Adaptive Clustering for Dynamic IoT Data Streams

Puschmann, Daniel; Barnaghi, Payam; Tafazolli, Rahim

doi:10.1109/jiot.2016.2618909

Cited by 111 publications

(54 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Given the scale of the IoT, a large amount of data is expected in the network and therefore requires a load balancing method, a load balancing algorithm based on restricted Boltzmann machine is proposed in [302]. Online clustering scheme form dynamic IoT data streams is described in [303]. Another work describing an ML application in IoT recommends a combination of PCA and regression for IoT to get better prediction [304].…”

Section: E Emerging Networking Applications Of Unsupervised Learningmentioning

confidence: 99%

Unsupervised Machine Learning for Networking: Techniques, Applications and Research Challenges

et al. 2019

View full text Add to dashboard Cite

While machine learning and artificial intelligence have long been applied in networking research, the bulk of such works has focused on supervised learning. Recently there has been a rising trend of employing unsupervised machine learning using unstructured raw network data to improve network performance and provide services such as traffic engineering, anomaly detection, Internet traffic classification, and quality of service optimization. The interest in applying unsupervised learning techniques in networking emerges from their great success in other fields such as computer vision, natural language processing, speech recognition, and optimal control (e.g., for developing autonomous self-driving cars). Unsupervised learning is interesting since it can unconstrain us from the need of labeled data and manual handcrafted feature engineering thereby facilitating flexible, general, and automated methods of machine learning. The focus of this survey paper is to provide an overview of the applications of unsupervised learning in the domain of networking. We provide a comprehensive survey highlighting the recent advancements in unsupervised learning techniques and describe their applications for various learning tasks in the context of networking. We also provide a discussion on future directions and open research issues, while also identifying potential pitfalls. While a few survey papers focusing on the applications of machine learning in networking have previously been published, a survey of similar scope and breadth is missing in literature. Through this paper, we advance the state of knowledge by carefully synthesizing the insights from these survey papers while also providing contemporary coverage of recent advances.

show abstract

Section: E Emerging Networking Applications Of Unsupervised Learningmentioning

confidence: 99%

Unsupervised Machine Learning for Networking: Techniques, Applications and Research Challenges

et al. 2019

View full text Add to dashboard Cite

show abstract

“…When developing the concept of a secured domain, the following assumptions were taken into account: sensor nodes are mobile, the communication medium used has a low bandwidth and low range, and sensor nodes have limited resources (a small memory resources, low computing power, and limited power capabilities). For this reason, protection of sensor nodes network should be designed in small sensor clusters . From the security point of view, each cluster of sensor nodes should be autonomous.…”

Section: The Concept Of Secured Domainmentioning

confidence: 99%

Procedures for sensor nodes operation in the secured domain

Furtak

Zieliński

Chudzikiewicz

2019

Concurrency and Computation

View full text Add to dashboard Cite

Summary Currently, the networks of mobile sensor nodes, which are IoT components, are increasingly a source of data for building situational awareness. Such networks can be a source of data for critical infrastructure systems if, in addition to the timeliness (freshness) of the data, it will also be possible to ensure the reliability of these data. Such requirements are met by a secure domain of sensor nodes, in which all sensor nodes are authenticated, sensitive data is stored in protected resources of sensor nodes, and data exchange in the domain and outside the domain is cryptographically protected. This paper presents the concept of a safe domain of sensor nodes for IoT and describes the data structures necessary to ensure security in the domain and the way of protecting the resources of each node. Particular attention was paid to the procedure of preparing a sensor node for work in a secured domain of sensor nodes.

show abstract

“…In the IoT domain and especially in smart city data analysis, we are interested in the second type of drift which will be referred as data drift in this paper [40]. Some examples where a data drift may occur in smart cities are related to the replacement of sensors (different calibration), sensor wear and tear [41] or drastic changes to the topics of discussion in social media used for crowdsensing [42].…”

Section: Motivation and Contributionsmentioning

confidence: 99%

“…There are several existing methods and solution addressing the concept drift for supervised learning [41], and some recent ones also for unsupervised learning [40]. However, we focus on the initial step of the analysis (i.e preprocessing).…”

Section: Motivation and Contributionsmentioning

confidence: 99%

“…In this case, we have evaluated the algorithms with the data generated by using the code from the Repository 3 described in [54], which was first used in [40]. The drift is introduced both by shifting the centroids in randomized intervals and by changing the data distribution function used to randomly draw the data from the centroids that are selected through Latin Hypercube Sampling (LHS).…”

Section: Random Lhs Generator Lift (Synthetic and With Drifts)mentioning

confidence: 99%

See 1 more Smart Citation

BEATS: Blocks of Eigenvalues Algorithm for Time series Segmentation

González-Vidal

Barnaghi

Skarmeta

2018

IEEE Trans. Knowl. Data Eng.

Self Cite

View full text Add to dashboard Cite

Abstract-The massive collection of data via emerging technologies like the Internet of Things (IoT) requires finding optimal ways to reduce the observations in the time series analysis domain. The IoT time series require aggregation methods that can preserve and represent the key characteristics of the data. In this paper, we propose a segmentation algorithm that adapts to unannounced mutations of the data (i.e. data drifts). The algorithm splits the data streams into blocks and groups them in square matrices, computes the Discrete Cosine Transform (DCT) and quantizes them. The key information is contained in the upper-left part of the resulting matrix. We extract this sub-matrix, compute the modulus of its eigenvalues and remove duplicates. The algorithm, called BEATS, is designed to tackle dynamic IoT streams, whose distribution changes over time. We implement experiments with six datasets combining real, synthetic, real-world data, and data with drifts. Compared to other segmentation methods like Symbolic Aggregate approXimation (SAX), BEATS shows significant improvements. Trying it with classification and clustering algorithms it provides efficient results. BEATS is an effective mechanism to work with dynamic and multi-variate data, making it suitable for IoT data sources. The datasets, code of the algorithm and the analysis results can be accessed publicly at: https://github.com/auroragonzalez/BEATS.

show abstract

Adaptive Clustering for Dynamic IoT Data Streams

Cited by 111 publications

References 31 publications

Unsupervised Machine Learning for Networking: Techniques, Applications and Research Challenges

Unsupervised Machine Learning for Networking: Techniques, Applications and Research Challenges

Procedures for sensor nodes operation in the secured domain

BEATS: Blocks of Eigenvalues Algorithm for Time series Segmentation

Contact Info

Product

Resources

About