Ensemble based Data Imputation at the Edge

Fountas, Panagiotis; Kolomvatsos, Kostas

doi:10.1109/ictai50040.2020.00150

Cited by 8 publications

(7 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…A similar approach, relying on the spatiotemporal correlations, has been adopted in [ 24 ], where a comparison among several data imputation techniques has been proposed. Yet, the authors in [ 25 ] propose a data imputation strategy that relies on the concept of “group opinion”. For instance, metrics, such as the Mahalanobis distance and the cosine similarity, are combined to evaluate the data replacement proposed by a group of peer devices.…”

Section: Related Workmentioning

confidence: 99%

“…While all the works discussed earlier present interesting and innovative techniques or frameworks to manage the problem of missing data, some important differences arise with respect to our proposal. First, in many works (see, e.g., [ 25 , 26 , 27 , 28 , 30 , 31 , 32 , 33 , 35 , 36 ]), despite the data coming from the IoT, the analysed imputation algorithms actually run on standard computer architectures (e.g., PCs, laptops, etc.). Moreover, the importance of tackling the missing data problem as close as possible to the devices is evident (e.g., at the Edge of the network [ 37 ]); thus, our assessments are carried out straight on the board of the devices.…”

Section: Related Workmentioning

confidence: 99%

“…Moreover, the importance of tackling the missing data problem as close as possible to the devices is evident (e.g., at the Edge of the network [ 37 ]); thus, our assessments are carried out straight on the board of the devices. Second, many existing experiments (see, e.g., [ 24 , 25 , 26 , 32 ]) are performed by simplistic data missing models, while we consider the problem of bursty missing values, which often arises when a sensor becomes unavailable for a certain (finite) period of time (a common situation for environmental sensors). Finally, compared to other works (see, e.g., [ 23 , 24 , 34 , 35 ]), which only account for one performance analysis (for instance, the imputation method accuracy), in our approach, we complement this analysis with a time assessment, which seems critical within real-time environmental data.…”

Section: Related Workmentioning

confidence: 99%

See 2 more Smart Citations

Embedded Data Imputation for Environmental Intelligent Sensing: A Case Study

Erhan

Mauro

Anjum

et al. 2021

Sensors

View full text Add to dashboard Cite

Recent developments in cloud computing and the Internet of Things have enabled smart environments, in terms of both monitoring and actuation. Unfortunately, this often results in unsustainable cloud-based solutions, whereby, in the interest of simplicity, a wealth of raw (unprocessed) data are pushed from sensor nodes to the cloud. Herein, we advocate the use of machine learning at sensor nodes to perform essential data-cleaning operations, to avoid the transmission of corrupted (often unusable) data to the cloud. Starting from a public pollution dataset, we investigate how two machine learning techniques (kNN and missForest) may be embedded on Raspberry Pi to perform data imputation, without impacting the data collection process. Our experimental results demonstrate the accuracy and computational efficiency of edge-learning methods for filling in missing data values in corrupted data series. We find that kNN and missForest correctly impute up to 40% of randomly distributed missing values, with a density distribution of values that is indistinguishable from the benchmark. We also show a trade-off analysis for the case of bursty missing values, with recoverable blocks of up to 100 samples. Computation times are shorter than sampling periods, allowing for data imputation at the edge in a timely manner.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Embedded Data Imputation for Environmental Intelligent Sensing: A Case Study

Erhan

Mauro

Anjum

et al. 2021

Sensors

View full text Add to dashboard Cite

show abstract

“…For instance, the Mahalanobis distance can offer a statistical view on the correlation of multivariate vectors being based on their covariance matrix. Such correlations can be also adopted to impute data in combination with outliers detection for creating efficient models that manage distributed data streams [17], [18]. Other statistical measures can be found in the Cook's distance [12], the leverage model [10], the χ 2 metric (it detects deviations from the multidimensional normality) and an extended version of the Mahalanobis distance [34].…”

Section: Related Workmentioning

confidence: 99%

Landmark based Outliers Detection in Pervasive Applications

Kolomvatsos

Anagnostopoulos

2021

2021 12th International Conference on Information and Communication Systems (ICICS)

Self Cite

View full text Add to dashboard Cite

The combination of the Internet of Things and the Edge Computing gives many opportunities to support innovative applications close to end users. Numerous devices present in both infrastructures can collect data upon which various processing activities can be performed. However, the quality of the outcomes may be jeopardized by the presence of outliers. In this paper, we argue on a novel model for outliers detection by elaborating on a 'soft' approach. Our mechanism is built upon the concepts of candidate and confirmed outliers. Any data object that deviates from the population is confirmed as an outlier only after the study of its sequence of magnitude values as new data are incorporated into our decision making model. We adopt the combination of a sliding with a landmark window model when a candidate outlier is detected to expand the sequence of data objects taken into consideration. The proposed model is fast and efficient as exposed by our experimental evaluation while a comparative assessment reveals its pros and cons.

show abstract

“…Como se puede apreciar, un único estudio primario realiza la comparativa de distintas técnicas de Inteligencia Computacional (IC) [9] y siete de ellos utilizan dichas técnicas para realizar el procesamiento de datos en soluciones particulares [9,20,21,22,23,24,25].…”

Section: Fuente: Autoría Propiaunclassified

Criterios de ubicación de algoritmos de inteligencia computacional en la web de las cosas: Revisión sistemática de la literatura

Parra

Zambrano

Cobos

2021

Investigación e Innovación en Ingenierías

View full text Add to dashboard Cite

Objetivo: Determinar y caracterizar el estado actual del conocimiento acerca de los criterios de ubicación del procesamiento de datos, utilizando técnicas de inteligencia computacional en un Ecosistema de Objetos Inteligentes de la Web de las Cosas. Metodología: La revisión sistemática que se presenta a continuación se basa en los estudios realizados por Petersen y Kitchenham , se plantearon cuatro preguntas de investigación, se aplicó el método PICOC para identificar las palabras clave, se planteó una cadena de búsqueda y cuatro motores de búsqueda, se plantearon los criterios de inclusión y exclusión de estudios primarios, así como los criterios de evaluación de la calidad, la estrategia de extracción de datos y el método de síntesis. Resultados: Se lograron responder las cuatro preguntas de investigación planteadas, encontrando que la mayoría de los estudios carecen de una implementación en las tres ubicaciones analizadas y un único estudio que compara el desempeño obtenido por un algoritmo de inteligencia computacional al procesar información en distintas ubicaciones del ecosistema. Conclusiones: Se demostró la necesidad de continuar realizando estudios en el área de la localización del procesamiento en ecosistemas inteligentes utilizando técnicas de inteligencia computacional para el procesamiento en distintas ubicaciones. Además, se evidencia una necesidad en hacer un mayor énfasis en la comparativa del rendimiento obtenido al realizar implementaciones teniendo en cuenta distintas técnicas de inteligencia computacional.

show abstract

Ensemble based Data Imputation at the Edge

Cited by 8 publications

References 33 publications

Embedded Data Imputation for Environmental Intelligent Sensing: A Case Study

Embedded Data Imputation for Environmental Intelligent Sensing: A Case Study

Landmark based Outliers Detection in Pervasive Applications

Criterios de ubicación de algoritmos de inteligencia computacional en la web de las cosas: Revisión sistemática de la literatura

Contact Info

Product

Resources

About