• International Journal of Internet, Broadcasting and Communication
• /
• 제12권4호
• /
• pp.232-237
• /
• 2020

Given the recent pace of development and expansion of Artificial Intelligence (AI) technology, the influence and ripple effects of AI technology on the whole of our lives will be very large and spread rapidly. The National Artificial Intelligence R&D Strategy, published in 2019, emphasizes the importance of artificial intelligence education for K-12 students. It also mentions STEM education, AI convergence curriculum, and budget for supporting the development of teaching materials and tools. However, it is necessary to create a new type of curriculum at a time when artificial intelligence curriculum has never existed before. With many attempts and discussions going very fast in all countries on almost the same starting line. Also, there is no suitable professor for K-12 students, and it is difficult to make K-12 students understand the concept of AI. In particular, it is difficult to teach elementary school students through professional programming in AI education. It is also difficult to learn tools that can teach AI concepts. In this paper, we propose an educational model for elementary school students to improve their understanding of AI through play or experience. This an experiential education model that combineds exploratory learning and discovery learning using multi-intelligence and the PLAY teaching-learning model to undertand the importance of data training or data required for AI education. This educational model is designed to learn how a computer that knows only binary numbers through UA recognizes images. Through code.org, students were trained to learn AI robots and configured to understand data bias like play. In addition, by learning images directly on a computer through TeachableMachine, a tool capable of supervised learning, to understand the concept of dataset, learning process, and accuracy, and proposed the process of AI inference.

• 한국컴퓨터정보학회논문지
• /
• 제28권1호
• /
• pp.27-38
• /
• 2023

본 논문에서는 시간의 흐름에 따라 새로운 데이터를 지속적으로 학습하고 성장할 수 있는 연속 학습 기반 대화형 AI 에이전트를 제안한다. 연속학습 기반 대화형 AI 에이전트는 태스크 관리자 (Task Manager), 사용자 속성 추출(User Attribute Extraction), 자동 확장 지식 그래프(Auto-growing Knowledge Graph), 크게 3가지 요소로 구성된다. 태스크 관리자는 사용자와의 대화에서 새로운 데이터를 발견하면 이전에 학습한 지식을 통해 새로운 태스크를 생성한다. 사용자 특성 추출 모델은 새로운 태스크에서 사용자의 특성을 추출하고, 자동 확장 지식 그래프는 새로운 외부 지식을 지속적으로 학습할 수 있도록 한다. 한정된 데이터셋을 기반으로 학습된 기존 대화형 AI 에이전트와 달리, 본 논문에서 제안하는 방법은 지속적인 사용자의 특성과 지식 학습을 기반으로 대화를 가능하게 한다. 연속학습 기술이 적용된 대화형 AI 에이전트는 사용자와의 대화가 축적될수록 개인 맞춤형 대응이 가능하며, 새로운 지식에도 대응이 가능하다. 본 논문에서는 시간에 따른 대화 생성 모델의 성능 변화 실험을 통해 제안하는 방법의 가능성을 검증한다.

• Korean Journal of Radiology
• /
• 제22권12호
• /
• pp.2017-2025
• /
• 2021

Objective: To evaluate the accuracy and clinical efficacy of a hybrid Greulich-Pyle (GP) and modified Tanner-Whitehouse (TW) artificial intelligence (AI) model for bone age assessment. Materials and Methods: A deep learning-based model was trained on an open dataset of multiple ethnicities. A total of 102 hand radiographs (51 male and 51 female; mean age ± standard deviation = 10.95 ± 2.37 years) from a single institution were selected for external validation. Three human experts performed bone age assessments based on the GP atlas to develop a reference standard. Two study radiologists performed bone age assessments with and without AI model assistance in two separate sessions, for which the reading time was recorded. The performance of the AI software was assessed by comparing the mean absolute difference between the AI-calculated bone age and the reference standard. The reading time was compared between reading with and without AI using a paired t test. Furthermore, the reliability between the two study radiologists' bone age assessments was assessed using intraclass correlation coefficients (ICCs), and the results were compared between reading with and without AI. Results: The bone ages assessed by the experts and the AI model were not significantly different (11.39 ± 2.74 years and 11.35 ± 2.76 years, respectively, p = 0.31). The mean absolute difference was 0.39 years (95% confidence interval, 0.33-0.45 years) between the automated AI assessment and the reference standard. The mean reading time of the two study radiologists was reduced from 54.29 to 35.37 seconds with AI model assistance (p < 0.001). The ICC of the two study radiologists slightly increased with AI model assistance (from 0.945 to 0.990). Conclusion: The proposed AI model was accurate for assessing bone age. Furthermore, this model appeared to enhance the clinical efficacy by reducing the reading time and improving the inter-observer reliability.

