Abstract ：

Deep learning technology can improve sensing efficiency and has the ability to discover potential patterns in data; the efficiency of user behavior recognition in the field of smart homes has been further improved, making the recognition process more intelligent and humanized. This paper analyzes the optical sensors commonly used in smart homes and their working principles through case studies and explores the technical framework of user behavior recognition based on optical sensors. At the same time, CiteSpace (Basic version 6.2.R6) software is used to visualize and analyze the related literature, elaborate the main research hotspots and evolutionary changes of optical sensor-based smart home user behavior recognition, and summarize the future research trends. Finally, fully utilizing the advantages of cloud computing technology, such as scalability and on-demand services, combining typical life situations and the requirements of smart home users, a smart home data collection and processing technology framework based on elderly fall monitoring scenarios is designed. Based on the comprehensive research results, the application and positive impact of optical sensors in smart home user behavior recognition were analyzed, and inspiration was provided for future smart home user experience research.

Keyword ：

intelligent sensing user behavior recognition smart home cloud computing deep learning optical sensors

Cite：

Copy from the list or Export to your reference management。

Abstract ：

Traffic violation is one of the leading causes of traffic crashes. In the context of global aging, it is important to study traffic violations by elderly drivers for improving traffic safety in preparation for a worldwide aging population. In this study, a hybrid approach of Latent Class Analysis (LCA) and XGBoost based SHAP is proposed to identify hidden clusters and to understand the key contributing factors on the severity of traffic violations by elderly drivers, based on the police-reported traffic violation dataset of Beijing (China). First, LCA is applied to segment the dataset into several latent homogeneous clusters, then XGBoost based SHAP is established on each cluster to identify feature contributions and the interaction effects of the key contributing factors on the severity of traffic violations by elderly drivers. Two comparison groups were set up to analyze factors, which are responsible for the different severities of traffic violations. The results show that elderly drivers can be classified into four groups by age, urban or not, license, and season; factors such as less annual number of traffic violations, national & provincial highway, night and winter are key contributing factors for higher severity of traffic violations, which are consistent with common cognition; key contributing factors for all clusters are similar but not identical, for example, more annual number of traffic violations contribute to more severe violation for all clusters except for Cluster 2; some factors which are not key contributing factors may affect the severity of traffic violations when they are combined with other factors, for example, the combination of lower annual number of traffic violations and county & township highway contributes to more severe violation for Cluster 1. These findings can help government to formulate targeted countermeasures to decrease the severity of traffic violations by specific elderly groups and improve road service for the driving population.

Keyword ：

unobserved heterogeneity XGBoost based SHAP latent class analysis Severity of traffic violations elderly drivers

Cite：

Copy from the list or Export to your reference management。

Abstract ：

The causes of traffic violations by elderly drivers are different from those of other age groups. To reduce serious traffic violations that are more likely to cause serious traffic crashes, this study divided the severity of traffic violations into three levels (i.e., slight, ordinary, severe) based on point deduction, and explore the patterns of serious traffic violations (i.e., ordinary, severe) using multi-source data. This paper designed an interpretable machine learning framework, in which four popular machine learning models were enhanced and compared. Specifically, adaptive synthetic sampling method was applied to overcome the effects of imbalanced data and improve the prediction accuracy of minority classes (i.e., ordinary, severe); multi-objective feature selection based on NSGA-II was used to remove the redundant factors to increase the computational efficiency and make the patterns discovered by the explainer more effective; Bayesian hyperparameter optimization aimed to obtain more effective hyperparameters combination with fewer iterations and boost the model adaptability. Results show that the proposed interpretable machine learning framework can significantly improve and distinguish the performance of four popular machine learning models and two post-hoc interpretation methods. It is found that six of the top ten important factors belong to multi-scale built environment attributes. By comparing the results of feature contribution and interaction effects, some findings can be summarized: ordinary and severe traffic violations have some identical influencing factors and interactive effects; have the same influencing factors or the same combinations of influencing factors, but the values of the factors are different; have some unique influencing factors and unique combinations of influencing factors.

