An improved extreme learning machine model for the prediction of human scenarios in smart homes

Liouane, Zaineb; Lemlouma, Tayeb; Roose, Philippe; Weis, Frédéric; Messaoud, Hassani

doi:10.1007/s10489-017-1062-5

Cited by 29 publications

(18 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Such networks of autonomous smart sensing devices are poised to advance the exchange of information in smart homes, offices, cities, and factories. [1][2][3][4] It is being argued that many aspects of our life will be mediated via 75 billion IoT devices by 2025, of which the majority will reside indoors. They will collect, communicate and process real-time data to optimize services and manufacturing processes, as well as to manage resources to reduce our energy consumption.…”

Section: Introductionmentioning

confidence: 99%

Dye-sensitized solar cells under ambient light powering machine learning: towards autonomous smart sensors for the internet of things

et al. 2020

View full text Add to dashboard Cite

show abstract

Section: Introductionmentioning

confidence: 99%

Dye-sensitized solar cells under ambient light powering machine learning: towards autonomous smart sensors for the internet of things

et al. 2020

View full text Add to dashboard Cite

show abstract

“…We also compared our proposed method with Liouane et al (17). Similar to them we have used the first three weeks of the eHealth dataset for the training phase and nine days for the test purposes.…”

Section: Action Detection and Predictionmentioning

confidence: 99%

“…Table 4 shows the results of the comparison by using three metrics of RMSE (Equation 14), cosine similarity (Equation 15) and percentage error (Equation 16) for each test day. Similar to Liouane et al study (17), for calculating RMSE, we first normalized the values using (Equation 13).…”

Section: Action Detection and Predictionmentioning

confidence: 99%

See 1 more Smart Citation

Elderly Action Prediction and Anomalous Activity Detection in Smart Homes through Profiling Residents’ Behavior

2019

View full text Add to dashboard Cite

Background: Elderly healthcare is one of the important issues in an aging society. Smart homes in healthcare domain help the elderly to be continuously monitored, instead of being under supervision at expensive health centers, and hence enable them to live independently. This service requires detecting and monitoring of residents' normal activities of daily living in smart homes. Objectives: By profiling residents' behavior and identifying changes in normal activities of the elderly over time, one can detect anomalous behavior and determine whether their health status declines. Hence, the possibility of preventive care for some elderly people or providing assistance to the elderly will be partly provided in case of occurrence of the anomalies. Methods: In this paper, first a method was proposed for detection and prediction of elderly activities by extracting several features from available information. In the second step, statistical measures were applied on the features to profile the elderly's behavior. The AdaBoost learning algorithm was used for detecting the anomalies and modeling normal/abnormal behavior. Results: For detection and prediction of the activity, the proposed method was tested using a dataset collected in the "eHealth Monitoring Open Data Project". The accuracy of 98.48% was obtained by considering features of start time, end time, duration, location, previous action, water, and electric device use. Anomalous behavior was detected in the same dataset with the average f-score of 90%. Conclusions: Results of the present study revealed that, the proposed methods are effective for detecting abnormal actions of the residents in smart homes to a fairly good level. This enables the elderly to live independently while being under continuous monitoring and significantly reduces the elderly health care costs.

show abstract

“…Besides, the ELM has been developed to solve the drawbacks of multilayer feedforward neural networks [7]. The ELM has various advantages such as unification of multiclassification, fast learning speed, minimal human intervention, ease of implementation and regression [8,9]. The ELM is computationally powerful single-hidden-layer feed-forward neural network, which is widely utilized in the several real-world problems due to its remarkable efficiency, simplicity, and impressive generalization performance [10].…”

Section: Introductionmentioning

confidence: 99%

Automotive Rank Based ELM Using Iterative Decomposition

Nagelli¹,

Ragala²,

B³

2019

IJIES

View full text Add to dashboard Cite

Nowadays, the Extreme learning machine (ELM) is playing a key role in machine intelligence and big data analytics due to its various advantages such as fast training rate, universal classification/regression and the capability of approximation. The standard ELM uses the Moore-Penrose generalized pseudo-inverse for solving the hidden layer activation matrix and also it identifies the output weights. Because of that, the standard ELM takes more time to train the features from the dataset. In ELM, scalability also considered as a one of the major concern while processing the large dataset. In order to overcome this concern, the Automotive Rank based ELM (AR-ELM) is proposed to obtain an effective tensor decomposition for diminishing the training time. Besides, the Bayesian approach is considered in this AR-ELM to remove the redundancy from the decomposed samples of the tensor. The major objective of this proposed AR-ELM is to process the large amount of dataset without depending on memory capacities. The recognition accuracy is improved by eliminating redundant information. The key idea of the AR-ELM is to reduce the training time while processing the huge dataset. The implementation and simulation of the AR-ELM is done in Spark Python 3.7. The performance of the AR-ELM is analysed in terms of accuracy, precision, recall and training time. The proposed methodology is compared with three existing methodologies such as basic ELM, ELM-TUCKER and ELM-PARAFAC. The recognition accuracy of the AR-ELM methodology with Hardlims activation function is 0.8879 for letter recognition dataset, it is high when compared to the basic ELM, ELM-TUCKER and ELM-PARAFAC that are 0.8102, 0.8375 and 0.834 respectively.

show abstract

An improved extreme learning machine model for the prediction of human scenarios in smart homes

Abstract: International audienc

Cited by 29 publications

References 29 publications

Dye-sensitized solar cells under ambient light powering machine learning: towards autonomous smart sensors for the internet of things

Dye-sensitized solar cells under ambient light powering machine learning: towards autonomous smart sensors for the internet of things

Elderly Action Prediction and Anomalous Activity Detection in Smart Homes through Profiling Residents’ Behavior

Automotive Rank Based ELM Using Iterative Decomposition

Contact Info

Product

Resources

About