• Journal of Korean Society for Quality Management
• /
• v.49 no.2
• /
• pp.201-211
• /
• 2021

Purpose: This paper proposes a density adaptive grid algorithm for the k-NN regression model to reduce the computation time for large datasets without significant prediction accuracy loss. Methods: The proposed method utilizes the concept of the grid with centroid to reduce the number of reference data points so that the required computation time is much reduced. Since the grid generation process in this paper is based on quantiles of original variables, the proposed method can fully reflect the density information of the original reference data set. Results: Using five real-life datasets, the proposed k-NN regression model is compared with the original k-NN regression model. The results show that the proposed density adaptive grid-based k-NN regression model is superior to the original k-NN regression in terms of data reduction ratio and time efficiency ratio, and provides a similar prediction error if the appropriate number of grids is selected. Conclusion: The proposed density adaptive grid algorithm for the k-NN regression model is a simple and effective model which can help avoid a large loss of prediction accuracy with faster execution speed and fewer memory requirements during the testing phase.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.23 no.6
• /
• pp.89-95
• /
• 2023

In recent decades, human action recognition (HAR) has demonstrated potential applications in sports analysis, human-robot interaction, and large-scale signage content. In this paper, spatial temporal attention graph convolutional network (STAGCN)-based HAR system is proposed. Spatioal-temmporal features of skeleton sequences are assigned different weights by STAGCN, enabling the consideration of key joints and viewpoints. From simulation results, it has been shown that the performance of the proposed model can be improved in terms of classification accuracy in the NTU RGB+D dataset.

• Journal of the Korean Data and Information Science Society
• /
• v.23 no.5
• /
• pp.949-959
• /
• 2012

This article is to find the right size of decision trees that performs better for boosting algorithm. First we defined the tree size D as the depth of a decision tree. Then we compared the performance of boosting algorithm with different tree sizes in the experiment. Although it is an usual practice to set the tree size in boosting algorithm to be small, we figured out that the choice of D has a significant influence on the performance of boosting algorithm. Furthermore, we found out that the tree size D need to be sufficiently large for some dataset. The experiment result shows that there exists an optimal D for each dataset and choosing the right size D is important in improving the performance of boosting. We also tried to find the model for estimating the right size D suitable for boosting algorithm, using variables that can explain the nature of a given dataset. The suggested model reveals that the optimal tree size D for a given dataset can be estimated by the error rate of stump tree, the number of classes, the depth of a single tree, and the gini impurity.

• The KIPS Transactions:PartA
• /
• v.18A no.3
• /
• pp.81-92
• /
• 2011

There has been a number of attempts to apply 3D CAD data created in the design stage of product life cycle to various applications of the other stages in related industries. But, 3D CAD data requires a large amount of computing resources for data processing, and it is not suitable for post applications such as distributed collaboration, marketing tool, or Interactive Electronic Technical Manual because of the design information security problem and the license cost. Therefore, various lightweight visualization formats and application systems have been suggested to overcome these problems. However, most of these lightweight formats are dependent on the companies or organizations which suggested them and cannot be shared with each other. In addition, product structure information is not represented along with the product geometric information. In this paper, we define a dataset called prod-X3D(Enhanced X3D Dataset for Web-based Visualization of 3D CAD Product Model) based on the international standard graphic format, X3D, which can represent the structure information as well as the geometry information of a product, and propose a translation method from 3D CAD data to an prod-X3D.

• The Journal of the Convergence on Culture Technology
• /
• v.8 no.2
• /
• pp.385-391
• /
• 2022

Quality evaluation of beef carcasses is an important issue in the livestock industry. Recently, through the AI monitor system based on artificial intelligence, the quality manager can receive help in making accurate decisions based on the analysis of beef carcass images or result information. This artificial intelligence dataset is an important factor in judging performance. Existing datasets may have different surface orientation or resolution. In this paper, we proposed a two-stage classification model that can efficiently manage the grades of beef carcass image using deep learning. And to overcome the problem of the various conditions of the image, a new dataset of 1,300 images was constructed. The recognition rate of deep network for 5-grade classification using the new dataset was 72.5%. Two-stage evaluation is a method to increase reliability by taking advantage of the large difference between grades 1++, 1+, and grades 1 and 2 and 3. With two experiments using the proposed two stage model, the recognition rates of 73.7% and 77.2% were obtained. As this, The proposed method will be an efficient method if we have a dataset with 100% recognition rate in the first stage.

• Smart Media Journal
• /
• v.13 no.2
• /
• pp.52-61
• /
• 2024

In the shipyard, aerial images are acquired at regular intervals using Unmanned Aerial Vehicles (UAVs) for the management of external storage yards. These images are then investigated by humans to manage the status of the storage yards. This method requires a significant amount of time and manpower especially for large areas. In this paper, we propose an automated management technology based on a semantic segmentation foundation model to address these challenges and accurately assess the status of external storage yards. In addition, as there is insufficient publicly available dataset for external storage yards, we collected a small-scale dataset for external storage yards objects and equipment. Using this dataset, we fine-tune an object detector and extract initial object candidates. They are utilized as prompts for the Segment Anything Model(SAM) to obtain precise semantic segmentation results. Furthermore, to facilitate continuous storage yards dataset collection, we propose a training data generation pipeline using SAM. Our proposed method has achieved 4.00%p higher performance compared to those of previous semantic segmentation methods on average. Specifically, our method has achieved 5.08% higher performance than that of SegFormer.

