• Journal of Periodontal and Implant Science
• /
• v.48 no.2
• /
• pp.114-123
• /
• 2018

Purpose: The aim of the current study was to develop a computer-assisted detection system based on a deep convolutional neural network (CNN) algorithm and to evaluate the potential usefulness and accuracy of this system for the diagnosis and prediction of periodontally compromised teeth (PCT). Methods: Combining pretrained deep CNN architecture and a self-trained network, periapical radiographic images were used to determine the optimal CNN algorithm and weights. The diagnostic and predictive accuracy, sensitivity, specificity, positive predictive value, negative predictive value, receiver operating characteristic (ROC) curve, area under the ROC curve, confusion matrix, and 95% confidence intervals (CIs) were calculated using our deep CNN algorithm, based on a Keras framework in Python. Results: The periapical radiographic dataset was split into training (n=1,044), validation (n=348), and test (n=348) datasets. With the deep learning algorithm, the diagnostic accuracy for PCT was 81.0% for premolars and 76.7% for molars. Using 64 premolars and 64 molars that were clinically diagnosed as severe PCT, the accuracy of predicting extraction was 82.8% (95% CI, 70.1%-91.2%) for premolars and 73.4% (95% CI, 59.9%-84.0%) for molars. Conclusions: We demonstrated that the deep CNN algorithm was useful for assessing the diagnosis and predictability of PCT. Therefore, with further optimization of the PCT dataset and improvements in the algorithm, a computer-aided detection system can be expected to become an effective and efficient method of diagnosing and predicting PCT.

• Journal of Digital Convergence
• /
• v.11 no.11
• /
• pp.409-414
• /
• 2013

Naive Bayesian classifiers are a powerful and well-known type of classifiers that can be easily induced from a dataset of sample cases. However, the strong conditional independence assumptions can sometimes lead to weak classification performance. Normally, naive Bayesian classifiers use Gaussian distributions to handle continuous attributes and to represent the likelihood of the features conditioned on the classes. The probability density of attributes, however, is not always well fitted by a Gaussian distribution. Another eminent type of classifier is the neuro-fuzzy classifier, which can learn fuzzy rules and fuzzy sets using supervised learning. Since there are specific structural similarities between a neuro-fuzzy classifier and a naive Bayesian classifier, the purpose of this study is to apply learning distribution graphs constructed by a neuro-fuzzy network to naive Bayesian classifiers. We compare the Gaussian distribution graphs with the fuzzy distribution graphs for the naive Bayesian classifier. We applied these two types of distribution graphs to classify leukemia and colon DNA microarray data sets. The results demonstrate that a naive Bayesian classifier with fuzzy distribution graphs is more reliable than that with Gaussian distribution graphs.

• The KIPS Transactions:PartB
• /
• v.15B no.1
• /
• pp.1-8
• /
• 2008

Generally, moving objects in surveillance video are extracted by background subtraction or frame difference method. However, moving cast shadows on object distort extracted figures which cause serious detection problems. Especially, analyzing vehicle information in video frames from a fixed surveillance camera on road, we obtain inaccurate results by shadow which vehicle causes. So, Shadow Elimination is essential to extract right objects from frames in surveillance video. And we use shadow removal algorithm for vehicle classification. In our paper, as we suppress moving cast shadow in object, we efficiently discriminate vehicle types. After we fit new object of shadow-removed object as three dimension object, we use extracted attributes for supervised learning to classify vehicle types. In experiment, we use 3 learning methods {IBL, C4.5, NN(Neural Network)} so that we evaluate the result of vehicle classification by shadow elimination.

• Journal of KIISE
• /
• v.42 no.9
• /
• pp.1124-1132
• /
• 2015

In order to train a supervised machine learning algorithm, it is necessary to have non-biased labels and a sufficient amount of training data. However, it is difficult to collect the required non-biased labels and a sufficient amount of training data to develop an automatic English Composition scoring system. In addition, an English writing assessment is carried out using a multi-faceted evaluation of the overall level of the answer. Therefore, it is difficult to choose an appropriate machine learning algorithm for such work. In this paper, we show that it is possible to alleviate these problems through ensemble learning. The results of the experiment indicate that the ensemble technique exhibited an overall performance that was better than that of other algorithms.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.29 no.3
• /
• pp.549-555
• /
• 2019

