An Activity-Embedding Approach for Next-Activity Prediction in a Multi-User Smart Space

Kim, Young-Gi; An, Jinung; Lee, Minseok; Lee, Younghee

doi:10.1109/smartcomp.2017.7946985

Cited by 15 publications

(9 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Because RNN/LSTM can deal with sequential data, it has been used not only to recognize and classify the current behavior but also to predict the next behavior or the time at which the next behavior occurs. In particular, in the case of the study of Kim et al [55], they used LSTM to recognize 23 activities of 7 participants and predict their next behavior in a multi-user smart space. Understanding the domain of smart homes and considering a multi-user environment was different from other studies.…”

Section: Data Extractionmentioning

confidence: 99%

“…The studies using sensor data can be seen in Table 6. Sensor data for activities Touch sensor, Tilt sensor, Height sensor, Weight sensor, Reed switch, Infrared sensor [52] Multi-user activity data Occupancy sensor, Ambient sensor (temperature, brightness, Humidity, Sound), Screen sensor, Door sensor, Seat occupancy sensor [55] ADL data Ultrasonic Positioning System (Position), Bluetooth watt checker (power consumption), CT Sensor (power consumption), ECHONET (appliance status), Motion sensor (motion detect) [64] User authentication ShakeLogin data Smartphone internal sensors: accelerometer gyroscope, rotation vector [48] Figure 10 shows the analysis of the public datasets used in the studies using public datasets. In total, 47.5% of the studies used the CASAS project dataset.…”

Section: Rq3: How Is Dataset Collected Analyzed and Used By Each Study?mentioning

confidence: 99%

See 1 more Smart Citation

Deep Learning (CNN, RNN) Applications for Smart Homes: A Systematic Review

Jiménez

Villalba-Mora

2022

Computers

View full text Add to dashboard Cite

In recent years, research on convolutional neural networks (CNN) and recurrent neural networks (RNN) in deep learning has been actively conducted. In order to provide more personalized and advanced functions in smart home services, studies on deep learning applications are becoming more frequent, and deep learning is acknowledged as an efficient method for recognizing the voices and activities of users. In this context, this study aims to systematically review the smart home studies that apply CNN and RNN/LSTM as their main solution. Of the 632 studies retrieved from the Web of Science, Scopus, IEEE Explore, and PubMed databases, 43 studies were selected and analyzed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) methodology. In this paper, we examine which smart home applications CNN and RNN/LSTM are applied to and compare how they were implemented and evaluated. The selected studies dealt with a total of 15 application areas for smart homes, where activity recognition was covered the most. This study provides essential data for all researchers who want to apply deep learning for smart homes, identifies the main trends, and can help to guide design and evaluation decisions for particular smart home services.

show abstract

Section: Data Extractionmentioning

confidence: 99%

Section: Rq3: How Is Dataset Collected Analyzed and Used By Each Study?mentioning

confidence: 99%

Deep Learning (CNN, RNN) Applications for Smart Homes: A Systematic Review

Jiménez

Villalba-Mora

2022

Computers

View full text Add to dashboard Cite

show abstract

“…We also find some examples of deep learning approaches (e.g. LSTM (Long Short Term Memory neural networks)) applied to activity prediction, as in [17,18,19 [20] a two-step module for activity prediction in homes. In the first step, sequence pattern mining is applied in order to discover temporal patterns of activity sequences.…”

Section: Machine Learning Prediction For Activity Predictionmentioning

confidence: 99%

“…• H i , the Hour of the day when the activity occurs. These values are discretized into 6 ranges of hours: [0, 3], [4,7], [8,11], [12,15], [16,19], and [20,23]; • D i , the Day of the week when the activity occurs, from 1 (Monday) to 7 (Sunday).…”

Section: Dbn For Activity and Availability Predictionmentioning

confidence: 99%

Psines

Cumin

Lefebvre

Ramparany

et al. 2020

ACM Trans. Auton. Adapt. Syst.

View full text Add to dashboard Cite

Autonomy and adaptability are essential components of ambient intelligence. For example, in smart homes, proactive acting and occupants advising, adapted to current and future contexts of living, are essential to go beyond limitations of previous domotic services. To reach such autonomy and adaptability, ambient systems need to automatically grasp their users’ ambient context. In particular, users’ activities and availabilities for communication are valuable pieces of contextual information that can help such systems to adapt to user needs and behaviours. While significant research work exists on activity recognition in homes, less attention has been given to prediction of future activities, as well as to availability recognition and prediction in general. In this article, we investigate several Dynamic Bayesian Network (DBN) architectures for activity and availability prediction of occupants in homes, including our novel model, called Past SItuations to predict the NExt Situation (PSINES). This predictive architecture utilizes context information, sensor event aggregations, and latent user cognitive states to accurately predict future home situations based on previous situations. We experimentally evaluate PSINES, as well as intermediate DBN architectures, on multiple state-of-the-art datasets, with prediction accuracies of up to 89.52% for activity and 82.08% for availability on the Orange4Home dataset.

show abstract

“…Therefore, it is essential to identify and distinguish the activities that occur within a multi-occupant environment [ 12 ], since multi-occupancy scenarios are far more realistic. Additional challenges exist when dealing with such environments because existing sensors are incapable of distinguishing who has activated them in the absence of any tagging system [ 13 , 14 , 15 ]. Considering the negative connotations and privacy issues associated with wearable tags/sensors, the wearable sensors are not widely accepted by older adults [ 6 , 7 , 8 , 9 ].…”

Section: Introductionmentioning

confidence: 99%

Exploring Entropy Measurements to Identify Multi-Occupancy in Activities of Daily Living

Howedi

Lotfi

Pourabdollah

2019

Entropy

View full text Add to dashboard Cite

Human Activity Recognition (HAR) is the process of automatically detecting human actions from the data collected from different types of sensors. Research related to HAR has devoted particular attention to monitoring and recognizing the human activities of a single occupant in a home environment, in which it is assumed that only one person is present at any given time. Recognition of the activities is then used to identify any abnormalities within the routine activities of daily living. Despite the assumption in the published literature, living environments are commonly occupied by more than one person and/or accompanied by pet animals. In this paper, a novel method based on different entropy measures, including Approximate Entropy (ApEn), Sample Entropy (SampEn), and Fuzzy Entropy (FuzzyEn), is explored to detect and identify a visitor in a home environment. The research has mainly focused on when another individual visits the main occupier, and it is, therefore, not possible to distinguish between their movement activities. The goal of this research is to assess whether entropy measures can be used to detect and identify the visitor in a home environment. Once the presence of the main occupier is distinguished from others, the existing activity recognition and abnormality detection processes could be applied for the main occupier. The proposed method is tested and validated using two different datasets. The results obtained from the experiments show that the proposed method could be used to detect and identify a visitor in a home environment with a high degree of accuracy based on the data collected from the occupancy sensors.

show abstract

An Activity-Embedding Approach for Next-Activity Prediction in a Multi-User Smart Space

Cited by 15 publications

References 13 publications

Deep Learning (CNN, RNN) Applications for Smart Homes: A Systematic Review

Deep Learning (CNN, RNN) Applications for Smart Homes: A Systematic Review

Psines

Exploring Entropy Measurements to Identify Multi-Occupancy in Activities of Daily Living

Contact Info

Product

Resources

About