• Journal of Korea Multimedia Society
• /
• v.22 no.11
• /
• pp.1251-1258
• /
• 2019

Intelligent tutoring system enables users to effectively learn by utilizing various artificial intelligence techniques. For instance, it can recommend a proper curriculum or learning method to individual users based on their learning history. To do this effectively, user's characteristics need to be analyzed and classified based on various aspects such as interest, learning ability, and personality. Even though data labeled by the characteristics are required for more accurate classification, it is not easy to acquire enough amount of labeled data due to the labeling cost. On the other hand, unlabeled data should not need labeling process to make a large number of unlabeled data be collected and utilized. In this paper, we propose a semi-supervised learning method based on feedback variational auto-encoder(FVAE), which uses both labeled data and unlabeled data. FVAE is a variation of variational auto-encoder(VAE), where a multi-layer perceptron is added for giving feedback. Using unlabeled data, we train FVAE and fetch the encoder of FVAE. And then, we extract features from labeled data by using the encoder and train classifiers with the extracted features. In the experiments, we proved that FVAE-based semi-supervised learning was superior to VAE-based method in terms with accuracy and F1 score.

• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.2237-2242
• /
• 2005

The original RRT is iteratively expanded by applying control inputs that drive the system slightly toward randomly-selected states, as opposed to requiring point-to-point convergence, as in the probabilistic roadmap approach. It is generally known that the performance of RRTs can be improved depending on the selection of the metrics in choosing the nearest vertex and bias techniques in choosing random states. We designed a path planning algorithm based on the RRT method for a remote-controlled mobile robot. First, we considered a bias technique that is goal-biased Gaussian random distribution along the command directions. Secondly, we selected the metric based on a weighted Euclidean distance of random states and a weighted distance from the goal region. It can save the effort to explore the unnecessary regions and help the mobile robot to find a feasible trajectory as fast as possible. Finally, the constraints of the actuator should be considered to apply the algorithm to physical mobile robots, so we select control inputs distributed with commanded inputs and constrained by the maximum rate of input change instead of random inputs. Simulation results demonstrate that the proposed algorithm is significantly more efficient for planning than a basic RRT planner. It reduces the computational time needed to find a feasible trajectory and can be practically implemented in a remote-controlled mobile robot.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.4
• /
• pp.1931-1953
• /
• 2017

The reverse skyline query plays an important role in information searching applications. This paper deals with continuous reverse skyline queries in sensor networks, which retrieves reverse skylines as well as the set of nodes that reported them for continuous sampling epochs. Designing an energy-efficient approach to answer continuous reverse skyline queries is non-trivial because the reverse skyline query is not decomposable and a huge number of unqualified nodes need to report their sensor readings. In this paper, we develop a new algorithm that avoids transmission of updates from nodes that cannot influence the reverse skyline. We propose a data mapping scheme to estimate sensor readings and determine their dominance relationships without having to know the true values. We also theoretically analyze the properties for reverse skyline computation, and propose efficient pruning techniques while guaranteeing the correctness of the answer. An extensive experimental evaluation demonstrates the efficiency of our approach.

• ETRI Journal
• /
• v.20 no.2
• /
• pp.115-132
• /
• 1998

This paper proposes an extended service filtering technique to prevent overload in service control point (SCP) due to televoting (VOT) or mass calling (MAS) services with the heavy traffic characteristics. Also, this paper compares this extended technique with the existing overload control techniques, and calculates steady state call blocking probabilities in intelligent network (IN) under overload conditions. The proposed technique considers SCP overload and IN Capability Set (CS)-1 services (such as VOT or MAS service) that have to use the specialized resources of intelligent peripheral (IP). This technique uses first an activating step in which SCP requests service filtering to service switching point (SSP). Then, in the filtering step, SSP sends filtering results to SCP periodically or each N-calls. Also, when filtering time-out expires, SSP stops service filtering, and sends service filtering response to SCP in the deactivating step. This paper applies this technique to VOT/MAS service, and calculates SCP and SSP-IP (circuit) call blocking probabilities by using an analytical VOT/MAS service model. With the modeling and analyzing of this new technique, it shows that this technique reduces the traffic flow into SCP from SSP and IP prominently.

• The KIPS Transactions:PartB
• /
• v.12B no.3 s.99
• /
• pp.329-336
• /
• 2005

This paper proposes a method to provide intelligence for characters in fighting action games by using genetic algorithm. The proposed characters without any knowledge on the rules of the game learn the rules and techniques for generations, and have the capability of evolving. To evaluate adaptability for varying circumstances, we changed the rules and compared the results. The experimental results show that the intelligent characters can adapt to the new rules. An advantage of the proposed method is that it can be easily applied to characters for other category of games such as PC games and internet online games.