According to a recent report by anti-virus vendors, the number of new and modified malware increased exponentially. Therefore, malware analysis research using machine learning has been actively researched in order to replace passive analysis method which has low analysis speed. However, when using supervised learning based machine learning, many studies use low-reliability malware family name provided by the antivirus vendor as the label. In order to solve the problem of low-reliability of malware label, this paper introduces a new labeling technique, "Unified Labeling", and further verifies the malicious behavior similarity through the feature analysis of the fine-grained method. To verify this study, various clustering algorithms were used and compared with existing labeling techniques.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.14 no.1
• /
• pp.169-178
• /
• 2019

Now a day, we can perform various predictions by applying machine learning, which is a field of artificial intelligence; however, the finding of best algorithm in the field is always the problem. This paper predicts monthly power trading amount, monthly power trading amount of money, monthly index of production extension, final consumption of energy, and diesel for automotive using machine learning supervised algorithms. Then, we find most fit algorithm among them for each case. To do this we show the probability of predicting the value for monthly power trading amount and monthly power trading amount of money, monthly index of production extension, final consumption of energy, and diesel for automotive. Then, we try to average each predicting values. Finally, we confirm which algorithm is the most superior algorithm among them.

• Journal of the Economic Geographical Society of Korea
• /
• v.25 no.1
• /
• pp.171-181
• /
• 2022

The real estate price index plays key roles as quantitative data in real estate market analysis. International organizations including OECD publish the real estate price indexes by country, and the Korea Real Estate Board announces metropolitan-level and municipal-level indexes. However, when the index is set on the smaller spatial unit level than metropolitan and municipal-level, problems occur: missing values. As the spatial scope is narrowed down, there are cases where there are few or no transactions depending on the unit period, which lead index calculation difficult or even impossible. This study suggests a supervised learning-based machine learning model to compensate for missing values that may occur due to no transaction in a specific range and period. The models proposed in our research verify the accuracy of predicting the existing values and missing values.

• The Journal of the Convergence on Culture Technology
• /
• v.8 no.4
• /
• pp.347-353
• /
• 2022

In the restaurant industry, start-ups are active due to high demand from consumers and low entry barriers. However, the restaurant industry has a high closure rate, and in the case of franchises, there is a large deviation in sales within the same brand. Thus, research is needed to prevent the closure of food franchises. Therefore, this study examines the factors affecting franchise sales and uses machine learning techniques to predict the success and failure of franchises. Various factors that affect franchise sales are extracted by using Point of Sale (PoS) data of food franchise and public data in Gangnam-gu, Seoul. And for more valid variable selection, multicollinearity is removed by using Variance Inflation Factor (VIF). Finally, classification models are used to predict the success and failure of food franchise stores. Through this method, we propose success and failure prediction model for food franchise stores with the accuracy of 0.92.

• Journal of the Korea Computer Graphics Society
• /
• v.27 no.3
• /
• pp.1-11
• /
• 2021

Slow sync is a photography technique where a user takes an image with long exposure and a camera flash to enlighten the foreground and background. Unlike short exposure with flash and long exposure without flash, slow sync guarantees the bright foreground and background in the dim environment. However, taking a slow sync image with a smartphone is difficult because the smartphone camera has continuous and weak flash and can not turn on flash if the exposure time is long. This paper proposes a deep learning method that input is a short exposure flash image and output is a slow sync image. We present a deep learning network with a weight map for spatially varying enlightenment. We also propose a dataset that consists of smartphone short exposure flash images and slow sync images for supervised learning. We utilize the linearity of a RAW image to synthesize a slow sync image from short exposure flash and long exposure no-flash images. Experimental results show that our method trained with our dataset synthesizes slow sync images effectively.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.46 no.1
• /
• pp.165-181
• /
• 2022

The fashion industry is creating innovative business models using artificial intelligence. To efficiently utilize artificial intelligence (AI), fashion data must be classified. Until now, such data have been classified focusing only on the objective properties of fashion products. Their subjective attributes, such as fashion brand sensibilities, are holistic and heuristic intuitions created by a combination of design elements. This study aims to improve the performance of collaborative filtering in the fashion industry by extracting fashion brand sensibility using computer vision technology. The image data set of fashion brand sensibility consists of high-end fashion brand photos that share sensibilities and communicate well in fashion. About 26,000 fashion photos of 11 high-end fashion brand sensibility labels have been collected from the 16FW to 21SS runway and 50 years of US Vogue magazines beginning from 1971. We use EfficientNet-B1 to establish the main architecture and fine-tune the network with ImageNet-ILSVRC. After training fashion brand sensibilities through deep learning, the proposed model achieved an F-1 score of 74% on accuracy tests. Furthermore, as a result of comparing AI machine and human experts, the proposed model is expected to be expanded to mass fashion brands.