• Journal of Intelligence and Information Systems
• /
• v.12 no.2
• /
• pp.167-183
• /
• 2006

Since data mining attempts to find unknown facts or rules by dealing with also vaguely-known data sets, it always suffers from high error rate. In order to reduce the error rate, many researchers have employed multiple models in solving a problem. In this research, we present a new type of multiple models, called DyMoS, whose unique feature is that it classifies the input data and applies the different model developed appropriately for each class of data. In order to evaluate the performance of DyMoS, we applied it to a real customer churn problem of an automobile insurance company, The result shows that the DyMoS outperformed any model which employed only one data mining technique such as artificial neural network, decision tree and case-based reasoning.

• Journal of Information Processing Systems
• /
• v.15 no.5
• /
• pp.1119-1130
• /
• 2019

With the wide spread of Social Network Services (SNS), fake news-which is a way of disguising false information as legitimate media-has become a big social issue. This paper proposes a deep learning architecture for detecting fake news that is written in Korean. Previous works proposed appropriate fake news detection models for English, but Korean has two issues that cannot apply existing models: Korean can be expressed in shorter sentences than English even with the same meaning; therefore, it is difficult to operate a deep neural network because of the feature scarcity for deep learning. Difficulty in semantic analysis due to morpheme ambiguity. We worked to resolve these issues by implementing a system using various convolutional neural network-based deep learning architectures and "Fasttext" which is a word-embedding model learned by syllable unit. After training and testing its implementation, we could achieve meaningful accuracy for classification of the body and context discrepancies, but the accuracy was low for classification of the headline and body discrepancies.

• International Journal of Internet, Broadcasting and Communication
• /
• v.13 no.2
• /
• pp.173-178
• /
• 2021

In this paper, we find a suitable methodology for Korean Sentiment Analysis through a comparative experiment in which methods of embedding and natural network models are learned at the highest accuracy and fastest speed. The embedding method compares word embeddeding and Word2Vec. The model compares and experiments representative neural network models CNN, RNN, LSTM, GRU, Bi-LSTM and Bi-GRU with IKEA review data. Experiments show that Word2Vec and BiGRU had the highest accuracy and second fastest speed with 94.23% accuracy and 42.30 seconds speed. Word2Vec and GRU were found to have the third highest accuracy and fastest speed with 92.53% accuracy and 26.75 seconds speed.

• Journal of Information Processing Systems
• /
• v.19 no.4
• /
• pp.439-449
• /
• 2023

This paper deals with a resource efficient artificial intelligence (AI) service architecture for multi-channel video streams. As an AI service, we consider the object detection model, which is the most representative for video applications. Since most object detection models are basically designed for a single channel video stream, the utilization of the additional resource for multi-channel video stream processing is inevitable. Therefore, we propose a resource efficient AI service framework, which can be associated with various AI service models. Our framework is designed based on the modular architecture, which consists of adaptive frame control (AFC) Manager, multiplexer (MUX), adaptive channel selector (ACS), and YOLO interface units. In order to run only a single YOLO process without regard to the number of channels, we propose a novel approach efficiently dealing with multi-channel input streams. Through the experiment, it is shown that the framework is capable of performing object detection service with minimum resource utilization even in the circumstance of multi-channel streams. In addition, each service can be guaranteed within a deadline.

• 대한상한금궤의학회지
• /
• v.14 no.1
• /
• pp.41-50
• /
• 2022

Objective : The purpose of this study is to explore the most suitable machine learning model algorithm for Shanghanlun diagnostic system classification using natural language processing (NLP). Methods : A total of 201 data items were collected from 『Shanghanlun』 and 『Clinical Shanghanlun』, 'Taeyangbyeong-gyeolhyung' and 'Eumyangyeokchahunobokbyeong' were excluded to prevent oversampling or undersampling. Data were pretreated using a twitter Korean tokenizer and trained by logistic regression, ridge regression, lasso regression, naive bayes classifier, decision tree, and random forest algorithms. The accuracy of the models were compared. Results : As a result of machine learning, ridge regression and naive Bayes classifier showed an accuracy of 0.843, logistic regression and random forest showed an accuracy of 0.804, and decision tree showed an accuracy of 0.745, while lasso regression showed an accuracy of 0.608. Conclusions : Ridge regression and naive Bayes classifier are suitable NLP machine learning models for the Shanghanlun diagnostic system classification.

