• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.05a
• /
• pp.676-677
• /
• 2022

In this paper, we propose a new method based on a deep learning model with an optimized dynamic decay learning rate and improved model structure to achieve fast and accurate classification of fashion clothing images. Experiments are performed using the model proposed in the Fashion-MNIST dataset and compared with methods of CNN, LeNet, LSTM and BiLSTM.

• Journal of the Semiconductor & Display Technology
• /
• v.22 no.3
• /
• pp.84-89
• /
• 2023

In water resource management, data prediction is performed using artificial intelligence, and companies, governments, and institutions continue to attempt to efficiently manage resources through this. LSTM is a model specialized for processing time series data, which can identify data patterns that change over time and has been attempted to predict groundwater level data. However, groundwater level data can cause sen-sor errors, missing values, or outliers, and these problems can degrade the performance of the LSTM model, and there is a need to improve data quality by processing them in the pretreatment stage. Therefore, in pre-dicting groundwater data, we will compare the LSTM model with the MSE and the model after normaliza-tion through distribution, and discuss the important process of analysis and data preprocessing according to the comparison results and changes in the results.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.12-14
• /
• 2021

A sensor registry system (SRS) provides semantic metadata about a sensor based on location information of a mobile device in order to solve a problem of interoperability between a sensor and a device. However, if the GPS of the mobile device is incorrectly received, the SRS receives incorrect sensor information and has a problem in that it cannot connect with the sensor. This paper proposes a dual collaboration strategy based on geographical embedding and LSTM-based path prediction to improve the probability of successful requests between mobile devices and sensors to address this problem and evaluate with the Monte Carlo approach. Through experiments, it was shown that the proposed method can compensate for location abnormalities and is an effective multicasting mechanism.

• International Journal of Computer Science & Network Security
• /
• v.23 no.8
• /
• pp.177-189
• /
• 2023

Malware detection is an increasingly important operational focus in cyber security, particularly given the fast pace of such threats (e.g., new malware variants introduced every day). There has been great interest in exploring the use of machine learning techniques in automating and enhancing the effectiveness of malware detection and analysis. In this paper, we present a deep recurrent neural network solution as a stacked Long Short-Term Memory (LSTM) with a pre-training as a regularization method to avoid random network initialization. In our proposal, we use global and short dependencies of the inputs. With pre-training, we avoid random initialization and are able to improve the accuracy and robustness of malware threat hunting. The proposed method speeds up the convergence (in comparison to stacked LSTM) by reducing the length of malware OpCode or bytecode sequences. Hence, the complexity of our final method is reduced. This leads to better accuracy, higher Mattews Correlation Coefficients (MCC), and Area Under the Curve (AUC) in comparison to a standard LSTM with similar detection time. Our proposed method can be applied in real-time malware threat hunting, particularly for safety critical systems such as eHealth or Internet of Military of Things where poor convergence of the model could lead to catastrophic consequences. We evaluate the effectiveness of our proposed method on Windows, Ransomware, Internet of Things (IoT), and Android malware datasets using both static and dynamic analysis. For the IoT malware detection, we also present a comparative summary of the performance on an IoT-specific dataset of our proposed method and the standard stacked LSTM method. More specifically, of our proposed method achieves an accuracy of 99.1% in detecting IoT malware samples, with AUC of 0.985, and MCC of 0.95; thus, outperforming standard LSTM based methods in these key metrics.

• Convergence Security Journal
• /
• v.23 no.5
• /
• pp.65-72
• /
• 2023

This paper proposes a method for detecting malicious domains considering human habitual characteristics by building a Deep Learning model based on LSTM (Long Short-Term Memory). DGA (Domain Generation Algorithm) malicious domains exploit human habitual errors, resulting in severe security threats. The objective is to swiftly and accurately respond to changes in malicious domains and their evasion techniques through typosquatting to minimize security threats. The LSTM-based Deep Learning model automatically analyzes and categorizes generated domains as malicious or benign based on malware-specific features. As a result of evaluating the model's performance based on ROC curve and AUC accuracy, it demonstrated 99.21% superior detection accuracy. Not only can this model detect malicious domains in real-time, but it also holds potential applications across various cyber security domains. This paper proposes and explores a novel approach aimed at safeguarding users and fostering a secure cyber environment against cyber attacks.