Keyword ：

Elderly drivers Multi-scale built environment Traffic violations Multi-source data Interpretable machine learning framework

Cite：

Copy from the list or Export to your reference management。

Abstract ：

Amid the escalating global climatic challenges, hydrological risks significantly influence human settlement patterns, underscoring the imperative for an in-depth comprehension of hydrological change's ramifications on human migration. However, predominant research has been circumscribed to the national level. The study delves into the nonlinear effects of hydrological risks on migration dynamics in 46,776 global subnational units. Meanwhile, leveraging remote sensing, we procured globally consistent metrics of hydrological intrusion exposure, offering a holistic risk assessment encompassing hazard, exposure, and vulnerability dimensions, thus complementing previous work. Here, we show that exposure is the primary migration driver, surpassing socioeconomic factors. Surrounding disparities further intensified exposure's impact. Vulnerable groups, especially the economically disadvantaged and elderly, tend to remain in high-risk areas, with the former predominantly migrating within proximate vicinities. The nonlinear analysis delineates an S-shaped trajectory for hydrological exposure, transitioning from resistance to migration and culminating in entrapment, revealing dependence on settlement resilience and adaptability. Hydrological risks drive migration more than socioeconomic factors. Vulnerable groups often stay in high-risk areas or migrate nearby. The study reveals an S-shaped migration pattern influenced by settlement resilience and adaptability.

Cite：

Copy from the list or Export to your reference management。

Abstract ：

The phenomenon of population aging is escalating, and China is rapidly reaching 100 million empty-nesters as a result of the country's uneven regional economic development and declining population growth. We created an emergency robot system for detecting geriatric behavior due to the issue that empty-nesters may experience acute disease or other emergency situation without prompt medical attention. To accomplish our goals, we employ the YOLO V3 algorithm to recognize persons in a complicated setting. Meanwhile we use the AlphaPose model firstly in order to detect the joint points of the elderly. And at that basis, we use the ST-GCN models to complete the posture estimation and detection, finish the categorization of postures. In this way, we could determine whether the elderly are in an emergency scenario or not. Also, the M5StickC open-source bracelets are used to improve the accuracy of the emergency detection. They were used concurrently to measure information including acceleration to assist in determining whether the elderly person is in emergency at the same time. Once the emergency situation has been verified, a robot car will be sent to quickly bring first aid supplies to the elderly. This approach can prevent losing the initial opportunity to perform first aid. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.

Keyword ：

Fall detection Image recognition Robots Population statistics Regional planning Drug delivery

Cite：

Copy from the list or Export to your reference management。

Abstract ：

Exposure to green environments is crucial for human health. However, urbanization has reduced the contact of urban residents with natural environments, causing a mismatch between the supply and demand for green exposure. Research in this field is hindered by the lack of long-term, reliable data sources and methodologies, leading to insufficient consideration of temporal variations in green exposure. This study presented a comprehensive methodology for assessing green exposure at a fine scale utilizing satellite images for urban tree canopy identification. We conducted a case study in the core area of Beijing from 2010 to 2020 and examined the effects of urban renewal and alleviation efforts. The results revealed a slight decrease in green exposure for the elderly over the decade, with minimal changes in equity. In contrast, green exposure for children has increased, with increasing inequality. Moreover, urban renewal has improved green exposure for nearly half of the low-supply blocks. However, a significant mismatch was observed between supply and demand for blocks with increased demand but limited supply. This study enhances the assessment of green exposure and provides guidance for planning and constructing a "Green Equal City".

Keyword ：

Green exposure Green space Accessibility Supply -demand Environmental equity

Cite：

Copy from the list or Export to your reference management。

Abstract ：

To examine the bus travel behaviour of the elderly in the context of the COVID-19 pandemic, this study analysed the mechanisms influencing the elderly’s risk perceptions regarding behavioural intention towards bus travel whilst focusing on the normalisation stage of pandemic prevention and control. Based on the theory of planned behaviour, a structural equation model of the elderly’s bus travel intention was constructed. The interactions among six factors—including attitudes, subjective norms, perceived behavioural control, cognitive risk perception, affective risk perception and the behavioural intention of the elderly’s bus travel—were quantitatively analysed. Valid sample data were used for empirical research. The results of this study show that perceived behavioural control, attitudes and subjective norms have a significant positive impact on the behavioural intentions of the elderly’s bus travel during the normalisation stage of pandemic prevention and control, with perceived behavioural control being the most influential factor. Moreover, perceived behavioural control also has a significant positive impact on attitudes, which indirectly influences behavioural intention. Cognitive risk perception has a direct and significant negative impact on attitudes, perceived behavioural control and subjective norms; however, affective risk perception only has a significant negative impact on subjective norms. Additionally, there is a positive correlation between the two, with both indirectly and negatively influencing the behavioural intentions of the elderly’s bus travel. This study can provide a basis for the formulation and improvement of pandemic prevention measures for bus travel during the normalisation stage of pandemic prevention and control to safeguard the elderly’s bus travel rights © This open access article is published under a Creative Commons [Attribution-NonCommercial-NoDerivatives 4.0 International] license. https://creativecommons.org/licenses/by-nc-nd/4.0/

