A Single Pass Trellis-Based Algorithm for Clustering Evolving Data Streams

The rapid spreading of positioning devices leads to the generation of massive spatiotemporal trajectories data. In some scenarios, spatiotemporal data are received in stream manner. Clustering of stream data is beneficial for different applications such as traffic management and weather forecasting. In this article, an algorithm for Continuous Clustering of Trajectory Stream Data Based on Micro Cluster Life is proposed. The algorithm consists of two phases. There is the online phase where temporal micro clusters are used to store summarized spatiotemporal information for each group of similar segments. The clustering task in online phase is based on temporal micro cluster lifetime instead of time window technique which divides stream data into time bins and clusters each bin separately. For offline phase, a density based clustering approach is used to generate macro clusters depending on temporal micro clusters. The evaluation of the proposed algorithm on real data sets shows the efficiency and the effectiveness of the proposed algorithm and proved it is efficient alternative to time window technique.

Recently data stream has been extensively explored due to its emergence in large deal of applications such as sensor networks,web click streams and network flows. Vast majority of researches in the context of data stream mining are devoted to superviselearning, whereas, in real word human practice label of data are rarely available to the learning algorithms. Hence, clustering asthe most important unsupervised learning has been in the gravity of focus of quite a lot number of the researchers in data streamcommunity. Clustering paradigms basically place the similar objects together and separate the dissimilar ones into differentclusters.In this paper, we propose a Statistical framework for data Stream Clustering, which abbreviated as StatisStreamClust that makesuse of two components to find clusters in data stream. The first component especially designed to detect concept change wheredata underlying distributions change from time to time. Upon detection of concept change by the first component, the secondcomponent is triggered to update the whole clustering model. StatisStreamClust brings great benefits to data stream clusteringincluding no sensitivity to the number of clusters and dimensions, reasonable complexity and in the meantime desirable performance,and finally no need to determine window size a priori. To explore the advantages of our approach, quite a lot ofexperiments with different settings and specifications are conducted. The obtained results are very promising.

Data stream clustering has emerged as a challenging and interesting problem over the past few years. Due to the evolving nature, and one-pass restriction imposed by the data stream model, traditional clustering algorithms are inapplicable for stream clustering. This problem becomes even more challenging when the data is high-dimensional and the clusters are not linearly separable in the input space. In this paper, we propose a nonlinear stream clustering algorithm that adapts to the stream's evolutionary changes. Using the kernel methods for dealing with the non-linearity of data separation, we propose a novel 2-tier stream clustering architecture. Tier-1 captures the temporal locality in the stream, by partitioning it into segments, using a kernel-based novelty detection approach. Tier-2 exploits this segment structure to continuously project the streaming data nonlinearly onto a low-dimensional space (LDS), before assigning them to a cluster. We demonstrate the effectiveness of our approach through extensive experimental evaluation on various real-world datasets.

© 2019 Association for Computing Machinery.Number of connected devices is steadily increasing and these devices continuously generate data streams. These data streams are often high dimensional and contain concept drift. Real-time processing of data streams is arousing interest despite many challenges. Clustering is a method that does not need labeled instances (it is unsupervised) and it can be applied with less prior information about the data. These properties make clustering one of the most suitable methods for real-time data stream processing. Moreover, data embedding is a process that may simplify clustering and makes visualization of high dimensional data possible. There exist several data stream clustering algorithms in the literature, however no data stream embedding method exists. UMAP is a data embedding algorithm that is suitable to be applied on data streams, but it cannot adopt concept drift. In this study, we have developed a new method to apply UMAP on data streams, adopt concept drift and cluster embedded data instances using any distance based clustering algorithms.

