PurposeIn 2018, an artificial intelligence (AI) interview platform was introduced and adopted by companies in Korea. This study aims to explore the perspectives of applicants who have experienced an ...AI-based interview through this platform and examines the opinions of companies, a platform developer and academia.Design/methodology/approachThis study uses a phenomenological approach. The participants, who had recent experience of AI video interviews, were recruited offline and online. Eighteen job applicants in their 20s, two companies that have adopted this interview platform, a software developer who created the platform and three professors participated in the study. To collect data, focus group interviews and in-depth interviews were conducted.FindingsAs a result, all of them believed that an AI-based interview was more efficient than a traditional one in terms of cost and time savings and is likely to be adopted by more companies in the future. They pointed to the possibility of data bias requiring an improvement in AI accountability. Applicants perceived an AI-based interview to be better than traditional evaluation procedures in procedural fairness, objectivity and consistency of algorithms. However, some applicants were dissatisfied about being assessed by AI. Digital divide and automated inequality were recurring themes in this study.Originality/valueThe study is important, as it addresses the real application of AI in detail, and a case study of smart hiring tools would be valuable in finding the practical and theoretical implications of such hiring in the fields of employment and AI.
Many studies have examined the negative impact on smartphone addiction in adolescents. Recent concerns have focused on predictors of smartphone addiction. This study aimed to investigate the ...association of adolescents' smartphone addiction with family environment (specifically, domestic violence and parental addiction). We further investigated whether self-control and friendship quality, as predictors of smartphone addiction, may reduce the observed risk.
We used the 2013 national survey on internet usage and utilization data from the National Information Agency of Korea. Information on exposure and covariates included self-reported experience of domestic violence and parental addiction, sociodemographic variables, and other variables potentially related to smartphone addiction. Smartphone addiction was estimated using a smartphone addiction proneness scale, a standardized measure developed by national institutions in Korea.
Adolescents who had experienced domestic violence (OR = 1.74; 95% CI: 1.23-2.45) and parental addiction (OR = 2.01; 95% CI: 1.24-3.27) were found to be at an increased risk for smartphone addiction after controlling for all potential variables. Furthermore, on classifying adolescents according to their level of self-control and friendship quality the association between domestic violence and parental addiction, and smartphone addiction was found to be significant in the group with adolescents with lower levels of self-control (OR = 2.87; 95% CI: 1.68-4.90 and OR = 1.95; 95% CI: 1.34-2.83) and friendship quality (OR = 2.33; 95% CI: 1.41-3.85 and OR = 1.83; 95% CI: 1.26-2.64).
Our findings suggest that family dysfunction was significantly associated with smartphone addiction. We also observed that self-control and friendship quality act as protective factors against adolescents' smartphone addiction.
Microrobots facilitate targeted therapy due to their small size, minimal invasiveness, and precise wireless control. A degradable hyperthermia microrobot (DHM) with a 3D helical structure is ...developed, enabling actively controlled drug delivery, release, and hyperthermia therapy. The microrobot is made of poly(ethylene glycol) diacrylate (PEGDA) and pentaerythritol triacrylate (PETA) and contains magnetic Fe3O4 nanoparticles (MNPs) and 5‐fluorouracil (5‐FU). Its locomotion is remotely and precisely controlled by a rotating magnetic field (RMF) generated by an electromagnetic actuation system. Drug‐free DHMs reduce the viability of cancer cells by elevating the temperature under an alternating magnetic field (AMF), a hyperthermic effect. 5‐FU is released from the proposed DHMs in normal‐, high‐burst‐, and constant‐release modes, controlled by the AMF. Finally, actively controlled drug release from the DHMs in normal‐ and high‐burst‐release mode results in a reduction in cell viability. The reduction in cell viability is of greater magnitude in high‐burst‐ than in normal‐release mode. In summary, biodegradable DHMs have potential for actively controlled drug release and hyperthermia therapy.
A degradable hyperthermia microrobot (DHM), which encapsulates 5‐fluorouracil and Fe3O4 nanoparticles is developed and demonstrates the potential for actively controlled drug release and hyperthermia therapy. The locomotion of the DHM is remotely controlled by a rotating magnetic field. The alternating magnetic field not only increases the temperature of the DHMs, but also actively controls the drug release from the DHMs.
The use of deep learning and machine learning (ML) in medical science is increasing, particularly in the visual, audio, and language data fields. We aimed to build a new optimized ensemble model by ...blending a DNN (deep neural network) model with two ML models for disease prediction using laboratory test results. 86 attributes (laboratory tests) were selected from datasets based on value counts, clinical importance-related features, and missing values. We collected sample datasets on 5145 cases, including 326,686 laboratory test results. We investigated a total of 39 specific diseases based on the International Classification of Diseases, 10th revision (ICD-10) codes. These datasets were used to construct light gradient boosting machine (LightGBM) and extreme gradient boosting (XGBoost) ML models and a DNN model using TensorFlow. The optimized ensemble model achieved an F1-score of 81% and prediction accuracy of 92% for the five most common diseases. The deep learning and ML models showed differences in predictive power and disease classification patterns. We used a confusion matrix and analyzed feature importance using the SHAP value method. Our new ML model achieved high efficiency of disease prediction through classification of diseases. This study will be useful in the prediction and diagnosis of diseases.