• Journal of Internet Computing and Services
• /
• v.25 no.1
• /
• pp.57-67
• /
• 2024

This research investigates training methods for large language models to accurately identify sentiments and comprehend information about increasing and decreasing fluctuations in the financial domain. The main goal is to identify suitable datasets that enable these models to effectively understand expressions related to financial increases and decreases. For this purpose, we selected sentences from Wall Street Journal that included relevant financial terms and sentences generated by GPT-3.5-turbo-1106 for post-training. We assessed the impact of these datasets on language model performance using Financial PhraseBank, a benchmark dataset for financial sentiment analysis. Our findings demonstrate that post-training FinBERT, a model specialized in finance, outperformed the similarly post-trained BERT, a general domain model. Moreover, post-training with actual financial news proved to be more effective than using generated sentences, though in scenarios requiring higher generalization, models trained on generated sentences performed better. This suggests that aligning the model's domain with the domain of the area intended for improvement and choosing the right dataset are crucial for enhancing a language model's understanding and sentiment prediction accuracy. These results offer a methodology for optimizing language model performance in financial sentiment analysis tasks and suggest future research directions for more nuanced language understanding and sentiment analysis in finance. This research provides valuable insights not only for the financial sector but also for language model training across various domains.

• International conference on construction engineering and project management
• /
• 2024.07a
• /
• pp.1307-1307
• /
• 2024

Recent research on heavy equipment has been conducted for the purposes of enhanced safety, productivity improvement, and carbon neutrality at construction sites. A sensor-based approach is being explored to monitor the location and movements of heavy equipment in real time. However, it poses significant challenges in terms of time and cost as multiple sensors should be installed on numerous heavy equipment at construction sites. In addition, there is a limitation in identifying the collaboration or interference between two or more heavy equipment. In light of this, a vision-based deep learning approach is being actively conducted to effectively respond to various working conditions and dynamic environments. To enhance the performance of a vision-based activity recognition model, it is essential to secure a sufficient amount of training datasets (i.e., video datasets collected from actual construction sites). However, due to safety and security issues at construction sites, there are limitations in adequately collecting training dataset under various situations and environmental conditions. In addition, the videos feature a sequence of multiple activities of heavy equipment, making it challenging to clearly distinguish the boundaries between preceding and subsequent activities. To address these challenges, this study proposed a domain adaptation in vision-based transfer learning for automated excavator activity recognition utilizing 3D ResNet (residual deep neural network). Particularly, this study aimed to identify the optimal depth of 3D ResNet (i.e., the number of layers of the feature extractor) suitable for domain adaptation via fine-tuning process. To achieve this, this study sought to evaluate the activity recognition performance of five 3D ResNet models with 18, 34, 50, 101, and 152 layers, which used two consecutive videos with multiple activities (5 mins, 33 secs and 10 mins, 6 secs) collected from actual construction sites. First, pretrained weights from large-scale datasets (i.e., Kinetic-700 and Moment in Time (MiT)) in other domains (e.g., humans, animals, natural phenomena) were utilized. Second, five 3D ResNet models were fine-tuned using a customized dataset (14,185 clips, 60,606 secs). As an evaluation index for activity recognition model, the F1 score showed 0.881, 0.689, 0.74, 0.684, and 0.569 for the five 3D ResNet models, with the 18-layer model performing the best. This result indicated that the activity recognition models with fewer layers could be advantageous in deriving the optimal weights for the target domain (i.e., excavator activities) when fine-tuning with a limited dataset. Consequently, this study identified the optimal depth of 3D ResNet that can maintain a reliable performance in dynamic and complex construction sites, even with a limited dataset. The proposed approach is expected to contribute to the development of decision-support systems capable of systematically managing enhanced safety, productivity improvement, and carbon neutrality in the construction industry.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.16 no.4
• /
• pp.193-202
• /
• 2023

In recent years, the outstanding performance of large language models (LLMs) trained on extensive datasets has become a hot topic. Since studies on LLMs are available on open-source approaches, the ecosystem is expanding rapidly. Models that are task-specific, lightweight, and high-performing are being actively disseminated using additional training techniques using pre-trained LLMs as foundation models. On the other hand, the performance of LLMs for Korean is subpar because English comprises a significant proportion of the training dataset of existing LLMs. Therefore, research is being carried out on Korean-specific LLMs that allow for further learning with Korean language data. This paper identifies trends of open source based LLMs and introduces research on Korean specific large language models; moreover, the applications and limitations of large language models are described.

• Journal of the Korea Society of Computer and Information
• /
• v.25 no.4
• /
• pp.29-35
• /
• 2020

In this paper, we propose a new structured entity recognition DeNERT model. Recently, the field of natural language processing has been actively researched using pre-trained language representation models with a large amount of corpus. In particular, the named entity recognition, which is one of the fields of natural language processing, uses a supervised learning method, which requires a large amount of training dataset and computation. Reinforcement learning is a method that learns through trial and error experience without initial data and is closer to the process of human learning than other machine learning methodologies and is not much applied to the field of natural language processing yet. It is often used in simulation environments such as Atari games and AlphaGo. BERT is a general-purpose language model developed by Google that is pre-trained on large corpus and computational quantities. Recently, it is a language model that shows high performance in the field of natural language processing research and shows high accuracy in many downstream tasks of natural language processing. In this paper, we propose a new named entity recognition DeNERT model using two deep learning models, DQN and BERT. The proposed model is trained by creating a learning environment of reinforcement learning model based on language expression which is the advantage of the general language model. The DeNERT model trained in this way is a faster inference time and higher performance model with a small amount of training dataset. Also, we validate the performance of our model's named entity recognition performance through experiments.