• Journal of Digital Convergence
• /
• v.17 no.12
• /
• pp.243-248
• /
• 2019

Multi-Task Learning(MTL) is a training method that trains a single neural network with multiple tasks influences each other. In this paper, we compare performance of MTL Named entity recognition(NER) model trained with Korean traditional culture corpus and other NER model. In training process, each Bi-LSTM layer of Part of speech tagging(POS-tagging) and NER are propagated from a Bi-LSTM layer to obtain the joint loss. As a result, the MTL based Bi-LSTM model shows 1.1%~4.6% performance improvement compared to single Bi-LSTM models.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.11
• /
• pp.13-20
• /
• 2023

In this paper, we propose a water temperature prediction method using feature extraction and reconstructed data based on LSTM-Autoencoder. We used multivariate time series data such as sea surface water temperature in the Naksan area of the East Sea where the cold water zone phenomenon occurred, and wind direction and wind speed that affect water temperature. Using the LSTM-Autoencoder model, we used three types of data: feature data extracted through dimensionality reduction of the original data combined with multivariate data of the original data, reconstructed data, and original data. The three types of data were trained by the LSTM model to predict sea surface water temperature and evaluated the accuracy. As a result, the sea surface water temperature prediction accuracy using feature extraction of LSTM-Autoencoder confirmed the best performance with MAE 0.3652, RMSE 0.5604, MAPE 3.309%. The result of this study are expected to be able to prevent damage from natural disasters by improving the prediction accuracy of sea surface temperature changes rapidly such as the cold water zone.

• Journal of Korea Water Resources Association
• /
• v.51 no.6
• /
• pp.503-514
• /
• 2018

In recent, the hydrological regime of the Mekong river is changing drastically due to climate change and haphazard watershed development including dam construction. Information of hydrologic feature like streamflow of the Mekong river are required for water disaster prevention and sustainable water resources development in the river sharing countries. In this study, runoff simulations at the Kratie station of the lower Mekong river are performed using SWAT (Soil and Water Assessment Tool), a physics-based hydrologic model, and LSTM (Long Short-Term Memory), a data-driven deep learning algorithm. The SWAT model was set up based on globally-available database (topography: HydroSHED, landuse: GLCF-MODIS, soil: FAO-Soil map, rainfall: APHRODITE, etc) and then simulated daily discharge from 2003 to 2007. The LSTM was built using deep learning open-source library TensorFlow and the deep-layer neural networks of the LSTM were trained based merely on daily water level data of 10 upper stations of the Kratie during two periods: 2000~2002 and 2008~2014. Then, LSTM simulated daily discharge for 2003~2007 as in SWAT model. The simulation results show that Nash-Sutcliffe Efficiency (NSE) of each model were calculated at 0.9(SWAT) and 0.99(LSTM), respectively. In order to simply simulate hydrological time series of ungauged large watersheds, data-driven model like the LSTM method is more applicable than the physics-based hydrological model having complexity due to various database pressure because it is able to memorize the preceding time series sequences and reflect them to prediction.

• Journal of Advanced Information Technology and Convergence
• /
• v.10 no.2
• /
• pp.167-175
• /
• 2020

In this paper, we propose a traffic accident detection on vehicle-mounted camera. In the proposed method, the minimum bounding box coordinates the central coordinates on the bird's eye view and motion vectors of each vehicle object, and ego-motions of the vehicle equipped with dash-cam are extracted from the dash-cam video. By using extracted 4 kinds features as the input of Bi-LSTM (bidirectional LSTM), the accident probability (score) is predicted. To investigate the effect of each input feature on the probability of an accident, we analyze the performance of the detection the case of using a single feature input and the case of using a combination of features as input, respectively. And in these two cases, different detection models are defined and used. Bi-LSTM is used as a cascade, especially when a combination of the features is used as input. The proposed method shows 76.1% precision and 75.6% recall, which is superior to our previous work.

• Journal of Software Assessment and Valuation
• /
• v.17 no.2
• /
• pp.39-46
• /
• 2021

To analyze and defend malware codes, reverse engineering is used as identify function location information. However, the stripped binary is not easy to find information such as function location because function symbol information is removed. To solve this problem, there are various binary analysis tools such as BAP and BitBlaze IDA Pro, but they are based on heuristics method, so they do not perform well in general. In this paper, we propose a technique to extract function information using LSTM-based models by applying algorithms of N-byte method that is extracted binaries corresponding to reverse assembling instruments in a recursive descent method. Through experiments, the proposed techniques were superior to the existing techniques in terms of time and accuracy.