Abstract With the advancement of Information and Communication Technology (ICT) proliferation sensor technologies, Internet Things (IoT) is now being widely used in smart home for purposes efficient resource management pervasive sensing. In homes, various IoT devices are connected to each other, these connections centered on gateways. The role gateways homes significant, however, its centralized structure presents multiple security vulnerabilities such as integrity, certification,...

The context-dependent nature of online aggression makes annotating large collections data extremely difficult. Previously studied datasets in abusive language detection have been insufficient size to efficiently train deep learning models. Recently, Hate and Abusive Speech on Twitter, a dataset much greater reliability, has released. However, this not comprehensively its potential. In paper, we conduct the first comparative study various models discuss possibility using additional features...

The broadly configured smart city network requires a variety of security considerations for heterogeneous device environment. Because devices facilitates an attacker’s intrusion through specific or node, management framework is required to manage each node comprehensively. This paper proposes blockchain-based efficient management, scalable firmware update and resiliences on attacks against network. offers four mechanisms based the performance requirements device: bidirectional mechanism...

Parameter calibration of complex environmental models remains a significant challenge in watershed management, particularly when integrating multiple biogeochemical processes. Reinforcement learning (RL) has emerged as promising approach solving optimization problems with its ability to learn optimal strategies through continuous interaction and feedback. This study presents SWAT-C-RL, novel that combines the Soil Water Assessment Tool-Carbon (SWAT-C) RL for efficient multi-objective...

The increasing complexity of water pollution and its impact on aquatic ecosystems necessitates the accurate prediction pollutant loads for effective river management. Total Organic Carbon (TOC), a key indicator organic levels, is central to assessing ecosystem health informing treatment strategies. However, conventional process-based modeling methods, while capable providing precise quality predictions, require extensive input data significant computational resources, limiting their...

Item categorization (IC) aims to classify product descriptions into leaf nodes in a categorical taxonomy, which is key technology used wide range of applications. Along with the fact that most datasets often has long-tailed distribution, classification performances on tail labels tend be poor due scarce supervision, causing many issues real-life To address IC task’s long-tail issue, K-positive contrastive loss (KCL) proposed image task and can applied when using text-based learning, e.g.,...

This study was performed to examine attitude, perception, and sensory evaluation of Jjigae HMR (Home Meal Replacement) for Americans in the L.A. area. Attitude perception were conducted by 128 consumers. The questions as follows: 1) frequencies attitude toward soup stew, 2) experience frequency Korean food intake, 3) awareness Jjigae. A total 69.5% American had previous with foods. However, intake once every few months (27.4%) or a year (18.9%). 20.2% consumers selected Kimchi-jjigae...

Although the concept of RTLS is very unfamiliar to construction industry, recently it popular in other such as logistics, ship building, mobile telecommunication based on state-of-the-art information technology. Effective resource management using cutting-edge technology makes possible succeed a project with saving time and cost. And effective can be achieved by new technologies RFID, WEB-based internet, DB technology, real-time monitoring etc. This paper suggest characteristics...

For the panoramic video streaming service, this letter proposes a visual perception-based view navigation trick mode (VP-VNTM) that reduces bandwidth requirements by adjusting quality of transmitting views in accordance with velocity without decreasing user's sensitivity. Experiments show proposed VP-VNTM more than 44%.

Understanding the dynamics of counseling conversations is an important task, yet it a challenging NLP problem regardless recent advance Transformer-based pre-trained language models. This paper proposes systematic approach to examine efficacy domain knowledge and large models (LLMs) in better representing between crisis counselor help seeker. We empirically show that state-of-the-art such as GPT fail predict conversation outcome. To provide richer context conversations, we incorporate...

Large Language Models (LLMs) have been widely used as general-purpose AI agents showing comparable performance on many downstream tasks. However, existing work shows that it is challenging for LLMs to integrate structured data (e.g. KG, tables, DBs) into their prompts; need either understand long text or select the most relevant evidence prior inference, and both approaches are not trivial. In this paper, we propose a framework, Learning Reduce, fine-tunes language model generate reduced...

Large Language Models (LLMs) have been achieving competent performance on a wide range of downstream tasks, yet existing work shows that inference structured data is challenging for LLMs. This because LLMs need to either understand long or select the most relevant evidence before inference, and both approaches are not trivial. paper proposes framework, Learning Reduce, fine-tunes language model with On-Policy generate reduced version an input data. When compared state-of-the-art like GPT-4,...

The context-dependent nature of online aggression makes annotating large collections data extremely difficult. Previously studied datasets in abusive language detection have been insufficient size to efficiently train deep learning models. Recently, Hate and Abusive Speech on Twitter, a dataset much greater reliability, has released. However, this not comprehensively its potential. In paper, we conduct the first comparative study various models discuss possibility using additional features...

The purpose of the study aims to develop comprehensive modeling alternatives for intraproduct line pricing effect and compare explanation power among them.The researchers want include highest model Shipment Timing Support System stored apple.Five intra-apple models were developed applied wholesale price data with four graded Fuji apple announced by SAMPC in Korea.Five divided into three dichotomies: first level proportional response, second third gap fourth fifth non-proportional...