Evolution Of Data Streams Research Articles

Evolving data stream clustering based on constant false clustering probability

Today’s world needs new methods to deal with and analyze the ever-increasingly generated data streams. Two of the most challenging aspects of data streams are (i) concept drift, i.e. evolution of data stream over time, which requires the ability to make timely decisions against the high speed of receiving new data; (ii) limited memory storage and the impracticality of using memory due to the large amount of data. Clustering is one of the common methods to process data streams. In this paper, we propose a novel, fully-online, density-based method for clustering evolving data streams. In recent years, a number of methods have been proposed, which also have the ability to cluster data streams. The main limitation of these methods is the use of parameters based on knowledge-expert. This work is among the first works to address this issue. Constant False Clustering Probability, CFCP, has tried to choose the algorithm’s parameters based on Statistics. The proposed method has also the ability to identify clusters with arbitrary shapes. It is robust to noise and offers high accuracy and efficiency in both low and high dimensions. In this method, we determine the value of the parameters by using statistical theories and do not require more information, taking advantage of expert-knowledge. The presented experimental results show that the method performs data clustering at high speeds without reducing the quality compared to the state-of-the-art algorithms.

An adaptive algorithm for dealing with data stream evolution and singularity

The analysis of representative features for multi-source and massive data streams, from which implicit knowledge is mined, has become a research hotspot in the era of rapid development of data industry. In response to the weak adaptability of traditional algorithms, an adaptive algorithm GNG-L is proposed based on growing neural gas (GNG) for monitoring and tracking the drift and singularity of real-time data stream in non-stationary environments, which includes three mechanisms, namely weight adaptation, neuron deletion and generation. Firstly, the mechanism of weight adaptation is proposed by analyzing the changes of the local characteristics for data streams, which ensures the network topology is adjusted accurately and quickly. Secondly, the adaptive deletion mechanism removes neural nodes that are no longer updated due to the evolution of data streams. Finally, the generation mechanism is trigged when the new feature of data stream evolution needs to be described in new regions of the feature space. The proposed model has been validated based on a number of data sets, and the results show that the algorithm proposed in this paper can effectively track changes of data sets in non-stationary environments.

EACD: evolutionary adaptation to concept drifts in data streams

This paper presents a novel ensemble learning method based on evolutionary algorithms to cope with different types of concept drifts in non-stationary data stream classification tasks. In ensemble learning, multiple learners forming an ensemble are trained to obtain a better predictive performance compared to that of a single learner, especially in non-stationary environments, where data evolve over time. The evolution of data streams can be viewed as a problem of changing environment, and evolutionary algorithms offer a natural solution to this problem. The method proposed in this paper uses random subspaces of features from a pool of features to create different classification types in the ensemble. Each such type consists of a limited number of classifiers (decision trees) that have been built at different times over the data stream. An evolutionary algorithm (replicator dynamics) is used to adapt to different concept drifts; it allows the types with a higher performance to increase and those with a lower performance to decrease in size. Genetic algorithm is then applied to build a two-layer architecture based on the proposed technique to dynamically optimise the combination of features in each type to achieve a better adaptation to new concepts. The proposed method, called EACD, offers both implicit and explicit mechanisms to deal with concept drifts. A set of experiments employing four artificial and five real-world data streams is conducted to compare its performance with that of the state-of-the-art algorithms using the immediate and delayed prequential evaluation methods. The results demonstrate favourable performance of the proposed EACD method in different environments.

RED-GENE: An Evolutionary Game Theoretic Approach to Adaptive Data Stream Classification

The extensive growth of digital technologies such as the Internet of Things (IoT), social media networks and forecasting systems has led to new challenges regarding computational complexity and big data mining. The classification task in such applications is not trivial due to the high volume of related data and limited time available for the task. It is particularly difficult when dealing with data streams, where each instance of data is typically processed once on its arrival (i.e. online) while the underlying data distribution often changes due to the changing environment. In this paper, we propose a novel ensemble-based framework called Replicator Dynamics & Genetic Algorithms Approach (RED-GENE) for effective data stream classification in the context of changing environment leading to concept drifts (i.e. evolution of data streams). RED-GENE employs three novel Replicator Dynamics (RD) strategies along with a Genetic Algorithm (GA) optimisation technique to flexibly adapt to different types of concept drifts when performing data stream classification tasks. The proposed framework works as follows. First, a set of random feature combinations is drawn from a given pool of features of the target data stream to create different classification types. Next, RD is used to allow the classification types achieving higher classification accuracy to grow and those with lower accuracy to shrink. A modified version of the classic GA is then employed to optimise the randomly drawn combinations of features in each classification type. The proposed framework was tested using nine data streams (including both real-world and synthetic datasets) to investigate different variations of the proposed framework and compare its performance to other state-of-the-art algorithms using immediate and delayed prequential evaluation methods. The results demonstrated that the proposed framework can provide the best accuracy on average when comparing to five other state-of-the-art algorithms.

A fast density-based data stream clustering algorithm with cluster centers self-determined for mixed data

Most data streams encountered in real life are data objects with mixed numerical and categorical attributes. Currently most data stream algorithms have shortcomings including low clustering quality, difficulties in determining cluster centers, poor ability for dealing with outliers’ issue. A fast density-based data stream clustering algorithm with cluster centers automatically determined in the initialization stage is proposed. Based on data attribute relationships analysis, mixed data sets are filed into three types whose corresponding distance measure metrics are designed. Based on field intensity-distance distribution graph for each data object, linear regression model and residuals analysis are used to find the outliers of the graph, enabling cluster centers automatic determination. After the cluster centers are found, all data objects can be clustered according to their distance with centers. The data stream clustering algorithm adopts an online/offline two-stage processing framework, and a new micro cluster characteristic vector to maintain the arriving data objects dynamically. Micro clusters decay function and deletion mechanism of micro clusters are used to maintain the micro clusters, which reflects the data stream evolution process accurately. Finally, the performances of the proposed algorithm are testified by a series of experiments on real-world mixed data sets in comparison with several outstanding clustering algorithms in terms of the clustering purity, efficiency and time complexity.