• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.1
• /
• pp.334-349
• /
• 2022

The effectiveness of Wi-Fi networks is greatly influenced by the optimization of contention window (CW) parameters. Unfortunately, the conventional approach employed by IEEE 802.11 wireless networks is not scalable enough to sustain consistent performance for the increasing number of stations. Yet, it is still the default when accessing channels for single-users of 802.11 transmissions. Recently, there has been a spike in attempts to enhance network performance using a machine learning (ML) technique known as reinforcement learning (RL). Its advantage is interacting with the surrounding environment and making decisions based on its own experience. Deep RL (DRL) uses deep neural networks (DNN) to deal with more complex environments (such as continuous state spaces or actions spaces) and to get optimum rewards. As a result, we present a new approach of CW control mechanism, which is termed as contention window threshold (CW_Threshold). It uses the DRL principle to define the threshold value and learn optimal settings under various network scenarios. We demonstrate our proposed method, known as a smart exponential-threshold-linear backoff algorithm with a deep Q-learning network (SETL-DQN). The simulation results show that our proposed SETL-DQN algorithm can effectively improve the throughput and reduce the collision rates.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.33B no.3
• /
• pp.169-175
• /
• 1996

In this paper, a new parallel neural network system that performs dynamic competitive learning is proposed. Conventional learning mehtods utilize the full dimension of the original input patterns. However, a particular attribute or dimension of the input patterns does not necessarily contribute to classification. The proposed system consists of parallel neural networks with the reduced input dimension in order to take advantage of the information in each dimension of the input patterns. Consensus schemes were developed to decide the netowrks performs a competitive learning that dynamically generates output neurons as learning proceeds. Each output neuron has it sown class threshold in the proposed dynamic competitive learning. Because the class threshold in the proposed dynamic learning phase, the proposed neural netowrk adapts properly to the input patterns distribution. Experimental results with remote sensing and speech data indicate the improved performance of the proposed method compared to the conventional learning methods.

• International Journal of Computer Science & Network Security
• /
• v.23 no.1
• /
• pp.33-40
• /
• 2023

Machine-learning techniques are discovering effective performance on data analytics. Classification and regression are supported for prediction on different kinds of data. There are various breeds of classification techniques are using based on nature of data. Threshold determination is essential to making better model for unlabelled data. In this paper, threshold value applied as range, based on min-max normalization technique for creating labels and multiclass classification performed on rainfall data. Binary classification is applied on autism data and classification techniques applied on child abuse data. Performance of each technique analysed with the evaluation metrics.

• Current Optics and Photonics
• /
• v.8 no.3
• /
• pp.239-245
• /
• 2024

This paper proposes single logarithmic amplification (single-LA) and deep learning (DL)-based fixed-threshold on-off keying (OOK) detection for free-space optical (FSO) communication. Multilevel LAs (MLAs) can be used to mitigate intensity fluctuations in the received OOK signal by their nonlinear gain characteristics; however, it is ineffective in the case of high scintillation, owing to degradation of the OOK signal's extinction ratio. Therefore, a DL technique is applied to realize effective scintillation compensation in single-LA applications. Fully connected (FC) networks and fully connected neural networks (FCNN), which have nonlinear modeling characteristics, are deployed in this work. The performance of the proposed method is evaluated through simulations under various scintillation effects. Simulation results show that the proposed method outperforms the conventional adaptive-threshold-decision, single-LA-based, MLA-based, FC-based, and FCNN-based OOK detection techniques.

• Asia-Pacific Journal of Atmospheric Sciences
• /
• v.54 no.4
• /
• pp.527-544
• /
• 2018

This paper presents a nighttime sea fog detection algorithm incorporating unsupervised learning technique. The algorithm is based on data sets that combine brightness temperatures from the $3.7{\mu}m$ and $10.8{\mu}m$ channels of the meteorological imager (MI) onboard the Communication, Ocean and Meteorological Satellite (COMS), with sea surface temperature from the Operational Sea Surface Temperature and Sea Ice Analysis (OSTIA). Previous algorithms generally employed threshold values including the brightness temperature difference between the near infrared and infrared. The threshold values were previously determined from climatological analysis or model simulation. Although this method using predetermined thresholds is very simple and effective in detecting low cloud, it has difficulty in distinguishing fog from stratus because they share similar characteristics of particle size and altitude. In order to improve this, the unsupervised learning approach, which allows a more effective interpretation from the insufficient information, has been utilized. The unsupervised learning method employed in this paper is the expectation-maximization (EM) algorithm that is widely used in incomplete data problems. It identifies distinguishing features of the data by organizing and optimizing the data. This allows for the application of optimal threshold values for fog detection by considering the characteristics of a specific domain. The algorithm has been evaluated using the Cloud-Aerosol Lidar with Orthogonal Polarization (CALIOP) vertical profile products, which showed promising results within a local domain with probability of detection (POD) of 0.753 and critical success index (CSI) of 0.477, respectively.

