• Title/Summary/Keyword: generative learning

Search Result 284, Processing Time 0.029 seconds

KOMUChat: Korean Online Community Dialogue Dataset for AI Learning (KOMUChat : 인공지능 학습을 위한 온라인 커뮤니티 대화 데이터셋 연구)

  • YongSang Yoo;MinHwa Jung;SeungMin Lee;Min Song
    • Journal of Intelligence and Information Systems
    • /
    • v.29 no.2
    • /
    • pp.219-240
    • /
    • 2023
  • Conversational AI which allows users to interact with satisfaction is a long-standing research topic. To develop conversational AI, it is necessary to build training data that reflects real conversations between people, but current Korean datasets are not in question-answer format or use honorifics, making it difficult for users to feel closeness. In this paper, we propose a conversation dataset (KOMUChat) consisting of 30,767 question-answer sentence pairs collected from online communities. The question-answer pairs were collected from post titles and first comments of love and relationship counsel boards used by men and women. In addition, we removed abuse records through automatic and manual cleansing to build high quality dataset. To verify the validity of KOMUChat, we compared and analyzed the result of generative language model learning KOMUChat and benchmark dataset. The results showed that our dataset outperformed the benchmark dataset in terms of answer appropriateness, user satisfaction, and fulfillment of conversational AI goals. The dataset is the largest open-source single turn text data presented so far and it has the significance of building a more friendly Korean dataset by reflecting the text styles of the online community.

A Study on Webtoon Background Image Generation Using CartoonGAN Algorithm (CartoonGAN 알고리즘을 이용한 웹툰(Webtoon) 배경 이미지 생성에 관한 연구)

  • Saekyu Oh;Juyoung Kang
    • The Journal of Bigdata
    • /
    • v.7 no.1
    • /
    • pp.173-185
    • /
    • 2022
  • Nowadays, Korean webtoons are leading the global digital comic market. Webtoons are being serviced in various languages around the world, and dramas or movies produced with Webtoons' IP (Intellectual Property Rights) have become a big hit, and more and more webtoons are being visualized. However, with the success of these webtoons, the working environment of webtoon creators is emerging as an important issue. According to the 2021 Cartoon User Survey, webtoon creators spend 10.5 hours a day on creative activities on average. Creators have to draw large amount of pictures every week, and competition among webtoons is getting fiercer, and the amount of paintings that creators have to draw per episode is increasing. Therefore, this study proposes to generate webtoon background images using deep learning algorithms and use them for webtoon production. The main character in webtoon is an area that needs much of the originality of the creator, but the background picture is relatively repetitive and does not require originality, so it can be useful for webtoon production if it can create a background picture similar to the creator's drawing style. Background generation uses CycleGAN, which shows good performance in image-to-image translation, and CartoonGAN, which is specialized in the Cartoon style image generation. This deep learning-based image generation is expected to shorten the working hours of creators in an excessive work environment and contribute to the convergence of webtoons and technologies.

Comparative analysis of the digital circuit designing ability of ChatGPT (ChatGPT을 활용한 디지털회로 설계 능력에 대한 비교 분석)

  • Kihun Nam
    • The Journal of the Convergence on Culture Technology
    • /
    • v.9 no.6
    • /
    • pp.967-971
    • /
    • 2023
  • Recently, a variety of AI-based platform services are available, and one of them is ChatGPT that processes a large quantity of data in the natural language and generates an answer after self-learning. ChatGPT can perform various tasks including software programming in the IT sector. Particularly, it may help generate a simple program and correct errors using C Language, which is a major programming language. Accordingly, it is expected that ChatGPT is capable of effectively using Verilog HDL, which is a hardware language created in C Language. Verilog HDL synthesis, however, is to generate imperative sentences in a logical circuit form and thus it needs to be verified whether the products are executed properly. In this paper, we aim to select small-scale logical circuits for ease of experimentation and to verify the results of circuits generated by ChatGPT and human-designed circuits. As to experimental environments, Xilinx ISE 14.7 was used for module modeling, and the xc3s1000 FPGA chip was used for module embodiment. Comparative analysis was performed on the use area and processing time of FPGA to compare the performance of ChatGPT products and Verilog HDL products.

