A sequential deep learning application for recognising human activities in smart homes

Liciotti, Daniele; Bernardini, Michele; Romeo, Luca; Frontoni, Emanuele

doi:10.1016/j.neucom.2018.10.104

Cited by 133 publications

(133 citation statements)

References 33 publications

Supporting

Mentioning

123

Contrasting

Unclassified

Order By: Relevance

“…One study [ 7 ] uses triplet ML models consisting of a Naïve Bayes Classifier (NBC), a Hidden Markov Model (HMM), and a Conditional Random Field (CRF). These models are effective in average noisy data and recognize activities and compare the advanced version of this study in [ 15 ] with Support Vector Machines (SVM).…”

Section: Literature Reviewmentioning

confidence: 99%

“…The scheme outperforms when the number of users is given through the mining stage. In [ 7 ], the authors use the DL model based on LSTM, which classifies and learns human activities in the smart home environment. LSTM shows a viable and significant improvement in the recognition of human activities.…”

Section: Literature Reviewmentioning

confidence: 99%

“…Since SDC is not a new topic, a number of studies have been proposed. However, some of these studies are restricted to use the WSN continuously and collect the data with no missing of useful activities [ 7 ]. Moreover, ambient wireless sensors, which incorporate light, temperature, door, and contact positioning sensors, are low-cost and battery- powered sensor.…”

Section: Introductionmentioning

confidence: 99%

“…Besides, such approaches are unsatisfactory in front of the dynamic environment. However, these algorithms have their own advantages despite that none of them overcome other problems in different domains [ 7 ]. The rising attraction in Deep Learning (DL) model for HAD has shown significant improvement in activity detection via prediction.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Scheduling Sensor Duty Cycling Based on Event Detection Using Bi-Directional Long Short-Term Memory and Reinforcement Learning

Diyan

Khan

Silva

et al. 2020

Sensors

View full text Add to dashboard Cite

A smart home provides a facilitated environment for the detection of human activity with appropriate Deep Learning algorithms to manipulate data collected from numerous sensors attached to various smart things in a smart home environment. Human activities comprise expected and unexpected behavior events; therefore, detecting these events consisting of mutual dependent activities poses a key challenge in the activities detection paradigm. Besides, the battery-powered sensor ubiquitously and extensively monitors activities, disputes, and sensor energy depletion. Therefore, to address these challenges, we propose an Energy and Event Aware-Sensor Duty Cycling scheme. The proposed model predicts the future expected event using the Bi-Directional Long-Short Term Memory model and allocates Predictive Sensors to the predicted event. To detect the unexpected events, the proposed model localizes a Monitor Sensor within a cluster of Hibernate Sensors using the Jaccard Similarity Index. Finally, we optimize the performance of our proposed scheme by employing the Q-Learning algorithm to track the missed or undetected events. The simulation is executed against the conventional Machine Learning algorithms for the sensor duty cycle, scheduling to reduce the sensor energy consumption and improve the activity detection accuracy. The experimental evaluation of our proposed scheme shows significant improvement in activity detection accuracy from 94.12% to 96.12%. Besides, the effective rotation of the Monitor Sensor significantly improves the energy consumption of each sensor with the entire network lifetime.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Section: Literature Reviewmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Scheduling Sensor Duty Cycling Based on Event Detection Using Bi-Directional Long Short-Term Memory and Reinforcement Learning

Diyan

Khan

Silva

et al. 2020

Sensors

View full text Add to dashboard Cite

show abstract

“…In these papers, the authors made use of different types of sensors, including motion sensors [18,25,28,[141][142][143][144][145]; temperature sensors [28,40,73,143,144]; wireless sensor networks [21,40,141,145]; door sensors [25,143]; smartphone inertial sensors [146] and a smartphone application [36]; cameras [18]; a two-dimensional acoustic array [27]; daily activity recognition sensors [28]; actuators [143]; tactile sensors, power meters, and microphones in the ceiling [144]; non-wearable sensors [147]; unobtrusive sensors [9]; environmental sensors [73,142]; weather sensors [12]; WiFi-enabled sensors for food nutrition quantification [36]; and binary sensors [148]. In the scientific papers selected and summarized in Table S16, the reasons for using Deep Learning techniques integrated with sensor devices in smart buildings were mainly related to human activity recognition [9,18,25,27,28,73,142,143,…”

