• Journal of the Korean Institute of Intelligent Systems
• /
• v.16 no.6
• /
• pp.736-741
• /
• 2006

The design of a classification system generally involves data acquisition module, learning module and decision module, considering their functions and it is often an important component of intelligent systems. The learning module provides a priori information and it has been playing a key role for the classification. The conventional learning techniques for classification are based on a winner take all fashion which does not reflect the description of real data where boundarues might be fuzzy Moreover they need all data for the learning of its problem domain. Generally, in many practical applications, it is not possible to prepare them at a time. In this paper, we design an adaptive classification model using incremental training fuzzy neural networks, FNN-I. To have a more useful information, it introduces the representation and membership degree by fuzzy theory. And it provides an incremental learning algorithm for continuously gathered data. We present tie experimental results on computer virus data. They show that the proposed system can learn incrementally and classify new viruses effectively.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.9
• /
• pp.1244-1253
• /
• 2021

In this paper, we propose a method for predicting in advance whether pedestrians will enter the hazardous area after the current time using the pedestrian trajectory prediction method and an efficient simplification method of the trajectory prediction network. In addition, we propose a method to apply KD(Knowledge Distillation) to a small network for real-time operation in an embedded environment. Using the correlation between predicted future paths and hazard zones, we determined whether to enter or not, and applied efficient KD when learning small networks to minimize performance degradation. Experimentally, it was confirmed that the model applied with the simplification method proposed improved the speed by 37.49% compared to the existing model, but led to a slight decrease in accuracy. As a result of learning a small network with an initial accuracy of 91.43% using KD, It was confirmed that it has improved accuracy of 94.76%.

• International Journal of Computer Science & Network Security
• /
• v.22 no.2
• /
• pp.159-166
• /
• 2022

Artificial Intelligence has stimulated every aspect of today's life. Human thinking quality is trying to be involved through digital tools in all research areas of the modern era. The education industry is also leveraging artificial intelligence magical power. Uses of digital technologies in pedagogical paradigms are being observed from the last century. The widespread involvement of artificial intelligence starts reshaping the educational landscape. Adaptive learning is an emerging pedagogical technique that uses computer-based algorithms, tools, and technologies for the learning process. These intelligent practices help at each learning curve stage, from content development to student's exam evaluation. The quality of information technology students and professionals training has also improved drastically with the involvement of artificial intelligence systems. In this paper, we will investigate adopted digital methods in the education sector so far. We will focus on intelligent techniques adopted for information technology students and professionals. Our literature review works on our proposed framework that entails four categories. These categories are communication between teacher and student, improved content design for computing course, evaluation of student's performance and intelligent agent. Our research will present the role of artificial intelligence in reshaping the educational process.

• Journal of Intelligence and Information Systems
• /
• v.28 no.2
• /
• pp.1-18
• /
• 2022

As digital technology converges into the e-commerce market across industries, online transactions have activated, and the use of online has increased. With the recent spread of infectious diseases such as COVID-19, this market flow is accelerating, and various product information can be provided to customers online. Providing a variety of information provides customers with various opportunities but causes difficulties in decision-making. The recommendation system can help customers to make a decision more effectively. However, the previous research on recommendation systems is limited to only quantitative data and does not reflect detailed factors of products and customers. In this study, we propose an intelligent recommendation system that quantifies the attributes of products and customers by applying text mining techniques to qualitative data based on online reviews and integrates the existing objective indicators of total star rating, sentiment, and emotion. The proposed integrated recommendation model showed superior performance to the overall rating-oriented recommendation model. It expects the new business value to be created through the recommendation result reflecting detailed factors of products and customers.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.17 no.5
• /
• pp.825-832
• /
• 2022

In this paper, we discuss the design and implementation of predictive and diagnostic models for realizing intelligent predictive models by collecting and storing the power output of agricultural photovoltaic power generation systems. Our model predicts the amount of photovoltaic power generation using RNN, LSTM, and GRU models, which are recurrent neural network techniques specialized for time series data, and compares and analyzes each model with different hyperparameters, and evaluates the performance. As a result, the MSE and RMSE indicators of all three models were very close to 0, and the R2 indicator showed performance close to 1. Through this, it can be seen that the proposed prediction model is a suitable model for predicting the amount of photovoltaic power generation, and using this prediction, it was shown that it can be utilized as an intelligent and efficient O&M function in an agricultural photovoltaic system.

