• KIPS Transactions on Software and Data Engineering
• /
• v.6 no.9
• /
• pp.445-456
• /
• 2017

In this paper, we propose a method to detect spam tweets containing unhealthy information by using an n-gram dictionary under limited labeling. Spam tweets that contain unhealthy information have a tendency to use similar words and sentences. Based on this characteristic, we show that spam tweets can be effectively detected by applying a Naive Bayesian classifier using n-gram dictionaries which are constructed from spam tweets and normal tweets. On the other hand, constructing an initial training set requires very high cost because a large amount of data flows in real time in a twitter. Therefore, there is a need for a spam detection method that can be applied in an environment where the initial training set is very small or non exist. To solve the problem, we propose a method to generate pseudo-labels by utilizing twitter's retweet function and use them for the configuration of the initial training set and the n-gram dictionary update. The results from various experiments using 1.3 million korean tweets collected from December 1, 2016 to December 7, 2016 prove that the proposed method has superior performance than the compared spam detection methods.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.7 no.3
• /
• pp.532-543
• /
• 2003

In this paper, we proposed structure of a reversible discrete-time cellular neural network (DTCNN) for labeling digital images to protect copylight. First, we present the concept and the structure of reversible DTCNN, which can be used to generate 2D binary pseudo-random images sequences. We presented some, output examples of different kinds of reversible DTCNNs to show their complex behaviors. Then both the original image and the copyright label, which is often another binary image, are used to generate a binary random key image. The key image is then used to scramble the original image. Since the reversibility of a reversible DTCNN, the same reversible DTCNN can recover the copyright label from a labeled image. Due to the high speed of a DTCNN chip, our method can be used to label image sequences, e.g., video sequences, in real time. Computer simulation results are presented.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.12
• /
• pp.67-77
• /
• 2023

This study represents an innovative research conducted in the smart farm environment, developing a deep learning-based disease and pest detection model and applying it to the Intelligent Internet of Things (IoT) platform to explore new possibilities in the implementation of digital agricultural environments. The core of the research was the integration of the latest ImageNet models such as Pseudo-Labeling, RegNet, EfficientNet, and preprocessing methods to detect various diseases and pests in complex agricultural environments with high accuracy. To this end, ensemble learning techniques were applied to maximize the accuracy and stability of the model, and the model was evaluated using various performance indicators such as mean Average Precision (mAP), precision, recall, accuracy, and box loss. Additionally, the SHAP framework was utilized to gain a deeper understanding of the model's prediction criteria, making the decision-making process more transparent. This analysis provided significant insights into how the model considers various variables to detect diseases and pests.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.24 no.1
• /
• pp.83-88
• /
• 2024

In this research, we propose a Semi-Supervised learning based railroad surface defect detection method. The Resnet50 model, pretrained on ImageNet, was employed for the training. Data without labels are randomly selected, and then labeled to train the ResNet50 model. The trained model is used to predict the results of the remaining unlabeled training data. The predicted values exceeding a certain threshold are selected, sorted in descending order, and added to the training data. Pseudo-labeling is performed based on the class with the highest probability during this process. An experiment was conducted to assess the overall class classification performance based on the initial number of labeled data. The results showed an accuracy of 98% at best with less than 10% labeled training data compared to the overall training data.

• ETRI Journal
• /
• v.42 no.4
• /
• pp.518-526
• /
• 2020

This paper presents an efficient hardware random-number generator based on a beta source. The proposed generator counts the values of "0" and "1" and provides a method to distinguish between pseudo-random and true random numbers by comparing them using simple cumulative operations. The random-number generator produces labeled data indicating whether the count value is a pseudo- or true random number according to its bit value based on the generated labeling data. The proposed method is verified using a system based on Verilog RTL coding and LabVIEW for hardware implementation. The generated random numbers were tested according to the NIST SP 800-22 and SP 800-90B standards, and they satisfied the test items specified in the standard. Furthermore, the hardware is efficient and can be used for security, artificial intelligence, and Internet of Things applications in real time.

• BMB Reports
• /
• v.30 no.2
• /
• pp.157-161
• /
• 1997

The present paper reports characteristics and specificity of the inhibitory action of $N^{\alpha}-tosyl-L-lysine-chloromethyl\;ketone$ (TLCK) and $N^{\alpha}-tosyl-L-phenylalanine-chloromethyl\;ketone$ (TPCK) on the glucose6-phosphate transporter of rat liver microsomes. The TLCK-induced inhibition was pH dependent. The inhibition constants for TPCK were determined by following pseudo-Lst order reaction mechanism. The inhibition was protected by preincubation with excess amount of glucose-6-phosphate. The results proved that (a) TLCK inactivates the microsomal glucose-6-phosphate transporter, (b) the inhibition results from the modification of sulfhydryl groups of the transporter.