• Journal of KIISE
• /
• v.45 no.1
• /
• pp.1-8
• /
• 2018

Recently, distribution of malicious codes using the Internet has been one of the most serious cyber threats. Technology of malicious code distribution with detection bypass techniques has been also developing and the research has focused on how to detect and analyze them. However, obfuscated malicious JavaScript is almost impossible to detect, because the existing malicious code distributed web page detection system is based on signature and another limitation is that it requires constant updates of the detection patterns. We propose to overcome these limitations by means of an intelligent malicious code distributed web page detection system using a real browser that can analyze and detect intelligent malicious code distributed web sites effectively.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.3 no.2
• /
• pp.194-199
• /
• 2003

Data are an expression of the language or numerical values that show some features. And the information is extracted from data for the specific purposes. The knowledge is utilized as information to construct rules that recognize patterns or make a decision. Today, knowledge extraction and application of that are broadly accomplished for the easy comprehension and the performance improvement of systems in the several industrial fields. The knowledge extraction can be achieved by some steps that include the knowledge acquisition, expression, and implementation. Such extracted knowledge is drawn by rules with data mining techniques. Clustering (CL), input space partition (ISP), neuro-fuzzy (NF), neural network (NN), extension matrix (EM), etc. are employed for the knowledge expression based upon rules. In this paper, the various approaches of the knowledge extraction are surveyed and categorized by methodologies and applied industrial fields. Also, the trend and examples of each approaches are shown in the tables and graphes using the categories such as CL, ISP, NF, NN, EM, and so on.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.9 no.1
• /
• pp.10-15
• /
• 2009

With the development of techniques, robots are getting smaller, and the number of robots needed for application is greater and greater. How to coordinate large number of autonomous robots through local interactions has becoming an important research issue in robot community. Swarm Robot Systems (SRS) is a system that independent autonomous robots in the restricted environments infer their status from pre-assigned conditions and operate their jobs through the cooperation with each other. In the SRS, a robot contains sensor part to percept the situation around them, communication part to exchange information, and actuator part to do a work. Especially, in order to cooperate with other robots, communicating with other robots is one of the essential elements. Because Bluetooth has many advantages such as low power consumption, small size module package, and various standard protocols, it is rated as one of the efficient communicating technologies which can apply to small-sized robot system. In this paper, we will develop Bluetooth communicating system for autonomous robots. And we will discuss how to construct and what kind of procedure to develop the communicating system for group behavior of the SRS under intelligent space.

• Smart Structures and Systems
• /
• v.7 no.3
• /
• pp.229-240
• /
• 2011

Recent developments in Smart Structures with very large scale embedded sensors and actuators have introduced new challenges in terms of data processing and sensor fusion. These smart structures are dynamically classified as a large-scale system with thousands of sensors and actuators that form the musculoskeletal of the structure, analogous to human body. In order to develop structural health monitoring and diagnostics with data provided by thousands of sensors, new sensor informatics has to be developed. The focus of our on-going research is to develop techniques and algorithms that would utilize this musculoskeletal system effectively; thus creating the intelligence for such a large-scale autonomous structure. To achieve this level of intelligence, three major research tasks are being conducted: development of a Bio-Inspired data analysis and information extraction from thousands of sensors; development of an analytical technique for Optimal Sensory System using Structural Observability; and creation of a bio-inspired decision-making and control system. This paper is focused on the results of our effort on the first task, namely development of a Neuro-Morphic Engineering approach, using a neuro-symbolic data manipulation, inspired by the understanding of human information processing architecture, for sensor fusion and structural diagnostics.

• Proceedings of the KSR Conference
• /
• 2008.06a
• /
• pp.653-658
• /
• 2008

Service areas in the urban transit need to construct the intelligent integrated surveillance system, because they are the public places that many people get together at one time. In past, analogue, closed-circuit televisions and analogue video recorders are used to construct the surveillance system. Now, a lot parts of the analogue systems that depend on the images have been changed to the complicated system, which consists of sensors and images and also, to be digitalized. In past, the surveillance system was used as an inspection devices to examine the spots after happening some events. But, with a high level of the computer and communication technologies, it is possible that the digitalized data leads the intelligence systems to prevent some accidents by using the various analysis techniques. And the data could be used to decide surveillance policies and provide some information on the safety and management policies as well as surveillance policies. In this paper, we define the intelligent surveillance system and suggest the major functions of the system. Also, we suggest the fundamental functions that every part should get and describe the way to develop the system.