Data Mining Techniques in Sensor Networks

Appice, Annalisa; Ciampi, Anna; Fumarola, Fabio; Malerba, Donato

doi:10.1007/978-1-4471-5454-9

Cited by 6 publications

(4 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Anomaly Detection is the process of identifying unusual patterns in data sets which do not comply with well-established normal behaviour [28]. These atypical patterns in data sets are called anomalies or outliers.…”

Section: Data Quality and Anomaly Detectionmentioning

confidence: 99%

“…Statistical and machine learning-based predictive algorithms can be used to develop the predictive system. The time-series of the filtered sensor-nodes are used to train the predictive model [28]. The data which will be used to train the predictive model is theoretically considered to be an openended time-series because there is a relatively long sequence of observations.…”

Section: ) the Predictive Analytics-based Anomaly Detection Unitmentioning

confidence: 99%

“…Using the whole time series may overfit the trained model and practically not possible. To tackle this issue, the sliding window modelling [28] will be used to select a part of the time-series only, the size of the sliding window will vary based on the application itself. Time series decomposition technique will be applied, combined with statistical and machine learning technique such as the Seasonal Autoregressive Integrated Moving Average, (SARIMA) as shown in Fig.…”

Section: ) the Predictive Analytics-based Anomaly Detection Unitmentioning

confidence: 99%

See 2 more Smart Citations

HADES: a Hybrid Anomaly Detection System for Large-Scale Cyber-Physical Systems

Alwan

Ciupala

Baravalle³

et al. 2020

2020 Fifth International Conference on Fog and Mobile Edge Computing (FMEC)

View full text Add to dashboard Cite

Smart cities rely on large-scale heterogeneous distributed systems known as Cyber-Physical Systems (CPS).Information systems based on CPS typically analyse a massive amount of data collected from various data sources that operate under noisy and dynamic conditions. How to determine the quality and reliability of such data is an open research problem that concerns the overall system safety, reliability and security. Our research goal is to tackle the challenge of real-time data quality assessment for large-scale CPS applications with a hybrid anomaly detection system. In this paper we describe the architecture of HADES, our Hybrid Anomaly DEtection System for sensors data monitoring, storage, processing, analysis, and management. Such data will be filtered with correlation-based outlier detection techniques, and then processed by predictive analytics for anomaly detection.

show abstract

Section: Data Quality and Anomaly Detectionmentioning

confidence: 99%

Section: ) the Predictive Analytics-based Anomaly Detection Unitmentioning

confidence: 99%

Section: ) the Predictive Analytics-based Anomaly Detection Unitmentioning

confidence: 99%

See 1 more Smart Citation

HADES: a Hybrid Anomaly Detection System for Large-Scale Cyber-Physical Systems

Alwan

Ciupala

Baravalle³

et al. 2020

2020 Fifth International Conference on Fog and Mobile Edge Computing (FMEC)

View full text Add to dashboard Cite

show abstract

“…Recent advances in data analytics and data mining provide techniques that can appropriately address the complex dynamics of sensor networks, i.e. processing large data volumes and accounting for spatio-temporal information (Nanni et al 2008) (Appice et al 2014).…”

Section: Introductionmentioning

confidence: 99%

Spatial autocorrelation and entropy for renewable energy forecasting

Ceci

Corizzo

Malerba

et al. 2019

Data Min Knowl Disc

Self Cite

View full text Add to dashboard Cite

In renewable energy forecasting, data are typically collected by geographically distributed sensor networks, which poses several issues. i) Data represent physical properties that are subject to concept drift, i.e., their characteristics could change over time. To address the concept drift phenomenon, adaptive online learning methods should be considered. ii) The error distribution is typically non-Gaussian. Therefore, traditional quality performance criteria during training, like the mean-squared error, are less suitable. In the literature, entropy-based criteria have been proposed to deal with this problem. iii) Spatially-located sensors introduce some form of autocorrelation, that is, values collected by sensors show a correlation strictly due to their relative spatial proximity. Although all these issues have already been investigated in the literature, they have not been investigated in combination. In this paper, we propose a new method which learns artificial neural networks by addressing all these issues. The method performs online adaptive training and enriches the entropy measures with spatial information of the data, in order to take into account spatial autocorrelation. Experimental results on two photovoltaic power production datasets are clearly favorable for entropy-based measures that take into account spatial autocorrelation, also when compared with state-of-the art methods.

show abstract

Towards Real-Time Processing of Massive Spatio-temporally Distributed Sensor Data: A Sequential Strategy Based on Kriging

Lorkowski

Brinkhoff

2015

Lecture Notes in Geoinformation and Cartography

View full text Add to dashboard Cite

Data Mining Techniques in Sensor Networks

Cited by 6 publications

References 5 publications

HADES: a Hybrid Anomaly Detection System for Large-Scale Cyber-Physical Systems

HADES: a Hybrid Anomaly Detection System for Large-Scale Cyber-Physical Systems

Spatial autocorrelation and entropy for renewable energy forecasting

Towards Real-Time Processing of Massive Spatio-temporally Distributed Sensor Data: A Sequential Strategy Based on Kriging

Contact Info

Product

Resources

About