• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.3
• /
• pp.361-367
• /
• 2021

Information search in video streaming services such as YouTube is replacing traditional information search services. To find desired detailed information in such a video, users should repeatedly navigate several points in the video, resulting in a waste of time and network traffic. In this paper, we propose a method to assist users in searching for information in a video by using DBSCAN clustering and LSTM. Our LSTM model is trained with a dataset that consists of user search sequences and their final target points categorized by DBSCAN clustering algorithm. Then, our proposed method utilizes the trained model to suggest an expected category for the user's desired target point based on a partial search sequence that can be collected at the beginning of the search. Our experiment results show that the proposed method successfully finds user destination points with 98% accuracy and 7s of the time difference by average.

• Journal of Korea Game Society
• /
• v.21 no.4
• /
• pp.3-12
• /
• 2021

In this study, models for predicting the final result of League of Legends game were constructed for each rank using data from the first 10 minutes of the game. Variable importance was extracted from the prediction models to derive strategic direction in early phase of the game. As a result, it was possible to predict final results with over 70% accuracy in all ranks. It was found that early game advantage tends to lead to the final win and this tendency appeared stronger as it goes to challenger ranks. Kill(death) was found to be the most influential factor for win, however, there were also variables whose importance rank changed according to rank. This indicates there is a difference in the strategic direction in the early stage of the game depending on the rank.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.1
• /
• pp.64-69
• /
• 2022

Commercialization of public technology is the transfer of government-led scientific and technological innovation and R&D results to the private sector, and is recognized as a key achievement driving economic growth. Therefore, in order to activate technology transfer, various machine learning methods are being studied to identify success factors or to match public technology with high commercialization potential and demanding companies. However, public technology commercialization data is in the form of a table and has a problem that machine learning performance is not high because it is in an imbalanced state with a large difference in success-failure ratio. In this paper, we present a method of utilizing CTGAN to resolve imbalances in public technology data in tabular form. In addition, to verify the effectiveness of the proposed method, a comparative experiment with SMOTE, a statistical approach, was performed using actual public technology commercialization data. In many experimental cases, it was confirmed that CTGAN reliably predicts public technology commercialization success cases.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.1
• /
• pp.70-75
• /
• 2022

Accurately measuring location is necessary to provide a variety of services. The data for indoor positioning measures the RSSI values from the WiFi device through an application of a smartphone. The measured data becomes the raw data of machine learning. The feature data is the measured RSSI value, and the label is the name of the space for the measured position. For this purpose, the machine learning technique is to study a technique that predicts the exact location only with the WiFi signal by applying an efficient technique to classification. Ensemble is a technique for obtaining more accurate predictions through various models than one model, including backing and boosting. Among them, Boosting is a technique for adjusting the weight of a model through a modeling result based on sampled data, and there are various algorithms. This study uses Xgboost among the above techniques and evaluates performance with other ensemble techniques.

• International Journal of Computer Science & Network Security
• /
• v.22 no.4
• /
• pp.193-202
• /
• 2022

The classification of hyperspectral imagery (HSI) is essential in the surface of earth observation. Due to the continuous large number of bands, HSI data provide rich information about the object of study; however, it suffers from the curse of dimensionality. Dimensionality reduction is an essential aspect of Machine learning classification. The algorithms based on feature extraction can overcome the data dimensionality issue, thereby allowing the classifiers to utilize comprehensive models to reduce computational costs. This paper assesses and compares two HSI classification techniques. The first is based on the Joint Spatial-Spectral Stacked Autoencoder (JSSSA) method, the second is based on a shallow Artificial Neural Network (SNN), and the third is used the SVM model. The performance of the JSSSA technique is better than the SNN classification technique based on the overall accuracy and Kappa coefficient values. We observed that the JSSSA based method surpasses the SNN technique with an overall accuracy of 96.13% and Kappa coefficient value of 0.95. SNN also achieved a good accuracy of 92.40% and a Kappa coefficient value of 0.90, and SVM achieved an accuracy of 82.87%. The current study suggests that both JSSSA and SNN based techniques prove to be efficient methods for hyperspectral classification of snow features. This work classified the labeled/ground-truth datasets of snow in multiple classes. The labeled/ground-truth data can be valuable for applying deep neural networks such as CNN, hybrid CNN, RNN for glaciology, and snow-related hazard applications.

• The Journal of the Korea Contents Association
• /
• v.22 no.4
• /
• pp.83-93
• /
• 2022

Research is underway to solve the traffic problem by using crowdsourcing, where drivers use their mobile devices to provide traffic information. If it is used for traffic event detection through crowdsourcing, the task of collecting related data is reduced, which lowers time cost and increases accuracy. In this paper, we propose a scheme to collect traffic-related data using crowdsourcing and to detect events affecting traffic through this. The proposed scheme uses machine learning algorithms for processing large amounts of data to determine the event type of the collected data. In addition, to find out the location where the event occurs, a keyword indicating the location is extracted from the collected data, and the administrative area of the keyword is returned. In this way, it is possible to resolve a location that is broadly defined in the existing location information or incorrect location information. Various performance evaluations are performed to prove the superiority and feasibility of the proposed scheme.

