• Journal of Practical Engineering Education
• /
• v.15 no.2
• /
• pp.365-371
• /
• 2023

In this study, we proposed a plan to educate computer students at industrial specialized high schools about artificial intelligence programming using Raspberry Pi. To create an educational program, we received advice from experts working in schools and industries, analyzed existing research and requirements, designed weekly learning plans, developed teaching materials, and conducted classes. Due to the small number of research subjects, interviews were conducted with students, and the results of the teacher's diary were also presented to derive qualitative research results. The main interview results show that although it is true that interest in the field of artificial intelligence has increased through the class, many responded that the learning content is still difficult. The teacher's diary mainly included information about the latest trends in the industry that informatics and computer teachers should not miss out on. We hope that this study will provide an opportunity to meet the needs of the industry by increasing the proportion of artificial intelligence programming in industrial specialized high schools.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.33 no.6
• /
• pp.265-274
• /
• 2021

Outlier detection research in ocean data has traditionally been performed using statistical and distance-based machine learning algorithms. Recently, AI-based methods have received a lot of attention and so-called supervised learning methods that require classification information for data are mainly used. This supervised learning method requires a lot of time and costs because classification information (label) must be manually designated for all data required for learning. In this study, an autoencoder based on unsupervised learning was applied as an outlier detection to overcome this problem. For the experiment, two experiments were designed: one is univariate learning, in which only SST data was used among the observation data of Deokjeok Island and the other is multivariate learning, in which SST, air temperature, wind direction, wind speed, air pressure, and humidity were used. Period of data is 25 years from 1996 to 2020, and a pre-processing considering the characteristics of ocean data was applied to the data. An outlier detection of actual SST data was tried with a learned univariate and multivariate autoencoder. We tried to detect outliers in real SST data using trained univariate and multivariate autoencoders. To compare model performance, various outlier detection methods were applied to synthetic data with artificially inserted errors. As a result of quantitatively evaluating the performance of these methods, the multivariate/univariate accuracy was about 96%/91%, respectively, indicating that the multivariate autoencoder had better outlier detection performance. Outlier detection using an unsupervised learning-based autoencoder is expected to be used in various ways in that it can reduce subjective classification errors and cost and time required for data labeling.

• Computers and Concrete
• /
• v.21 no.4
• /
• pp.407-417
• /
• 2018

Concrete which is a composite material is one of the most important construction materials. Compressive strength is a commonly used parameter for the assessment of concrete quality. Accurate prediction of concrete compressive strength is an important issue. In this study, we utilized an experimental procedure for the assessment of concrete quality. Firstly, the concrete mix was prepared according to C 20 type concrete, and slump of fresh concrete was about 20 cm. After the placement of fresh concrete to formworks, compaction was achieved using a vibrating screed. After 28 day period, a total of 100 core samples having 75 mm diameter were extracted. On the core samples pulse velocity determination tests and compressive strength tests were performed. Besides, Windsor probe penetration tests and Schmidt hammer tests were also performed. After setting up the data set, twelve artificial intelligence (AI) models compared for predicting the concrete compressive strength. These models can be divided into three categories (i) Functions (i.e., Linear Regression, Simple Linear Regression, Multilayer Perceptron, Support Vector Regression), (ii) Lazy-Learning Algorithms (i.e., IBk Linear NN Search, KStar, Locally Weighted Learning) (iii) Tree-Based Learning Algorithms (i.e., Decision Stump, Model Trees Regression, Random Forest, Random Tree, Reduced Error Pruning Tree). Four evaluation processes, four validation implements (i.e., 10-fold cross validation, 5-fold cross validation, 10% split sample validation & 20% split sample validation) are used to examine the performance of predictive models. This study shows that machine learning regression techniques are promising tools for predicting compressive strength of concrete.

• Journal of Convergence for Information Technology
• /
• v.11 no.3
• /
• pp.1-6
• /
• 2021

Most of the match-3 puzzle games supports automatic play using the MCTS algorithm. However, implementing reinforcement learning agents is not an easy job because it requires both the knowledge of machine learning and the way of complex interactions within the development environment. This study proposes a method in which we can easily design reinforcement learning agents and implement game play agents by applying PPO(Proximal Policy Optimization) algorithms. And we could identify the performance was increased about 44% than the conventional method. The tools we used are the Unity 3D game engine and Unity ML SDK. The experimental result shows that agents became to learn game rules and make better strategic decisions as experiments go on. On average, the puzzle gameplay agents implemented in this study played puzzle games better than normal people. It is expected that the designed agent could be used to speed up the game level design process.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.21 no.1
• /
• pp.105-122
• /
• 2022

According to the statistics about the fatal crashes that have occurred on the expressways for the last 5 years, those who died on the shoulders of the road has been as 3 times high as the others who died on the expressways. It suggests that the crashes on the shoulders of the road should be fatal, and that it would be important to prevent the traffic crashes by cracking down on the vehicles intruding the shoulders of the road. Therefore, this study proposed a method to detect a vehicle that violates the shoulder lane by using the Faster R-CNN. The vehicle was detected based on the Faster R-CNN, and an additional reading module was configured to determine whether there was a shoulder violation. For experiments and evaluations, GTAV, a simulation game that can reproduce situations similar to the real world, was used. 1,800 images of training data and 800 evaluation data were processed and generated, and the performance according to the change of the threshold value was measured in ZFNet and VGG16. As a result, the detection rate of ZFNet was 99.2% based on Threshold 0.8 and VGG16 93.9% based on Threshold 0.7, and the average detection speed for each model was 0.0468 seconds for ZFNet and 0.16 seconds for VGG16, so the detection rate of ZFNet was about 7% higher. The speed was also confirmed to be about 3.4 times faster. These results show that even in a relatively uncomplicated network, it is possible to detect a vehicle that violates the shoulder lane at a high speed without pre-processing the input image. It suggests that this algorithm can be used to detect violations of designated lanes if sufficient training datasets based on actual video data are obtained.