• Computers and Concrete
• /
• v.29 no.3
• /
• pp.187-199
• /
• 2022

The usability of waste materials as raw materials is necessary for sustainable production. This study investigates the effects of different powder materials used to replace cement (0%, 5% and 10%) and standard sand (0%, 20% and 30%) (basalt, limestone, and dolomite) on the compressive strength (f_c), flexural strength (f_r), and ultrasonic pulse velocity (UPV) of mortars exposed to freeze-thaw cycles (56, 86, 126, 186 and 226 cycles). Furthermore, the usability of artificial intelligence models is compared, and the prediction accuracy of the outputs is examined according to the inputs (powder type, replacement ratio, and the number of cycles). The results show that the variability of the outputs was significantly high under the freeze-thaw effect in mortars produced with waste powder instead of those produced with cement and with standard sand. The highest prediction accuracy for all outputs was obtained using the adaptive-network-based fuzzy inference system model. The significantly high prediction accuracy was obtained for the UPV, f_c, and f_r of mortars produced using waste powders instead of standard sand (R² of UPV, f_c and f_f is 0.931, 0.759 and 0.825 respectively), when under the freeze-thaw effect. However, for the mortars produced using waste powders instead of cement, the prediction accuracy of UPV was significantly high (R²=0.889) but the prediction accuracy of f_c and f_r was low (R²f_c=0.612 and R²f_f=0.334).

• Journal of Internet Computing and Services
• /
• v.24 no.2
• /
• pp.67-74
• /
• 2023

Stock price prediction research in the financial sector aims to ensure trading stability and achieve profit realization. Conventional statistical prediction techniques are not reliable for actual trading decisions due to low prediction accuracy compared to randomly predicted results. Artificial intelligence models improve accuracy by learning data characteristics and fluctuation patterns to make predictions. However, predicting stock prices using long-term time series data remains a challenging problem. This paper proposes a stable and reliable stock price prediction method using K-means clustering-based data augmentation and normalization techniques and LSTM models specialized in time series learning. This enables obtaining more accurate and reliable prediction results and pursuing high profits, as well as contributing to market stability.

• The Journal of Bigdata
• /
• v.7 no.1
• /
• pp.37-47
• /
• 2022

Autonomous delivery operation data is the key to driving a paradigm shift for last-mile delivery in the Corona era. To bridge the technological gap between domestic autonomous delivery robots and overseas technology-leading countries, large-scale data collection and verification that can be used for artificial intelligence training is required as the top priority. Therefore, overseas technology-leading countries are contributing to verification and technological development by opening AI training data in public data that anyone can use. In this paper, 326 objects were collected to trainn autonomous delivery robots, and artificial intelligence models such as Mask r-CNN and Yolo v3 were trained and verified. In addition, the two models were compared based on comparison and the elements required for future autonomous delivery robot research were considered.

• Journal of the Microelectronics and Packaging Society
• /
• v.30 no.4
• /
• pp.69-78
• /
• 2023

With the rapid growth of artificial intelligence, the demand for semiconductors is enormously increasing everywhere. To ensure the manufacturing quality and quantity simultaneously, the importance of automatic defect detection during the packaging process has been re-visited by adapting various deep learning-based methodologies into automatic packaging defect inspection. Deep learning (DL) models require a large amount of data for training, but due to the nature of the semiconductor industry where security is important, sharing and labeling of relevant data is challenging, making it difficult for model training. In this study, we propose a new framework for securing sufficient data for DL models with fewer computing resources through a divide-and-conquer approach. The proposed method divides high-resolution images into pre-defined sub-regions and assigns conditional labels to each region, then trains individual sub-regions and boundaries with boundary loss inducing the globally coherent and seamless images. Afterwards, full-size image is reconstructed by combining divided sub-regions. The experimental results show that the images obtained through this research have high efficiency, consistency, quality, and generality.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.23 no.6
• /
• pp.15-20
• /
• 2023

This paper is a study on the development of an artificial intelligence (AI) system algorithm that recommends indirect advertising products suitable for character web dramas. The goal of this study is to increase viewers' content immersion and help them understand the story of the drama more deeply by recommending indirect advertising products that are suitable for writing lines for web dramas. In this study, we analyze dialogue and plot using the natural language processing model GPT, and develop two types of indirect advertising product recommendation systems, including prop type and background type, based on the analysis results. Through this, products that fit the story of the web drama are appropriately placed, allowing indirect advertisements to be exposed naturally, thereby increasing viewer immersion and enhancing the effectiveness of product promotion. There are limitations of artificial intelligence models, such as the difficulty in fully understanding hidden meanings or cultural nuances, and the difficulty in securing sufficient data for learning. However, this study will provide new insights into how AI can contribute to the production of creative works, and will be an important stepping stone to expand the possibilities of using natural language processing models in the creative industry.