Laser‐induced graphene (LIG) is a newly emerging 3D porous material produced when irradiating a laser beam on certain carbon materials. LIG exhibits high porosity, excellent electrical conductivity, ...and good mechanical flexibility. Predesigned LIG patterns can be directly fabricated on diverse carbon materials with controllable microstructure, surface property, electrical conductivity, chemical composition, and heteroatom doping. This selective, low‐cost, chemical‐free, and maskless patterning technology minimizes the usage of raw materials, diminishes the environmental impact, and enables a wide range of applications ranging from academia to industry. In this review, the recent developments in 3D porous LIG are comprehensively summarized. The mechanism of LIG formation is first introduced with a focus on laser‐material interactions and material transformations during laser irradiation. The effects of laser types, fabrication parameters, and lasing environment on LIG structures and properties are thoroughly discussed. The potentials of LIG for advanced applications including biosensors, physical sensors, supercapacitors, batteries, triboelectric nanogenerators, and so on are also highlighted. Finally, current challenges and future prospects of LIG research are discussed.
Laser‐induced graphene (LIG) is an emerging porous material produced when irradiating a laser beam on certain carbon materials. This in‐depth review highlights the recent advances in LIG research, including the mechanism of LIG formation, typical lasers in LIG fabrication, effects of lasing parameters on LIG structures and properties, and applications of LIG in flexible electronics.
The rate of detection of thyroid nodules and carcinomas has increased with the widespread use of ultrasonography (US), which is the mainstay for the detection and risk stratification of thyroid ...nodules as well as for providing guidance for their biopsy and nonsurgical treatment. The Korean Society of Thyroid Radiology (KSThR) published their first recommendations for the US-based diagnosis and management of thyroid nodules in 2011. These recommendations have been used as the standard guidelines for the past several years in Korea. Lately, the application of US has been further emphasized for the personalized management of patients with thyroid nodules. The Task Force on Thyroid Nodules of the KSThR has revised the recommendations for the ultrasound diagnosis and imaging-based management of thyroid nodules. The review and recommendations in this report have been based on a comprehensive analysis of the current literature and the consensus of experts.
The thermal capacity of buildings enables heating, ventilating, and air-conditioning (HVAC) systems to be exploited as demand response (DR) resources. Optimal DR of HVAC units is challenging, ...particularly for multi-zone buildings, because this requires detailed physics-based models of zonal temperature variations for HVAC system operation and building thermal conditions. Using supervised learning (SL), this paper proposes a new strategy for optimal DR of an HVAC system in a multi-zone building. Artificial neural networks (ANNs) are trained with data obtained under normal building operating conditions. The ANNs are replicated using piecewise linear equations, which are explicitly integrated into an optimization problem for price-based DR. The problem is solved for various electricity prices and building thermal conditions. The solutions are used to train a deep neural network (DNN) to directly determine the optimal DR schedule, which is termed as SL-aided meta-prediction (SLAMP) here; the DNN can work as a price-and-optimal-demand curve. Case studies are performed using three different methods: explicit ANN replication, SLAMP, and physics-based modeling. The case study results verify that the proposed SL-based strategy is effective in terms of both practical applicability and computational time, while also ensuring the thermal comfort of occupants and cost-effective operation of the HVAC system.
The development of green flexible micro‐supercapacitors (MSCs) is one of the biggest challenges in future wearable electronics. Flexible MSCs are mainly produced from non‐biodegradable synthetic ...polymers, resulting in massive electronic waste. Moreover, complex multi‐step fabrication increases their production cost. Here, the direct fabrication of highly conductive, intrinsically flexible, and green microelectrodes from naturally fallen leaves in ambient air using femtosecond laser pulses without any additional materials is reported. Hierarchically porous graphene is patterned on different types of leaves via a facile, mask‐less, scalable, and one‐step laser writing. Leaves consist of biominerals, which decompose into inorganic crystals that serve as nucleation sites for the growth of 3D mesoporous few‐layer graphene. The femtosecond laser‐induced graphene (FsLIG) microelectrodes formed on leaves have lower sheet resistance (23.3 Ω sq−1) than their synthetic polymer counterparts and exhibit an outstanding areal capacitance (34.68 mF cm−2 at 5 mV s−1) and capacitance retention (≈99% after 50 000 charge/discharge cycles). The FsLIG MSCs on a single leaf could easily power a light‐emitting diode or a table clock and could be applied in wearable electronics, smart houses, and Internet of Things.
Arbitrary graphene microelectrodes are directly patterned on fallen leaves in ambient air by ultraviolet ultrafast laser. The graphene microelectrodes on leaves exhibit superior electrical conductivity compared with their synthetic polymer counterparts. Due to unique hierarchical porous structures, flexible graphene micro‐supercapacitors on leaves show excellent performance that can be utilized for green wearable electronics.
This paper describes an analysis of a variable speed heat pump (VSHP), which responds to direct load control (DLC) signals to provide grid frequency regulation (GFR) ancillary service, while ensuring ...the comfort of building occupants. A data-driven dynamic model of the VSHP is developed through real-time experimental studies with a time horizon ranging from seconds to hours. The model is simple, yet still sufficiently comprehensive to analyze the operational characteristics of the VSHP. The DLC scheme is then experimentally applied to the VSHP to evaluate its demand response (DR) capability. Two control methods are considered for a practical implementation of the DLC-enabled VSHP and a further improvement of the DR capability, respectively. Additionally, a small-signal analysis is carried out using the aggregated dynamic response of a number of DLC-enabled VSHPs to analyze their contribution to GFR in an isolated power grid. For experimental case studies, a laboratory-scale microgrid is then implemented with generator and load emulators. We show that the DLC-enabled VSHP can effectively reduce grid frequency deviations and required reserve capacities of generators.