Relationship between Science Education Researchers' Views on Science Educational Theories for Pre-service Science Teachers and the Examination for Appointing Secondary School Science Teachers (예비과학교사에게 필요한 과학교육학 이론에 대한 과학교육 연구자들의 의견과 중등과학교사임용시험의 연관성)

  • Lee, Bongwoo;Shim, Kew-Cheol;Shin, Myeong-Kyeong;Kim, Jonghee;Choi, Jaehyeok;Park, Eunmi;Yoon, Jihyun;Kwon, Yongju;Kim, Yong-Jin
    • Journal of The Korean Association For Science Education
    • /
    • v.33 no.4
    • /
    • pp.826-839
    • /
    • 2013
  • The purpose of this study is to examine science education researchers' views on what and how much science educational theories would be needed for pre-service science teachers, and to investigate the relationship between their views and the Examination for Appointing Secondary School Science Teachers(EASST). For this study, the views of science education professors on science education theories have been analyzed in terms of their priorities for contributing to the improvement of science teacher competency and literacy. Their views have been compared with proportions of questions related to science education theories of the EASST in terms of what kinds of science education theories have been used for solving each item. As results of this study show, they have perceived that more essential things are needed for the improvement of science teacher competency and literacy including science inquiry process, methods of experimental equipments and tools, laboratory safety, misconception of students, discussion, writing, evaluation of scientific knowledges, and evaluation of scientific inquiry ability other than science philosophy, changes of science curricula, science curricula of foreign countries, Bruner's instructional theory, Karplus's Learning Cycle model, generative learning model, discovery learning model, and Klopfer's taxonomy of educational objectives. There is a higher proportion of questions related to science curriculum and Ausubel's learning theory in the EASST. They are hardly correlated with science education professors' selections of science educational theories for EASST questions. This study advocates the needs of exploring a new method of narrowing down the gap between science educators' opinions and questions of ESSAT in terms of science educaiton theories.

A study on age distortion reduction in facial expression image generation using StyleGAN Encoder (StyleGAN Encoder를 활용한 표정 이미지 생성에서의 연령 왜곡 감소에 대한 연구)

  • Hee-Yeol Lee;Seung-Ho Lee
    • Journal of IKEEE
    • /
    • v.27 no.4
    • /
    • pp.464-471
    • /
    • 2023
  • In this paper, we propose a method to reduce age distortion in facial expression image generation using StyleGAN Encoder. The facial expression image generation process first creates a face image using StyleGAN Encoder, and changes the expression by applying the learned boundary to the latent vector using SVM. However, when learning the boundary of a smiling expression, age distortion occurs due to changes in facial expression. The smile boundary created in SVM learning for smiling expressions includes wrinkles caused by changes in facial expressions as learning elements, and it is determined that age characteristics were also learned. To solve this problem, the proposed method calculates the correlation coefficient between the smile boundary and the age boundary and uses this to introduce a method of adjusting the age boundary at the smile boundary in proportion to the correlation coefficient. To confirm the effectiveness of the proposed method, the results of an experiment using the FFHQ dataset, a publicly available standard face dataset, and measuring the FID score are as follows. In the smile image, compared to the existing method, the FID score of the smile image generated by the ground truth and the proposed method was improved by about 0.46. In addition, compared to the existing method in the smile image, the FID score of the image generated by StyleGAN Encoder and the smile image generated by the proposed method improved by about 1.031. In non-smile images, compared to the existing method, the FID score of the non-smile image generated by the ground truth and the method proposed in this paper was improved by about 2.25. In addition, compared to the existing method in non-smile images, it was confirmed that the FID score of the image generated by StyleGAN Encoder and the non-smile image generated by the proposed method improved by about 1.908. Meanwhile, as a result of estimating the age of each generated facial expression image and measuring the estimated age and MSE of the image generated with StyleGAN Encoder, compared to the existing method, the proposed method has an average age of about 1.5 in smile images and about 1.63 in non-smile images. Performance was improved, proving the effectiveness of the proposed method.

