• International conference on construction engineering and project management
• /
• 2024.07a
• /
• pp.1326-1327
• /
• 2024

The construction industry is renowned for its dynamic and intricate characteristics, which demand proficient leadership skills for successful project management. However, the existing training platforms within this sector often overlook the significance of soft skills in leadership development. These platforms primarily focus on safety, work processes, and technical modules, leaving a noticeable gap in preparing future leaders, especially students in the construction domain, for the complex challenges they will encounter in their professional careers. It is crucial to recognize that effective leadership in construction projects requires not only technical expertise but also the ability to communicate effectively, collaborate with diverse stakeholders, and navigate complex relationships. These soft skills are critical for managing teams, resolving conflicts, and driving successful project outcomes. In addition, the construction sector has been slow in adopting and harnessing the potential of advanced emerging technologies such as virtual reality, artificial intelligence, to enhance the soft skills of future leaders. Therefore, there is a need for a platform where students can practice complex situations and conversations in a safe and repeatable training environment. To address these challenges, this study proposes a pioneering approach by integrating conversational AI techniques using large language models (LLMs) within virtual worlds. Although LLMs like ChatGPT possess extensive knowledge across various domains, their responses may lack relevance in specific contexts. Prompt engineering techniques are utilized to ensure more accurate and effective responses, tailored to the specific requirements of the targeted users. This involves designing and refining the input prompts given to the language model to guide its response generation. By carefully crafting the prompts and providing context-specific instructions, the model can generate responses that are more relevant and aligned with the desired outcomes of the training program. The proposed system offers interactive engagement to students by simulating diverse construction site roles through conversational AI based agents. Students can face realistic challenges that test and enhance their soft skills in a practical context. They can engage in conversations with AI-based avatars representing different construction site roles, such as machine operators, laborers, and site managers. These avatars are equipped with AI capabilities to respond dynamically to user interactions, allowing students to practice their communication and negotiation skills in realistic scenarios. Additionally, the introduction of AI instructors can provide guidance, feedback, and coaching tailored to the individual needs of each student, enhancing the effectiveness of the training program. The AI instructors can provide immediate feedback and guidance, helping students improve their decision-making and problem-solving abilities. The proposed immersive learning environment is expected to significantly enhance leadership competencies of students, such as communication, decision-making and conflict resolution in the practical context. This study highlights the benefits of utilizing conversational AI in educational settings to prepare construction students for real-world leadership roles. By providing hands-on, practical experience in dealing with site-specific challenges, students can develop the necessary skills and confidence to excel in their future roles.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.18 no.3
• /
• pp.89-100
• /
• 2023

Recently, on-device artificial intelligence (AI) solutions using mobile devices and embedded edge devices have emerged in various fields, such as computer vision, to address network traffic burdens, low-energy operations, and security problems. Although vision transformer deep learning models have outperformed conventional convolutional neural network (CNN) models in computer vision, they require more computations and parameters than CNN models. Thus, they are not directly applicable to embedded edge devices with limited hardware resources. Many researchers have proposed various model compression methods or lightweight architectures for vision transformers; however, there are only a few studies evaluating the effects of model compression techniques of vision transformers on performance. Regarding this problem, this paper presents a performance evaluation of vision transformers on embedded platforms. We investigated the behaviors of three vision transformers: DeiT, LeViT, and MobileViT. Each model performance was evaluated by accuracy and inference time on edge devices using the ImageNet dataset. We assessed the effects of the quantization method applied to the models on latency enhancement and accuracy degradation by profiling the proportion of response time occupied by major operations. In addition, we evaluated the performance of each model on GPU and EdgeTPU-based edge devices. In our experimental results, LeViT showed the best performance in CPU-based edge devices, and DeiT-small showed the highest performance improvement in GPU-based edge devices. In addition, only MobileViT models showed performance improvement on EdgeTPU. Summarizing the analysis results through profiling, the degree of performance improvement of each vision transformer model was highly dependent on the proportion of parts that could be optimized in the target edge device. In summary, to apply vision transformers to on-device AI solutions, either proper operation composition and optimizations specific to target edge devices must be considered.