• Korean Journal of Artificial Intelligence
• /
• v.10 no.1
• /
• pp.27-32
• /
• 2022

As the need for eco-friendly transportation increases due to the deepening climate crisis, many local governments in Korea are introducing shared bicycles. Due to anxiety about public transportation after COVID-19, bicycles have firmly established themselves as the axis of daily transportation. The use of shared bicycles is spread, and the demand for bicycles is increasing by rental offices, but there are operational and management difficulties because the demand is managed under a limited budget. And unfortunately, user behavior results in a spatial imbalance of the bike inventory over time. So, in order to easily operate the maintenance of shared bicycles in Seoul, bicycles should be prepared in large quantities at a time of high demand and withdrawn at a low time. Therefore, in this study, by using machine learning, the linear regression algorithm and MS Azure ML are used to predict and analyze when demand is high. As a result of the analysis, the demand for bicycles in 2018 is on the rise compared to 2017, and the demand is lower in winter than in spring, summer, and fall. It can be judged that this linear regression-based prediction can reduce maintenance and management costs in a shared society and increase user convenience. In a further study, we will focus on shared bike routes by using GPS tracking systems. Through the data found, the route used by most people will be analyzed to derive the optimal route when installing a bicycle-only road.

• Journal of Information Technology Services
• /
• v.23 no.3
• /
• pp.91-101
• /
• 2024

Based on the analysis of 174 Young Startups Project Papers from university students majoring in entrepreneurship, this study aimed to identify the thoughts and characteristics of prospective entrepreneurs and derive their main agendas. Detailed information was collected on their preparation and ambitions for entrepreneurship, market research in their fields of interest, analysis of potential target users, requirements, and team management. We refined, categorized, and interpreted 84,818 words across 396 digital pages on LMS, and visualized using word clouds program. Additionally, interviews about peer learning were conducted by Zoom. The analysis revealed unique differences in students' perspectives on entrepreneurship based on their major, academic year, and gender. Also we could find that they actively utilize AI tech with a strong spirit of challenge in the startups. Interestingly, they emphasized horizontal communication and collaboration capabilities to improve productivity. This suggests the need to consider diversity in the design of entrepreneurship incubating content and programs for generation Z technically and academically.

• Phonetics and Speech Sciences
• /
• v.15 no.3
• /
• pp.75-82
• /
• 2023

Automatic speech recognition (ASR) has been revolutionized with deep learning-based approaches, among which self-supervised learning methods have proven to be particularly effective. In this study, we aim to enhance the performance of OpenAI's Whisper model, a multilingual ASR system on the Korean language. Whisper was pretrained on a large corpus (around 680,000 hours) of web speech data and has demonstrated strong recognition performance for major languages. However, it faces challenges in recognizing languages such as Korean, which is not major language while training. We address this issue by fine-tuning the Whisper model with an additional dataset comprising about 1,000 hours of Korean speech. We also compare its performance against a Transformer model that was trained from scratch using the same dataset. Our results indicate that fine-tuning the Whisper model significantly improved its Korean speech recognition capabilities in terms of character error rate (CER). Specifically, the performance improved with increasing model size. However, the Whisper model's performance on English deteriorated post fine-tuning, emphasizing the need for further research to develop robust multilingual models. Our study demonstrates the potential of utilizing a fine-tuned Whisper model for Korean ASR applications. Future work will focus on multilingual recognition and optimization for real-time inference.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1998.06a
• /
• pp.296-301
• /
• 1998

The Rough Set theory suggested by Pawlak in 1982 has been useful in AI, machine learning, knowledge acquisition, knowledge discovery from databases, expert system, inductive reasoning. etc. The main advantages of rough set are that it does not need any preliminary or additional information about data and reduce the superfluous informations. but it is a significant disadvantage in the real application that the inference result form is not the real control value but the divided disjoint interval attribute. In order to overcome this difficulty, we will propose approach in which Rough set theory and Neuro-fuzzy fusion are combined to obtain the optimal rule base from lots of input/output datum. These results are applied to the rule construction for infering the temperatures of refrigerator's specified points.

• International Journal of Internet, Broadcasting and Communication
• /
• v.15 no.3
• /
• pp.58-65
• /
• 2023

Bankruptcy is a significant risk for start-up companies, but with the help of cutting-edge artificial intelligence technology, we can now predict bankruptcy with detailed explanations. In this paper, we implemented the Category Boosting algorithm following data cleaning and editing using OpenRefine. We further explained our model using the Shapash library, incorporating domain knowledge. By leveraging the 5C's credit domain knowledge, financial analysts in banks or investors can utilize the detailed results provided by our model to enhance their decision-making processes, even without extensive knowledge about AI. This empowers investors to identify potential bankruptcy risks in their business models, enabling them to make necessary improvements or reconsider their ventures before proceeding. As a result, our model serves as a "glass-box" model, allowing end-users to understand which specific financial indicators contribute to the prediction of bankruptcy. This transparency enhances trust and provides valuable insights for decision-makers in mitigating bankruptcy risks.