming Active Learning

Anomaly detection (AD) task corresponds to identifying the true anomalies among a given set of data instances. AD algorithms score the data instances and produce a ranked list of candidate anomalies. The ranked list of anomalies is then analyzed by a human to discover the true anomalies. Ensemble of tree-based anomaly detectors trained in an unsupervised manner and scoring based on uniform weights for ensembles are shown to work well in practice. However, the manual process of analysis can be laborious for the human analyst when the number of false-positives is very high. Therefore, in many real-world AD applications including computer security and fraud prevention, the anomaly detector must be configurable by the human analyst to minimize the effort on false positives. One important way to configure the detector is by providing true labels (nominal or anomaly) for a few instances. Recent work on active anomaly discovery has shown that greedily querying the top-scoring instance and tuning the weights of ensembles based on label feedback allows us to quickly discover true anomalies. This paper makes four main contributions to improve the state-of-the-art in anomaly discovery using tree-based ensembles. First, we provide an important insight that explains the practical successes of unsupervised tree-based ensembles and active learning based on greedy query selection strategy. We also show empirical results on real-world data to support our insights and theoretical analysis to support active learning. Second, we develop a novel batch active learning algorithm to improve the diversity of discovered anomalies based on a formalism called compact description to describe the discovered anomalies. Third, we develop a novel active learning algorithm to handle streaming data setting. We present a data drift detection algorithm that not only detects the drift robustly, but also allows us to take corrective actions to adapt the anomaly detector in a principled manner. Fourth, we present extensive experiments to evaluate our insights and our tree-based active anomaly discovery algorithms in both batch and streaming data settings. Our results show that active learning allows us to discover significantly more anomalies than state-of-the-art unsupervised baselines, our batch active learning algorithm discovers diverse anomalies, and our algorithms under the streaming-data setup are competitive with the batch setup.

Parameterisation of demand-driven material requirements planning: a multi-objective genetic algorithm

Demand-Driven Material Requirements Planning (DDMRP) is a recent inventory management method that has generated considerable interest in both academia and industry. Many recent papers have demonstrated the superiority of DDMRP over classical methods like MRP or Kanban, an observation confirmed by companies that have implemented DDMRP. However, DDMRP depends on many parameters that affect its performance. Only general rules are given by the authors of the method to fix these parameters but no algorithm. The present paper aims to fill this gap by proposing a multi-objective optimisation algorithm to fix a set of eight identified parameters. The suggested genetic algorithm is coupled with a simulation algorithm that computes the objective functions. Two opposing objective functions are considered: first, the maximisation of orders delivered on-time to the customer and, second, the minimisation of on-hand inventory. A set of data instances was generated to test the suggested method. Fronts of non-dominated solutions are found for all these instances.

A hybrid Electromagnetism-like Mechanism/tabu search procedure for the single machine scheduling problem with a maximum lateness objective

This paper presents a hybrid meta-heuristic search procedure to solve the well-known single machine scheduling problem to minimize the maximum lateness over all jobs, where precedence relations may exist between some of the jobs. The hybridization consists of a well-designed balance between the principles borrowed from an Electromagnetism-like Mechanism algorithm and the characteristics used in a tabu search procedure. The Electromagnetism-like Mechanism (EM) algorithm follows a search pattern based on the theory of physics to simulate attraction and repulsion of solutions in order to move towards more promising solutions. The well-known tabu search enhances the performance of a local search method by using memory structures by prohibiting visited solutions during a certain time of the search process. The hybridization of both algorithms results in an important trade-off between intensification and diversification strategies. These strategies will be discussed in detail. To that purpose, a new set of data instances is used to compare different elements of the hybrid search procedure and to validate the performance of the algorithm.