• The Journal of Bigdata
• /
• v.8 no.2
• /
• pp.83-95
• /
• 2023

Big data analytics and AI play a critical role in the development of smart plants. This study presents a big data platform for plant data and an 'AutoML platform' for AI-based plant O&M(Operation and Maintenance). The big data platform collects, processes and stores large volumes of data generated in plants using Hadoop, Spark, and Kafka. The AutoML platform is a machine learning automation system aimed at constructing predictive models for equipment prognostics and process optimization in plants. The developed platforms configures a data pipeline considering compatibility with existing plant OISs(Operation Information Systems) and employs a web-based GUI to enhance both accessibility and convenience for users. Also, it has functions to load user-customizable modules into data processing and learning algorithms, which increases process flexibility. This paper demonstrates the operation of the platforms for a specific process of an oil company in Korea and presents an example of an effective data utilization platform for smart plants.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.3
• /
• pp.217-223
• /
• 2022

In this paper, the needs and environment of the 'AI Convergence Education Method' were analyzed for elementary and secondary teachers, and based on this, teaching efficacy of informatics education was verified. For the research, elementary and secondary teachers who take the subject were selected, and based on the results of analyzing the general characteristics, pre-knowledge level, and needs of the subject, curriculum for 15 weeks was developed. As a result of verifying the teaching efficacy effectiveness of the developed 'AI Convergence Education Method' subject for 15 weeks, the effectiveness of the information education teaching efficacy was verified. Among the factors, there were statistically significant differences in information teaching values and information teaching strategies. In the future, it is necessary to conduct follow-up research to secure teachers' professionalism, such as linking with schools and convergence with other subjects. Various teaching and learning materials and teaching and learning methods such as educational contents and materials, reference literature, and artificial intelligence education platforms need to be prepared.

• 한국정보교육학회:학술대회논문집
• /
• 2021.08a
• /
• pp.213-221
• /
• 2021

With the advent of the 4th industrial revolution, the need for artificial intelligence education has increased. The online learning environment caused by COVID-19 made it possible to use variety of artificial intelligence platforms. In this study, an aritificial intelligence class plan was developed and proposed to achieve the goal of artificial intelligence education using an AI platform. The AI platform used is AI for Oceans, With the theme of creating a program for the environment, designed a 6-hour project class using Novel Engineering-based on STEAM model. Students experience AI for Oceans enough time and learn supervised learning by experience. Based on understanding of supervised learning, students design their own programs for the environment using Entry's AI blocks. In this study, for AI convergence education, this lesson was developed and presented with the goal of acquiring the creative problem solving ability and integrated thinking ability by using the principles of artificial intelligence to solve problems.

• International journal of advanced smart convergence
• /
• v.10 no.4
• /
• pp.72-83
• /
• 2021

The use of the predictive analytics (PA) powered by the artificial intelligence (AI) is more important in the movie sector during the COVID-19 pandemic, because Hollywood witnessed the impact of the 'Netflix Effect' and began to invest in data and AI. Our purpose is to discover a few cases of the AI centered PA in the movie industry value chain based on five objectives of PA: Compete, grow, enforce, improve, and satisfy. Even if movie companies' interest is to predict future success for competing with over-the-tops (OTTs) at a first glance, it is observed, once they start to use the PA with the AI, they try to utilize the enhanced PA platforms for remaining four objectives. As a result, ScriptBook, Vault, Pilot, Cinelytic and Merlin Video (Merlin) are use cases for the objective 'compete.' Movio of Vista Group International and Datorama of Salesforce are use cases for the objective 'grow.' Industrial Light & Magic (ILM) and Geena Davis Institute on Gender in Media (GDI) with Disney are use cases for the objective 'enforce.' Watson, Benjamin, and Greenlight Essential are use cases for the objective 'improve.' Disney Research (DR) with Simon Fraser University and California Institute of Technology is the use case for the objective 'satisfy.'