• Journal of Positioning, Navigation, and Timing
• /
• v.12 no.3
• /
• pp.215-228
• /
• 2023

The State Space Representation (SSR) method provides individual corrections for each Global Navigation Satellite System (GNSS) error components. This method can lead to less bandwidth for transmission and allows selective use of each correction. Precise Point Positioning (PPP) - Real-Time Kinematic (RTK) is one of the carrier-based precise positioning techniques using SSR correction. This technique enables high-precision positioning with a fast convergence time by providing atmospheric correction as well as satellite orbit and clock correction. Currently, the positioning service that supports PPP-RTK technology is the Quazi-Zenith Satellite System Centimeter Level Augmentation System (QZSS CLAS) in Japan. A system that provides correction for each GNSS error component, such as QZSS CLAS, requires monitoring of each error component to provide reliable correction and integrity information to the user. In this study, we conducted an analysis of the performance of residual error bounding for each error component. To assess this performance, we utilized the correction and quality indicators provided by QZSS CLAS. Performance analyses included the range domain, dispersive part, non-dispersive part, and satellite orbit/clock part. The residual root mean square (RMS) of CLAS correction for the range domain approximated 0.0369 m, and the residual RMS for both dispersive and non-dispersive components is around 0.0363 m. It has also been confirmed that the residual errors are properly bounded by the integrity parameters. However, the satellite orbit and clock part have a larger residual of about 0.6508 m, and it was confirmed that this residual was not bounded by the integrity parameters. Users who rely solely on satellite orbit and clock correction, particularly maritime users, thus should exercise caution when utilizing QZSS CLAS.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.23 no.1
• /
• pp.182-196
• /
• 2024

This study introduces driving pattern analysis and change detection methods using smartphone sensors, based on contrastive learning. These methods characterize driving patterns without labeled data, allowing accurate classification with minimal labeling. In addition, they are robust to domain changes, such as different vehicle types. The study also examined the applicability of these methods to smartphones by comparing them with six lightweight deep-learning models. This comparison supported the development of smartphone-based driving pattern analysis and assistance systems, utilizing smartphone sensors and contrastive learning to enhance driving safety and efficiency while reducing the need for extensive labeled data. This research offers a promising avenue for addressing contemporary transportation challenges and advancing intelligent transportation systems.

• Journal of Advanced Navigation Technology
• /
• v.27 no.6
• /
• pp.770-775
• /
• 2023

Recently, as the amount of digital data increases exponentially, the importance of data processing systems is being emphasized. In this situation, the selection and construction of data processing systems are becoming more important. In this study, the performance of cloud computing (CC), edge computing (EC), and UAV-based intelligent edge computing (UEC) was compared as a way to solve this problem. The characteristics, strengths, and weaknesses of each method were analyzed. In particular, this study focused on real-time large-capacity data processing situations such as corridor monitoring. When conducting the experiment, a specific scenario was assumed and a penalty was given to the infrastructure. In this way, it was possible to evaluate performance in real situations more accurately. In addition, the effectiveness and limitations of each computing method were more clearly understood, and through this, the help was provided to enable more effective system selection.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.24 no.1
• /
• pp.1-8
• /
• 2024

On-device AI technology, which can operate AI models at the edge devices to support real-time processing and privacy enhancement, is attracting attention. As intelligent IoT is applied to various industries, services utilizing the on-device AI technology are increasing significantly. However, general deep learning models require a lot of computational resources for inference and learning. Therefore, various lightweighting methods such as quantization and pruning have been suggested to operate deep learning models in embedded edge devices. Among the lightweighting methods, we analyze how to lightweight and apply deep learning models to edge computing devices, focusing on pruning technology in this paper. In particular, we utilize dynamic and static pruning techniques to evaluate the inference speed, accuracy, and memory usage of a lightweight AI vision model. The content analyzed in this paper can be used for intelligent video control systems or video security systems in autonomous vehicles, where real-time processing are highly required. In addition, it is expected that the content can be used more effectively in various IoT services and industries.