• Nuclear Engineering and Technology
• /
• 제54권10호
• /
• pp.3943-3948
• /
• 2022

Single-photon emission computed tomography is one of the reliable pin-by-pin verification techniques for spent-fuel assemblies. One of the challenges with this technique is to increase the total fuel assembly verification speed while maintaining high verification accuracy. The aim of the present study, therefore, was to develop an artificial intelligence (AI) algorithm-based tomographic image analysis technique for partial-defect verification of fuel assemblies. With the Monte Carlo (MC) simulation technique, a tomographic image dataset consisting of 511 fuel-rod patterns of a 3 × 3 fuel assembly was generated, and with these images, the VGG16, GoogLeNet, and ResNet models were trained. According to an evaluation of these models for different training dataset sizes, the ResNet model showed 100% pattern estimation accuracy. And, based on the different tomographic image qualities, all of the models showed almost 100% pattern estimation accuracy, even for low-quality images with unrecognizable fuel patterns. This study verified that an AI model can be effectively employed for accurate and fast partial-defect verification of fuel assemblies.

• 말소리와 음성과학
• /
• 제15권3호
• /
• pp.83-88
• /
• 2023

This study investigates the fine-tuning of large-scale Automatic Speech Recognition (ASR) models, specifically OpenAI's Whisper model, for domain-specific applications using the KsponSpeech dataset. The primary research questions address the effectiveness of targeted lexical item emphasis during fine-tuning, its impact on domain-specific performance, and whether the fine-tuned model can maintain generalization capabilities across different languages and environments. Experiments were conducted using two fine-tuning datasets: Set A, a small subset emphasizing specific lexical items, and Set B, consisting of the entire KsponSpeech dataset. Results showed that fine-tuning with targeted lexical items increased recognition accuracy and improved domain-specific performance, with generalization capabilities maintained when fine-tuned with a smaller dataset. For noisier environments, a trade-off between specificity and generalization capabilities was observed. This study highlights the potential of fine-tuning using minimal domain-specific data to achieve satisfactory results, emphasizing the importance of balancing specialization and generalization for ASR models. Future research could explore different fine-tuning strategies and novel technologies such as prompting to further enhance large-scale ASR models' domain-specific performance.

• 한국군사과학기술학회지
• /
• 제27권2호
• /
• pp.167-176
• /
• 2024

Lip-reading is the task of inferring the speaker's utterance from silent video based on learning of lip movements. It is very challenging due to the inherent ambiguities present in the lip movement such as different characters that produce the same lip appearances. Recent advances in deep learning models such as Transformer and Temporal Convolutional Network have led to improve the performance of lip-reading. However, most previous works deal with English lip-reading which has limitations in directly applying to Korean lip-reading, and moreover, there is no a large scale Korean lip-reading dataset. In this paper, we introduce the first large-scale Korean lip-reading dataset with more than 120 k utterances collected from TV broadcasts containing news, documentary and drama. We also present a preprocessing method which uniformly extracts a facial region of interest and propose a transformer-based model based on grapheme unit for sentence-level Korean lip-reading. We demonstrate that our dataset and model are appropriate for Korean lip-reading through statistics of the dataset and experimental results.