Section: Deep Learning Techniquesmentioning

confidence: 99%

A Review of the Recent Developments in Integrating Machine Learning Models with Sensor Devices in the Smart Buildings Sector with a View to Attaining Enhanced Sensing, Energy Efficiency, and Optimal Building Management

et al. 2019

View full text Add to dashboard Cite

Lately, many scientists have focused their research on subjects like smart buildings, sensor devices, virtual sensing, buildings management, Internet of Things (IoT), artificial intelligence in the smart buildings sector, improving life quality within smart homes, assessing the occupancy status information, detecting human behavior with a view to assisted living, maintaining environmental health, and preserving natural resources. The main purpose of our review consists of surveying the current state of the art regarding the recent developments in integrating supervised and unsupervised machine learning models with sensor devices in the smart building sector with a view to attaining enhanced sensing, energy efficiency and optimal building management. We have devised the research methodology with a view to identifying, filtering, categorizing, and analyzing the most important and relevant scientific articles regarding the targeted topic. To this end, we have used reliable sources of scientific information, namely the Elsevier Scopus and the Clarivate Analytics Web of Science international databases, in order to assess the interest regarding the above-mentioned topic within the scientific literature. After processing the obtained papers, we finally obtained, on the basis of our devised methodology, a reliable, eloquent and representative pool of 146 papers scientific works that would be useful for developing our survey. Our approach provides a useful up-to-date overview for researchers from different fields, which can be helpful when submitting project proposals or when studying complex topics such those reviewed in this paper. Meanwhile, the current study offers scientists the possibility of identifying future research directions that have not yet been addressed in the scientific literature or improving the existing approaches based on the body of knowledge. Moreover, the conducted review creates the premises for identifying in the scientific literature the main purposes for integrating Machine Learning techniques with sensing devices in smart environments, as well as purposes that have not been investigated yet.

show abstract

Deep learning support for intelligent transportation systems

Ibáñez

Contreras-Castillo

Zeadally

2020

Trans Emerging Tel Tech

View full text Add to dashboard Cite

Intelligent Transportation Systems (ITS) help improve the ever-increasing vehicular flow and traffic efficiency in urban traffic to reduce the number of accidents. The generation of massive amounts of data generated by all the digital devices connected to the transportation network enables the creation of datasets to perform an in-depth analysis of the data using deep learning methods. Such methods can help predict traffic performance, automated traffic light management, lane detection, and identifying objects near vehicles to increase the safety and efficiency of ITS. We discuss some of the challenges that need to be solved to achieve seamless integration between ITS and deep learning methods to address issues such as (1) improving traffic flow/transportation logistics, (2) predicting best routes for the transportation of goods, (3) optimal fuel consumption, (4) intelligent environmental conditions perception, (5) traffic speed management, and accident prevention.

show abstract

A sequential deep learning application for recognising human activities in smart homes

Cited by 133 publications

References 33 publications

Scheduling Sensor Duty Cycling Based on Event Detection Using Bi-Directional Long Short-Term Memory and Reinforcement Learning

Scheduling Sensor Duty Cycling Based on Event Detection Using Bi-Directional Long Short-Term Memory and Reinforcement Learning

A Review of the Recent Developments in Integrating Machine Learning Models with Sensor Devices in the Smart Buildings Sector with a View to Attaining Enhanced Sensing, Energy Efficiency, and Optimal Building Management

Deep learning support for intelligent transportation systems

Contact Info

Product

Resources

About