This article studies a new problem of data stream clustering, namely, multiview data stream (MVStream) clustering. Although many data stream clustering algorithms have been developed, they are restricted to the single-view streaming data, and clustering MVStreams still remains largely unsolved. In addition to the many issues encountered by the conventional single-view data stream clustering, such as capturing cluster evolution and discovering clusters of arbitrary shapes under the limited computational resources, the main challenge of MVStream clustering lies in integrating information from multiple views in a streaming manner and abstracting summary statistics from the integrated features simultaneously. In this article, we propose a novel MVStream clustering algorithm for the first time. The main idea is to design a multiview support vector domain description (MVSVDD) model, by which the information from multiple insufficient views can be integrated, and the outputting support vectors (SVs) are utilized to abstract the summary statistics of the historical multiview data objects. Based on the MVSVDD model, a new multiview cluster labeling method is designed, whereby clusters of arbitrary shapes can be discovered for each view. By tracking the cluster labels of SVs in each view, the cluster evolution associated with concept drift can be captured. Since the SVs occupy only a small portion of data objects, the proposed MVStream algorithm is quite efficient with the limited computational resources. Extensive experiments are conducted to demonstrate the effectiveness and efficiency of the proposed method.

Summary form only given. Data streams are continuous flows of data. Examples of data streams include network traffic, sensor data, call center records and so on. Their sheer volume and speed pose a great challenge for the data mining community to mine them. Data streams demonstrate several unique properties: infinite length, concept-drift, concept-evolution, and feature-evolution. Concept-drift occurs in data streams when the underlying concept of data changes over time. Concept-evolution occurs when new classes evolve in streams. Feature-evolution occurs when feature set varies with time in data streams. Each of these properties adds a challenge to data stream mining. This invited talk will present an organized picture on how to handle various data mining techniques in data streams: in particular, how to handle classification in evolving data streams by addressing these challenges.

Long short-term memory self-adapting online random forests for evolving data stream regression

Mining data streams poses great challenges due to the limited memory availability and real-time query response requirement. Clustering an evolving data stream is especially interesting because it captures not only the changing distribution of clusters but also the evolving behaviors of individual clusters. In this paper, we present a novel method for tracking the evolution of clusters over sliding windows. In our SWClustering algorithm, we combine the exponential histogram with the temporal cluster features, propose a novel data structure, the Exponential Histogram of Cluster Features (EHCF). The exponential histogram is used to handle the in-cluster evolution, and the temporal cluster features represent the change of the cluster distribution. Our approach has several advantages over existing methods: (1) the quality of the clusters is improved because the EHCF captures the distribution of recent records precisely; (2) compared with previous methods, the mechanism employed to adaptively maintain the in-cluster synopsis can track the cluster evolution better, while consuming much less memory; (3) the EHCF provides a flexible framework for analyzing the cluster evolution and tracking a specific cluster efficiently without interfering with other clusters, thus reducing the consumption of computing resources for data stream clustering. Both the theoretical analysis and extensive experiments show the effectiveness and efficiency of the proposed method.

Nowadays many applications need to deal with evolving data streams. In this work, we propose an incremental clustering approach for the exploitation of user constraints on data streams. Conventional constraints do not make sense on streaming data, so we extend the classic notion of constraint set into a constraint stream. We propose methods for using the constraint stream as data items are forgotten or new items arrive. Also we present an on-line clustering approach for the cost-based enforcement of the constraints during cluster adaptation on evolving data streams. Our method introduces the concept of multi-clusters (m-clusters) to capture arbitrarily shaped clusters. An m-cluster consists of multiple dense overlapping regions, named s-clusters, each of which can be efficiently represented by a single point. Also it proposes the definition of outliers clusters in order to handle outliers while it provides methods to observe changes in structure of clusters as data evolves.Keywordsstream clusteringsemi-supervised learningconstraint-based clustering

Most machine learning algorithms rely on a set of hyperparameters, which are tuned either by domain experts or using automated methods. In the context of data streams, the complexity of this task is compounded by the need for tuning every time a concept drift occurs. The problem of automated tuning may lead to classification performance improvements. Current methods optimise the hyperparameters from scratch every time concept drift is detected. Additionally, meta-learning techniques developed for static machine learning tasks allow pre-built knowledge to adapt faster to new tasks. We propose a meta-learning approach for automated tuning in evolving data streams, specifically for tuning parameters in Adaptive Random Forest (ARF) and parameters for several drift detectors. We compare our approach against non-adaptive methods on both synthetic and real-world datasets. Empirical results show that our approach makes a compromises between predictive performance and resource consumption for ARF, and between true positives and false positives for drift detectors.