• 식품보건융합연구
• /
• 제9권4호
• /
• pp.19-23
• /
• 2023

The integration of ChatGPT, an AI-powered language model, is causing a profound transformation within the food industry, impacting various domains. It offers novel capabilities in recipe creation, personalized dining, menu development, food safety, customer service, and culinary education. ChatGPT's vast culinary dataset analysis aids chefs in pushing flavor boundaries through innovative ingredient combinations. Its personalization potential caters to dietary preferences and cultural nuances, democratizing culinary knowledge. It functions as a virtual mentor, empowering enthusiasts to experiment creatively. For personalized dining, ChatGPT's language understanding enables customer interaction, dish recommendations based on preferences. In menu development, data-driven insights identify culinary trends, guiding chefs in crafting menus aligned with evolving tastes. It suggests inventive ingredient pairings, fostering innovation and inclusivity. AI-driven data analysis contributes to quality control, ensuring consistent taste and texture. Food writing and marketing benefit from ChatGPT's content generation, adapting to diverse strategies and consumer preferences. AI-powered chatbots revolutionize customer service, improving ordering experiences, and post-purchase engagement. In culinary education, ChatGPT acts as a virtual mentor, guiding learners through techniques and history. In food safety, data analysis prevents contamination and ensures compliance. Overall, ChatGPT reshapes the industry by uniting AI's analytics with culinary expertise, enhancing innovation, inclusivity, and efficiency in gastronomy.

• 한국멀티미디어학회논문지
• /
• 제25권8호
• /
• pp.1049-1058
• /
• 2022

HAR(Human Action Recognition) such as anomaly and object detection has become a trend in research field(s) that focus on utilizing Artificial Intelligence (AI) methods to analyze patterns of human action in crime-ridden area(s), media services, and industrial facilities. Especially, in real-time system(s) using video streaming data, HAR has become a more important AI-based research field in application development and many different research fields using HAR have currently been developed and improved. In this paper, we propose and analyze a deep-learning-based HAR that provides more efficient scheme(s) using an intelligent AI models, such system can be applied to media services using RGB video streaming data usage without feature extraction pre-processing. For the method, we adopt Slowfast based on the Deep Neural Network(DNN) model under an open dataset(HMDB-51 or UCF101) for improvement in prediction accuracy.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2022년도 추계학술발표대회
• /
• pp.869-871
• /
• 2022

현재 사회적 문제로 잘못된 자원 재활용 방법 및 경비 노동자 근로 환경 개선 필요성이 지속해서 대두되고 있으며, 최근 발생한 코로나바이러스로 인하여 배달 음식의 수요가 증가하여 각 가정에서 배출되는 쓰레기의 양이 매우 증가하였다. 이러한 사회적 문제를 효율적으로 대처하기 위하여 본 논문에서는 분리수거가 가능한 사물을 인식하여 AI 모듈로 객체 정보를 전송하고 전송된 정보에 따라 적절한 분리수거를 수행하는 스마트 분리수거 자동화 시스템을 개발하였다. 본 연구에서는 잘못된 객체 정보 전송을 최소화하고, 객체 인식률의 정확도를 높이기 위하여 많은 종류의 Custom dataset을 Yolo_Mark, Scaling Annoter Tool을 이용하여 직접 라벨링 하였으며 K-means Clustering 알고리즘을 적용하여 더욱 정확한 분리수거 자동화 시스템을 구현하였다. 본 연구를 바탕으로 불필요한 자원과 인력 낭비를 줄일 수 있으며, 인간이 아닌 시스템에 의해 통제되므로 더욱 정확한 분리수거가 가능하다.

• 한국지진공학회논문집
• /
• 제25권2호
• /
• pp.71-81
• /
• 2021

This paper presents a real-time, false-pick filter based on deep learning to reduce false alarms of an onsite Earthquake Early Warning (EEW) system. Most onsite EEW systems use P-wave to predict S-wave. Therefore, it is essential to properly distinguish P-waves from noises or other seismic phases to avoid false alarms. To reduce false-picks causing false alarms, this study made the EEWNet Part 1 'False-Pick Filter' model based on Convolutional Neural Network (CNN). Specifically, it modified the Pick_FP (Lomax et al.) to generate input data such as the amplitude, velocity, and displacement of three components from 2 seconds ahead and 2 seconds after the P-wave arrival following one-second time steps. This model extracts log-mel power spectrum features from this input data, then classifies P-waves and others using these features. The dataset consisted of 3,189,583 samples: 81,394 samples from event data (727 events in the Korean Peninsula, 103 teleseismic events, and 1,734 events in Taiwan) and 3,108,189 samples from continuous data (recorded by seismic stations in South Korea for 27 months from 2018 to 2020). This model was trained with 1,826,357 samples through balancing, then tested on continuous data samples of the year 2019, filtering more than 99% of strong false-picks that could trigger false alarms. This model was developed as a module for USGS Earthworm and is written in C language to operate with minimal computing resources.