• Journal of Chest Surgery
• /
• v.54 no.3
• /
• pp.232-234
• /
• 2021

Ependymomas arise from ependymal cells and can grow at any site in the central nervous system (CNS), as well as in some locations outside of the CNS. The latter is rare, contributing to the frequent misdiagnoses of such cases. Herein, we present the case of a 54-year-old man with a history of lower limb weakness and numbness. Magnetic resonance imaging revealed an extradural, heterogeneously enhanced solid lesion with a regular and well-defined border in the posterior mediastinum. A post-resection histopathological examination revealed tumor-forming perivascular pseudo-rosettes that showed immunoreactivity against glial fibrillary acidic protein, epithelial membrane antigen, and vimentin, as well as a high Ki-67 labeling index. Based on pathological features, a diagnosis of anaplastic ependymoma was established.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2024.01a
• /
• pp.73-75
• /
• 2024

본 논문에서는 밀 수확량을 증가시키기 위한 일반화된 검출 모델을 제안한다. 일반화 성능을 높이기 위해 CutMix 알고리즘으로 데이터를 증식시켰고, 라벨링 되지 않은 데이터를 최대한 활용하기 위해 Fast R-CNN 기반 Pseudo labeling을 사용하였다. 학습의 정확성과 효율성을 높이기 위해 사전에 훈련된 EfficientDet 모델로 학습하였으며, OOF를 이용하여 검증하였다. 최신 객체 검출 모델과 IoU(Intersection over Union)를 이용한 성능 평가 결과, 제안된 모델이 가장 높은 성능을 보이는 것을 확인하였다.

• International Journal of Internet, Broadcasting and Communication
• /
• v.16 no.1
• /
• pp.124-131
• /
• 2024

Breast cancer classification using ultrasound, while widely employed, faces challenges due to its relatively low predictive value arising from significant overlap in characteristics between benign and malignant lesions, as well as operator-dependency. To alleviate these challenges and reduce dependency on radiologist interpretation, the implementation of automatic breast cancer classification in ultrasound image can be helpful. To deal with this problem, we propose a semi-supervised deep learning framework for breast cancer classification. In the proposed method, we could achieve reasonable performance utilizing less than 50% of the training data for supervised learning in comparison to when we utilized a 100% labeled dataset for training. Though it requires more modification, this methodology may be able to alleviate the time-consuming annotation burden on radiologists by reducing the number of annotation, contributing to a more efficient and effective breast cancer detection process in ultrasound images.

• Korean Journal of Radiology
• /
• v.22 no.2
• /
• pp.243-252
• /
• 2021

Objective: To compare and correlate the findings of intravoxel incoherent motion (IVIM) magnetic resonance (MR) imaging and arterial spin labeling (ASL) imaging in characterizing parotid gland tumors. Materials and Methods: We retrospectively reviewed 56 patients with parotid gland tumors evaluated by MR imaging. The true diffusion coefficient (D), pseudo-diffusion coefficient (D^*), and fraction of perfusion (f) values of IVIM imaging and tumor-to-parotid gland signal intensity ratio (SIR) on ASL imaging were calculated. Spearman rank correlation coefficient, chi-squared, Mann-Whitney U, and Kruskal-Wallis tests with the post-hoc Dunn-Bonferroni method and receiver operating characteristic curve assessments were used for statistical analysis. Results: Malignant parotid gland tumors showed significantly lower D than benign tumors (p = 0.019). Within subgroup analyses, pleomorphic adenomas (PAs) showed significantly higher D than malignant tumors (MTs) and Warthin's tumors (WTs) (p < 0.001). The D^* of WTs was significantly higher than that of PAs (p = 0.031). The f and SIR on ASL imaging of WTs were significantly higher than those of MTs and PAs (p < 0.05). Significantly positive correlation was found between SIR on ASL imaging and f (r = 0.446, p = 0.001). In comparison with f, SIR on ASL imaging showed a higher area under curve (0.853 vs. 0.891) in discriminating MTs from WTs, although the difference was not significant (p = 0.720). Conclusion: IVIM and ASL imaging could help differentiate parotid gland tumors. SIR on ASL imaging showed a significantly positive correlation with f. ASL imaging might hold potential to improve the ability to discriminate MTs from WTs.