Random forest is an ensemble method used to improve the performance of single tree classifiers. In evolving data streams, the classifier needs to be adaptive and work under constraints of space and time. One benefit of random forest is its ability to be executed in parallel. In our research we introduce a random forest model utilizing a hybrid of both GPU and CPU, called GPU-based State-Adaptive Random Forest (GSARF). We address the pre-existing challenges of adapting random forest for data streams, specifically in the area of continual learning. Our novel approach reuses previously seen trees in the random forest when previous concepts reappear. This allows us to retain prior knowledge and provide a more stable predictive accuracy when changes occur in the data stream. Our random forest for data streams stores three types of trees, foreground trees which are trees that are currently used in prediction, background trees which are trees that are built when we are aware of possible changes in the data streams, and candidate trees which are trees that had been highly used in the previous concepts, but are now discarded due to changes in the data stream. We store candidate trees as they may be potentially useful at a later period in a repository and can be accessed when needed. We empirically show our technique performs up to 138 times the speed compared to current CPU-based random forest benchmarks. Our approach has shown to outperform a baseline GPU-based approach in terms of cumulative accuracy performance.

Mining top- k frequent closed itemsets over data streams using the sliding window model

Contraction Clustering (RASTER) is a single-pass algorithm for density-based clustering of 2D data. It can process arbitrary amounts of data in linear time and in constant memory, quickly identifying approximate clusters. It also exhibits good scalability in the presence of multiple CPU cores. RASTER exhibits very competitive performance compared to standard clustering algorithms, but at the cost of decreased precision. Yet, RASTER is limited to batch processing and unable to identify clusters that only exist temporarily. In contrast, S-RASTER is an adaptation of RASTER to the stream processing paradigm that is able to identify clusters in evolving data streams. This algorithm retains the main benefits of its parent algorithm, i.e. single-pass linear time cost and constant memory requirements for each discrete time step within a sliding window. The sliding window is efficiently pruned, and clustering is still performed in linear time. Like RASTER, S-RASTER trades off an often negligible amount of precision for speed. Our evaluation shows that competing algorithms are at least 50% slower. Furthermore, S-RASTER shows good qualitative results, based on standard metrics. It is very well suited to real-world scenarios where clustering does not happen continually but only periodically.

The data stream domain has become increasingly important in recent years because of its applicability to a wide variety of applications. Problems such as data mining and privacy preservation which have been studied for traditional data sets cannot be easily solved for the data stream domain. This is because the large volume of data arriving in a stream renders most algorithms to inefficient as most mining and privacy preservation algorithms require multiple scans of data which is unrealistic for stream data. More importantly, the characteristics of the data stream can change over time and the evolving pattern needs to be captured. In this talk, I’ll discuss the issued and focus on how to mine evolving data streams and preserve privacy.

Processing data streams gained much importance in recent years. Standard machine learning algorithms do not cope well with non-stationary streaming data, where decision models evolve and generate so-called concept drift. Online adaptive algorithms emerged to solve these issues. They learn incrementally and generally require explicit forgetting mechanisms to adapt to concept drift. In this paper, we propose the application of Kalman filtering to handle evolving data streams. This novel approach addresses data stream mining and concept drift management challenges from a new perspective, directly modelling a representation suitable for the data streams. First, we study a Kalman filter based learning a pproach and investigate its integration into the Naïve Bayes algorithm, namely KalmanNB. Additionally, we propose the Hoeffding Kalman Tree, a combination of the Hoeffding Tree with KalmanNB. Empirical results demonstrate that the Kalman filter based approach inherently manages concept drifts, and it adapts to the emerging concept more rapidly than the state-of-the-art algorithms. Moreover, it is an accurate and robust approach and requires less storage while still being faster.