True Orthoimage Generation from LiDAR Intensity Using Deep Learning (딥러닝에 의한 라이다 반사강도로부터 엄밀정사영상 생성)

  • Shin, Young Ha;Hyung, Sung Woong;Lee, Dong-Cheon
    • Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
    • /
    • v.38 no.4
    • /
    • pp.363-373
    • /
    • 2020
  • During last decades numerous studies generating orthoimage have been carried out. Traditional methods require exterior orientation parameters of aerial images and precise 3D object modeling data and DTM (Digital Terrain Model) to detect and recover occlusion areas. Furthermore, it is challenging task to automate the complicated process. In this paper, we proposed a new concept of true orthoimage generation using DL (Deep Learning). DL is rapidly used in wide range of fields. In particular, GAN (Generative Adversarial Network) is one of the DL models for various tasks in imaging processing and computer vision. The generator tries to produce results similar to the real images, while discriminator judges fake and real images until the results are satisfied. Such mutually adversarial mechanism improves quality of the results. Experiments were performed using GAN-based Pix2Pix model by utilizing IR (Infrared) orthoimages, intensity from LiDAR data provided by the German Society for Photogrammetry, Remote Sensing and Geoinformation (DGPF) through the ISPRS (International Society for Photogrammetry and Remote Sensing). Two approaches were implemented: (1) One-step training with intensity data and high resolution orthoimages, (2) Recursive training with intensity data and color-coded low resolution intensity images for progressive enhancement of the results. Two methods provided similar quality based on FID (Fréchet Inception Distance) measures. However, if quality of the input data is close to the target image, better results could be obtained by increasing epoch. This paper is an early experimental study for feasibility of DL-based true orthoimage generation and further improvement would be necessary.

A Case Study of "Engineering Design" Education with Emphasize on Hands-on Experience (기계공학과에서 제시하는 Hands-on Experience 중심의 "엔지니어링 디자인" 교과목의 강의사례)

  • Kim, Hong-Chan;Kim, Ji-Hoon;Kim, Kwan-Ju;Kim, Jung-Soo
    • Journal of Engineering Education Research
    • /
    • v.10 no.2
    • /
    • pp.44-61
    • /
    • 2007
  • The present investigation is concerned chiefly with new curriculum development at the Department of Mechanical System & Design Engineering at Hongik University with the aim of enhancing creativity, team working and communication capability which modern engineering education is emphasizing on. 'Mechanical System & Design Engineering' department equipped with new curriculum emphasizing engineering design is new name for mechanical engineering department in Hongik University. To meet radically changing environment and demands of industries toward engineering education, the department has shifted its focus from analog-based and machine-centered hard approach to digital-based and human-centered soft approach. Three new programs of Introduction to Mechanical System & Design Engineering, Creative Engineering Design and Product Design emphasize hands-on experiences through project-based team working. Sketch model and prototype making process is strongly emphasized and cardboard, poly styrene foam and foam core plate are provided as working material instead of traditional hard engineering material such as metals material because these three programs focus more on creative idea generation and dynamic communication among team members rather than the end results. With generative, visual and concrete experiences that can compensate existing engineering classes with traditional focus on analytic, mathematical and reasoning, hands-on experiences can play a significant role for engineering students to develop creative thinking and engineering sense needed to face ill-defined real-world design problems they are expected to encounter upon graduation.