• Journal of Korean Neurosurgical Society
• /
• v.66 no.1
• /
• pp.53-62
• /
• 2023

Objective : Deep learning is a machine learning approach based on artificial neural network training, and object detection algorithm using deep learning is used as the most powerful tool in image analysis. We analyzed and evaluated the diagnostic performance of a deep learning algorithm to identify skull fractures in plain radiographic images and investigated its clinical applicability. Methods : A total of 2026 plain radiographic images of the skull (fracture, 991; normal, 1035) were obtained from 741 patients. The RetinaNet architecture was used as a deep learning model. Precision, recall, and average precision were measured to evaluate the deep learning algorithm's diagnostic performance. Results : In ResNet-152, the average precision for intersection over union (IOU) 0.1, 0.3, and 0.5, were 0.7240, 0.6698, and 0.3687, respectively. When the intersection over union (IOU) and confidence threshold were 0.1, the precision was 0.7292, and the recall was 0.7650. When the IOU threshold was 0.1, and the confidence threshold was 0.6, the true and false rates were 82.9% and 17.1%, respectively. There were significant differences in the true/false and false-positive/false-negative ratios between the anterior-posterior, towne, and both lateral views (p=0.032 and p=0.003). Objects detected in false positives had vascular grooves and suture lines. In false negatives, the detection performance of the diastatic fractures, fractures crossing the suture line, and fractures around the vascular grooves and orbit was poor. Conclusion : The object detection algorithm applied with deep learning is expected to be a valuable tool in diagnosing skull fractures.

• Journal of the Korean Chemical Society
• /
• v.58 no.1
• /
• pp.126-129
• /
• 2014

Concerning the difficulty of learning science and reduced interest in science, the authors of this study searched for potential threshold concepts which are portals or gateways in the field of science (particularly chemistry). The nature of these concepts and how to overcome their troublesomeness were further questioned. For this study, 239 high school students completed chemistry II provided information about what difficult concepts and potential threshold concepts in high school chemistry are and how they affect learning chemistry. In particular, the mastery experience of the threshold concepts was explored in detail. Two, "mole and atomic structure" were selected as threshold concepts in chemistry. Not only as important but also as threshold, this study emphasized the importance of the two concepts in terms of features characterizing them as threshold concepts. In particular, the features objectify subjective experiences of students and provide information describing the scientific meaning and distinctive nature of threshold concepts in science. Along with the data from teachers, this study shows the integrative feature as key criteria for students to make meaningful understanding of the two threshold concepts.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.27 no.3
• /
• pp.319-328
• /
• 2024

Semi-supervised learning is a good way to train a classification model using a small number of labeled and large number of unlabeled data. We applied semi-supervised learning to a synthetic aperture radar(SAR) image classification model with a limited number of datasets that are difficult to create. To address the previous difficulties, semi-supervised learning uses a model trained with a small amount of labeled data to generate and learn pseudo labels. Besides, a lot of number of papers use a single fixed threshold to create pseudo labels. In this paper, we present a semi-supervised synthetic aperture radar(SAR) image classification method that applies different thresholds for each class instead of all classes sharing a fixed threshold to improve SAR classification performance with a small number of labeled datasets.

• The Transactions of the Korean Institute of Electrical Engineers
• /
• v.40 no.7
• /
• pp.718-724
• /
• 1991

In this paper, we proposed an algorithm to improve the memory efficiency by means of learning thresholds in spite of correlations among input patterns to be memorized. The proposed algorithm does not need preprocess correlations among input patterns but processes them with a threshold on a neural network. When memory contents are destroyed by correlation, nearly all patterns can be properly recovered with past learning. Through experiments we show how out algorithm can improve the memory efficiency.

• Journal of the Korean Chemical Society
• /
• v.59 no.4
• /
• pp.308-319
• /
• 2015

As a follow-up study to identify references for threshold concepts in science, 20 high school chemistry teachers were interviewed. Seven concepts were identified as threshold concepts. The data revealed that teachers overcome the thresholds while they are teaching as well as learning during their school years. This explains that the mastery experience of threshold concepts involve not only the process of creating subject matter knowledge of a learner but also the reflection on or preparation for teaching. Hence, the current study proposes that a strong relationship exists between the mastery experience of threshold concepts and the development of teachers’ pedagogical content knowledge (PCK). In this regard, findings from this study will provide valuable information to understand the nature of threshold concepts and suggests the value of mastery experience of threshold concepts in terms of PCK development.