• The Journal of the Korea Contents Association
• /
• v.22 no.1
• /
• pp.58-73
• /
• 2022

The purpose of this study is to construct a predictable estimation model that reflects the relationship between the variables of mobile app performance and to verify how app reviews affect app performance. In study 1 and 2, the relationship between app performance indicators was derived using correlation analysis and random forest regression estimation of machine learning, and app performance estimation modeling was performed. In study 3, sentiment scores for app reviews were by using sentiment analysis of text mining, and it was found that app review sentiment scores have an effect one lag ahead of the number of daily installations of apps when using multivariate time series analysis. By analyzing the dissatisfaction and needs raised by app performance indicators and reviews of apps, companies can improve their apps in a timely manner and derive the timing and direction of marketing promotions.

• Nuclear Engineering and Technology
• /
• v.54 no.1
• /
• pp.36-48
• /
• 2022

As a valid numerical method to obtain a high-resolution result of a flow field, computational fluid dynamics (CFD) have been widely used to study coolant flow and heat transfer characteristics in fuel rod bundles. However, the time-consuming, iterative calculation of Navier-Stokes equations makes CFD unsuitable for the scenarios that require efficient simulation such as sensitivity analysis and uncertainty quantification. To solve this problem, a reduced-order model (ROM) based on proper orthogonal decomposition (POD) and machine learning (ML) is proposed to simulate the flow field efficiently. Firstly, a validated CFD model to output the flow field data set of the rod bundle is established. Secondly, based on the POD method, the modes and corresponding coefficients of the flow field were extracted. Then, an deep feed-forward neural network, due to its efficiency in approximating arbitrary functions and its ability to handle high-dimensional and strong nonlinear problems, is selected to build a model that maps the non-linear relationship between the mode coefficients and the boundary conditions. A trained surrogate model for modes coefficients prediction is obtained after a certain number of training iterations. Finally, the flow field is reconstructed by combining the product of the POD basis and coefficients. Based on the test dataset, an evaluation of the ROM is carried out. The evaluation results show that the proposed POD-ROM accurately describe the flow status of the fluid field in rod bundles with high resolution in only a few milliseconds.

• International Journal of Computer Science & Network Security
• /
• v.22 no.6
• /
• pp.390-399
• /
• 2022

Cyber security and resilience are phrases that describe safeguards of ICTs (information and communication technologies) from cyber-attacks or mitigations of cyber event impacts. The sole purpose of Risk models are detections, analyses, and handling by considering all relevant perceptions of risks. The current research effort has resulted in the development of a new paradigm for safeguarding services offered online which can be utilized by both service providers and users. customers. However, rather of relying on detailed studies, this approach emphasizes task selection and execution that leads to successful risk treatment outcomes. Modelling intelligent CSGs (Cyber Security Games) using MLTs (machine learning techniques) was the focus of this research. By limiting mission risk, CSGs maximize ability of systems to operate unhindered in cyber environments. The suggested framework's main components are the Threat and Risk models. These models are tailored to meet the special characteristics of online services as well as the cyberspace environment. A risk management procedure is included in the framework. Risk scores are computed by combining probabilities of successful attacks with findings of impact models that predict cyber catastrophe consequences. To assess successful attacks, models emulating defense against threats can be used in topologies. CSGs consider widespread interconnectivity of cyber systems which forces defending all multi-step attack paths. In contrast, attackers just need one of the paths to succeed. CSGs are game-theoretic methods for identifying defense measures and reducing risks for systems and probe for maximum cyber risks using game formulations (MiniMax). To detect the impacts, the attacker player creates an attack tree for each state of the game using a modified Extreme Gradient Boosting Decision Tree (that sees numerous compromises ahead). Based on the findings, the proposed model has a high level of security for the web sources used in the experiment.

• Journal of Creative Information Culture
• /
• v.7 no.4
• /
• pp.235-242
• /
• 2021

Recently, many human activity recognition(HAR) researches using smartphone sensor data have been studied. HAR can be utilized in various fields, such as life pattern analysis, exercise measurement, and dangerous situation detection. However researches have been focused on recognition of basic human behaviors or efficient battery use. In this paper, exercising activities performed indoors and outdoors were defined and recognized. Data collection and pre-processing is performed to recognize the defined activities by SVM, random forest and gradient boosting model. In addition, the recognition result is determined based on voting class approach for accuracy and stable performance. As a result, the proposed activities were recognized with high accuracy and in particular, similar types of indoor and outdoor exercising activities were correctly classified.