An Analysis on the Responses and the Behavioral Characteristics between Mathematically Promising Students and Normal Students in Solving Open-ended Mathematical Problems (수학 영재교육 대상 학생과 일반 학생의 개방형 문제해결 전략 및 행동 특성 분석)

  • Kim, Eun-Hye;Park, Man-Goo
    • Journal of Elementary Mathematics Education in Korea
    • /
    • v.15 no.1
    • /
    • pp.19-38
    • /
    • 2011
  • The purpose of this study was to analyze the responses and the behavioral characteristics between mathematically promising students and normal students in solving open-ended problems. For this study, 55 mathematically promising students were selected from the Science Education Institute for the Gifted at Seoul National University of Education as well as 100 normal students from three 6th grade classes of a regular elementary school. The students were given 50 minutes to complete a written test consisting of five open-ended problems. A post-test interview was also conducted and added to the results of the written test. The conclusions of this study were summarized as follows: First, analysis and grouping problems are the most suitable in an open-ended problem study to stimulate the creativity of mathematically promising students. Second, open-ended problems are helpful for mathematically promising students' generative learning. The mathematically promising students had a tendency to find a variety of creative methods when solving open-ended problems. Third, mathematically promising students need to improve their ability to make-up new conditions and change the conditions to solve the problems. Fourth, various topics and subjects can be integrated into the classes for mathematically promising students. Fifth, the quality of students' former education and its effect on their ability to solve open-ended problems must be taken into consideration. Finally, a creative thinking class can be introduce to the general class. A number of normal students had creativity score similar to those of the mathematically promising students, suggesting that the introduction of a more challenging mathematics curriculum similar to that of the mathematically promising students into the general curriculum may be needed and possible.

  • PDF

Convergence of Artificial Intelligence Techniques and Domain Specific Knowledge for Generating Super-Resolution Meteorological Data (기상 자료 초해상화를 위한 인공지능 기술과 기상 전문 지식의 융합)

  • Ha, Ji-Hun;Park, Kun-Woo;Im, Hyo-Hyuk;Cho, Dong-Hee;Kim, Yong-Hyuk
    • Journal of the Korea Convergence Society
    • /
    • v.12 no.10
    • /
    • pp.63-70
    • /
    • 2021
  • Generating a super-resolution meteological data by using a high-resolution deep neural network can provide precise research and useful real-life services. We propose a new technique of generating improved training data for super-resolution deep neural networks. To generate high-resolution meteorological data with domain specific knowledge, Lambert conformal conic projection and objective analysis were applied based on observation data and ERA5 reanalysis field data of specialized institutions. As a result, temperature and humidity analysis data based on domain specific knowledge showed improved RMSE by up to 42% and 46%, respectively. Next, a super-resolution generative adversarial network (SRGAN) which is one of the aritifial intelligence techniques was used to automate the manual data generation technique using damain specific techniques as described above. Experiments were conducted to generate high-resolution data with 1 km resolution from global model data with 10 km resolution. Finally, the results generated with SRGAN have a higher resoltuion than the global model input data, and showed a similar analysis pattern to the manually generated high-resolution analysis data, but also showed a smooth boundary.

A review on the method of coined words by Korean and Chinese characters (한·중 인물지칭 신어 조어방식에 관한 고찰 - 2017년과 2018년을 중심으로 -)

  • Wang, Yan
    • Journal of Convergence for Information Technology
    • /
    • v.12 no.3
    • /
    • pp.178-185
    • /
    • 2022
  • This study compared and analyzed the characteristics of new words by classifying 197 newly coined Korean and Chinese characters in 2017 and 2018 into single, compound, derivative, abbreviated, and hybrid words according to the coined method. In the case of a single language, Korean is all words borrowed from Chinese and English. However, no monolingual language appeared in Chinese. In the case of compound words, the format of the Chinese synthesis method was much more diverse and the generative power was stronger than that of Korea. In the case of derivatives, there are not many prefixes in both countries, and Korean suffixes have the strongest productivity of Chinese suffixes and weak productivity of foreign and native suffixes. Korean foreign language suffixes were characterized by relatively more appearance than Chinese. In the case of abbreviations, it can be seen that the productivity of dark syllables is stronger for Korean abbreviations, and the productivity of empty syllables is stronger for Chinese abbreviations. In the case of mixed languages, the hybrid form of Korean was much more diverse than that of Chinese. Through this study, it will be possible to help Chinese Korean learners understand the process of forming a new language, and to develop their ability to guess the meaning of Korean words while learning a new language.