Keyword ：

Urban traffic Structural equation model COVID-19 Elderly Theory of planned behaviour Risk perception

Cite：

Copy from the list or Export to your reference management。

Abstract ：

Fall detection is of great significance to the elderly. The wearable fall detection device can collect real-time data to identify falling events, thereby helping to protect the elderly from suffering further injuries. The limited sensor data from older adults in the SisFall dataset is insufficient for training a fall detection classifier that is specifically tailored to older adults. This paper proposes a fall detection method based on online transfer learning. The method uses the weighted online sequential extreme learning machine with a forgetting factor as an online classifier, which can effectively update the model in real time based on continuously collected data, thereby improving the accuracy of fall detection among elder adults. By dynamically updating the combination weights of offline classifiers and online classifiers to transfer source domain knowledge to the target domain, the proposed method improves classification accuracy in the target domain. Moreover, we incorporate concept drift detection to adapt to changes in the data distribution over time. Experimental results show that the improved algorithm has a higher online accuracy. © 2023 Technical Committee on Control Theory, Chinese Association of Automation.

Keyword ：

E-learning Fall detection Classification (of information) Machine learning Knowledge acquisition

Cite：

Copy from the list or Export to your reference management。

Abstract ：

With the deepening of aging, chronic diseases of the elderly are the main burden of disease in most countries in the world. The prevalence of chronic diseases in urban areas in China is as high as 75%. Many elderly people use multiple drugs for a long time. Home self-medication problems occur frequently. In order to alleviate this problem to a certain extent, knowledge graph technology and a deep learning model are used to design a home self-medication question-answering system for the elderly and their caregivers. Explore a feasible way of providing automated online consultation intelligent services. In this paper, we have collected medication as well as professional Q&A (question and answer) data in the field of aging health, and constructed a knowledge graph that meets the characteristics of medication use in the elderly. Based on the matching rules in the question judging module, the problems entered by users are classified. For professional knowledge related to diseases and medications of the elderly, the question-answering system uses the knowledge graph to search for answers. For other basic knowledge related to elderly health, the system uses the BERT model to vectorize its users’ questions, then matches the questions by calculating cosine similarity, thus finding the corresponding answers. The system adds the Seq2Seq model as a supplement to the answer retrieval method of the knowledge graph. The testing results shows that the system provides online consultation services more accurately and efficiently for home self-medication for the elderly and their caregivers. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd 2023.

Keyword ：

Template matching Diseases Deep learning Knowledge graph

Cite：

Copy from the list or Export to your reference management。

Abstract ：

Objective. The purpose is to understand the depression status of the elderly in the community, explore its influencing factors, formulate a comprehensive psychological intervention plan according to the influencing factors, implement demonstration psychological intervention, and evaluate and feedback the effect, so as to provide a reference for improving the mental health of the elderly. Method. In order to make the output of different emotional data in LSTM more discriminative, a method to dynamically filter the output of LSTM is proposed. Combining the methods of Attention-LSTM, time-dimensional AI attention, and feature-dimensional AI attention, the best model in this paper is obtained. The multistage stratified cluster sampling method was used to conduct a questionnaire survey on the elderly aged 60 and above in a certain area, including the general demographic characteristics questionnaire of the elderly, the self-rating scale of mental health symptoms, and the health self-management ability of adults. All data were entered into a database using Excel software, and SPSS 19.0 statistical software was used for statistical analysis. Results/Discussion. The detection rate of depression (GDS >= 11 points) among the elderly in a community in a certain area was 39.38%. Multivariate logistic regression analysis showed that family history of mental illness, more negative life events, decreased ability of daily living, living alone, and suffering from physical diseases in the past six months were the risk factors for depression in the elderly. Community health education can partially alleviate depression in the elderly. The detection rate and degree of depression of the elderly in the comprehensive psychological intervention group were significantly lower than those in the control group, and the difference was statistically significant (P < 0:05).

Cite：

Copy from the list or Export to your reference management。