• Journal of the Korea Convergence Society
• /
• v.13 no.3
• /
• pp.11-22
• /
• 2022

Recently, research and development using various Artificial Intelligence (AI) technologies are being conducted in the field of education. Among the AI in Education (AIEd), conversational agents are not limited by time and space, and can learn more effectively by combining them with various AI technologies such as voice recognition and translation. This paper conducted a trend analysis on platforms that have a large number of users and used conversational agents for English learning among commercialized application. Currently commercialized educational platforms using conversational agent through trend analysis has several limitations and problems. To analyze specific problems and limitations, a comparative experiment was conducted with the latest pre-trained large-capacity dialogue model. Sensibleness and Specificity Average (SSA) human evaluation was conducted to evaluate conversational human-likeness. Based on the experiment, this paper propose the need for trained with large-capacity parameters dialogue models, educational data, and information retrieval functions for effective English conversation learning.

• Journal of Korea Multimedia Society
• /
• v.23 no.2
• /
• pp.241-246
• /
• 2020

Hadoop is a framework to process large data sets in a distributed way across clusters of nodes. It has been a popular platform to process big data, but in recent years, other platforms became competitive ones depending on the characteristics of the application. Spark is one of distributed platforms to enable real-time data processing and improve overall processing performance over Hadoop by introducing in-memory processing instead of disk I/O. Whereas Hadoop is designed to work on Java and data analysis is processed using Java API, Spark provides a variety of APIs with Scala, Python, Java and R. In this paper, the goal is to find out whether the APIs of different programming languages af ect the performances in Spark. We chose two popular APIs: Python and Scala. Python is easy to learn and is used in AI domain in a wide range. Scala is a programming language with advantages of parallelism. Our experiment shows much faster processing with Scala API than Python API. For the performance issues on AI-based analysis, further study is needed.

• Journal of the Korean Society of Radiology
• /
• v.17 no.6
• /
• pp.939-945
• /
• 2023

This study analyzes the 'performance efficiency' of AI-based reading assistance systems in the field of radiology by measuring their 'time behavior' properties. Due to the increase in medical images and the limited number of radiologists, the adoption of AI-based solutions is escalating, stimulating a multitude of studies in this area. Contrary to the majority of past research which centered on AI's diagnostic precision, this study underlines the significance of time behavior. Using 50 chest X-ray PA images, the system processed images in an average of 15.24 seconds, demonstrating high consistency and reliability, which is on par with leading global AI platforms, suggesting the potential for significant improvements in radiology workflow efficiency. We expect AI technology to play a large role in the field of radiology and help improve overall healthcare quality and efficiency.

• Journal of Internet Computing and Services
• /
• v.25 no.4
• /
• pp.109-119
• /
• 2024

This study aims to contribute to the initial research on on-device AI in an environment where generative AI-based services on mobile and other on-device platforms are increasing. To derive success strategies for generative AI-based chatbot services in a mobile environment, over 200,000 actual user experience review data collected from the Google Play Store were analyzed using the LDA topic modeling technique. Interpreting the derived topics based on the Information System Success Model (ISSM), the topics such as tutoring, limitation of response, and hallucination and outdated informaiton were linked to information quality; multimodal service, quality of response, and issues of device interoperability were linked to system quality; inter-device compatibility, utility of the service, quality of premium services, and challenges in account were linked to service quality; and finally, creative collaboration was linked to net benefits. Humanization of generative AI emerged as a new experience factor not explained by the existing model. By explaining specific positive and negative experience dimensions from the user's perspective based on theory, this study suggests directions for future related research and provides strategic insights for companies to improve and supplement their services for successful business operations.