• ETRI Journal
• /
• v.45 no.4
• /
• pp.713-723
• /
• 2023

Game bots are illegal programs that facilitate account growth and goods acquisition through continuous and automatic play. Early detection is required to minimize the damage caused by evolving game bots. In this study, we propose a game bot detection method based on action time intervals (ATIs). We observe the actions of the bots in a game and identify the most frequently occurring actions. We extract the frequency, ATI average, and ATI standard deviation for each identified action, which is to used as machine learning features. Furthermore, we measure the performance using actual logs of the Aion game to verify the validity of the proposed method. The accuracy and precision of the proposed method are 97% and 100%, respectively. Results show that the game bots can be detected early because the proposed method performs well using only data from a single day, which shows similar performance with those proposed in a previous study using the same dataset. The detection performance of the model is maintained even after 2 months of training without any revision process.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2001.01a
• /
• pp.59-64
• /
• 2001

The robot soccer simulation game is a dynamic multi-agent environment. In this paper we suggest a new reinforcement learning approach to each agent's dynamic positioning in such dynamic environment. Reinforcement learning is the machine learning in which an agent learns from indirect, delayed reward an optimal policy to chose sequences of actions that produce the greatest cumulative reward. Therefore the reinforcement learning is different from supervised learning in the sense that there is no presentation of input pairs as training examples. Furthermore, model-free reinforcement learning algorithms like Q-learning do not require defining or learning any models of the surrounding environment. Nevertheless it can learn the optimal policy if the agent can visit every state- action pair infinitely. However, the biggest problem of monolithic reinforcement learning is that its straightforward applications do not successfully scale up to more complex environments due to the intractable large space of states. In order to address this problem. we suggest Adaptive Mediation-based Modular Q-Learning (AMMQL)as an improvement of the existing Modular Q-Learning (MQL). While simple modular Q-learning combines the results from each learning module in a fixed way, AMMQL combines them in a more flexible way by assigning different weight to each module according to its contribution to rewards. Therefore in addition to resolving the problem of large state effectively, AMMQL can show higher adaptability to environmental changes than pure MQL. This paper introduces the concept of AMMQL and presents details of its application into dynamic positioning of robot soccer agents.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.8 no.10
• /
• pp.3378-3393
• /
• 2014

We have witnessed the rapid development of information technology in recent years. One of the key phenomena is the fast, near-exponential increase of data. Consequently, most of the traditional data classification methods fail to meet the dynamic and real-time demands of today's data processing and analyzing needs--especially for continuous data streams. This paper proposes an improved incremental learning algorithm for a large-scale data stream, which is based on SVM (Support Vector Machine) and is named DS-IILS. The DS-IILS takes the load condition of the entire system and the node performance into consideration to improve efficiency. The threshold of the distance to the optimal separating hyperplane is given in the DS-IILS algorithm. The samples of the history sample set and the incremental sample set that are within the scope of the threshold are all reserved. These reserved samples are treated as the training sample set. To design a more accurate classifier, the effects of the data volumes of the history sample set and the incremental sample set are handled by weighted processing. Finally, the algorithm is implemented in a cloud computing system and is applied to study user behaviors. The results of the experiment are provided and compared with other incremental learning algorithms. The results show that the DS-IILS can improve training efficiency and guarantee relatively high classification accuracy at the same time, which is consistent with the theoretical analysis.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.18 no.3
• /
• pp.151-157
• /
• 2018

Most previous studies of software fault prediction have focused on supervised learning models for binary classification that determines whether an input module has faults or not. However, binary classification model determines only the presence or absence of faults in the module without considering the complex characteristics of the fault, and supervised model has the limitation that it requires a training data set that most development groups do not have. To solve these two problems, this paper proposes severity-based ternary classification model using unsupervised learning algorithms, and experimental results show that the proposed model has comparable performance to the supervised models.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.15 no.4
• /
• pp.217-231
• /
• 2015

A novel fuzzy learning framework that employs fuzzy inference to solve the problem of Multiple Instance Learning (MIL) is presented. The framework introduces a new class of fuzzy inference systems called Multiple Instance Mamdani Fuzzy Inference Systems (MI-Mamdani). In multiple instance problems, the training data is ambiguously labeled. Instances are grouped into bags, labels of bags are known but not those of individual instances. MIL deals with learning a classifier at the bag level. Over the years, many solutions to this problem have been proposed. However, no MIL formulation employing fuzzy inference exists in the literature. Fuzzy logic is powerful at modeling knowledge uncertainty and measurements imprecision. It is one of the best frameworks to model vagueness. However, in addition to uncertainty and imprecision, there is a third vagueness concept that fuzzy logic does not address quiet well, yet. This vagueness concept is due to the ambiguity that arises when the data have multiple forms of expression, this is the case for multiple instance problems. In this paper, we introduce multiple instance fuzzy logic that enables fuzzy reasoning with bags of instances. Accordingly, a MI-Mamdani that extends the standard Mamdani inference system to compute with multiple instances is introduced. The proposed framework is tested and validated using a synthetic dataset suitable for MIL problems. Additionally, we apply the proposed multiple instance inference to fuse the output of multiple discrimination algorithms for the purpose of landmine detection using Ground Penetrating Radar.

• ETRI Journal
• /
• v.28 no.1
• /
• pp.59-66
• /
• 2006

Learning behaviors of a radial basis function network (RBFN) using a singular value decomposition (SVD) and stochastic gradient (SG) algorithm, together named RBF-SVD-SG, for odor sensing systems are analyzed, and a fast training method is proposed. RBF input data is from a conducting polymer sensor array. It is revealed in this paper that the SG algorithm for the fine-tuning of centers and widths still shows ill-behaving learning results when a sufficiently small convergence coefficient is not used. Since the tuning of centers in RBFN plays a dominant role in the performance of RBFN odor sensing systems, our analysis is focused on the center-gradient variance of the RBFN-SVD-SG algorithm. We found analytically that the steadystate weight fluctuation and large values of a convergence coefficient can lead to an increase in variance of the center-gradient estimate. Based on this analysis, we propose to use the least mean square algorithm instead of SVD in adjusting the weight for stable steady-state weight behavior. Experimental results of the proposed algorithm have shown faster learning speed and better classification performance.

• Computers and Concrete
• /
• v.20 no.6
• /
• pp.627-634
• /
• 2017

Predicting the compressive strength of concrete is important to assess the load-carrying capacity of a structure. However, the use of blended cements to accrue the technical, economic and environmental benefits has increased the complexity of prediction models. Artificial Neural Networks (ANNs) have been used for predicting the compressive strength of ordinary Portland cement concrete, i.e., concrete produced without the addition of supplementary cementing materials. In this study, models to predict the compressive strength of blended cement concrete prepared with a natural pozzolan were developed using regression models and single- and 2-phase learning ANNs. Back-propagation (BP), Levenberg-Marquardt (LM) and Conjugate Gradient Descent (CGD) methods were used for training the ANNs. A 2-phase learning algorithm is proposed for the first time in this study for predictive modeling of the compressive strength of blended cement concrete. The output of these predictive models indicates that the use of a 2-phase learning algorithm will provide better results than the linear regression model or the traditional single-phase ANN models.

• Smart Structures and Systems
• /
• v.32 no.1
• /
• pp.37-47
• /
• 2023

Indirect determination of pan evaporation (PE) has been highly regarded, due to the advantages of intelligent models employed for this objective. This work pursues improving the reliability of a popular intelligent model, namely multi-layer perceptron (MLP) through surmounting its computational knots. Available climatic data of Fresno weather station (California, USA) is used for this study. In the first step, testing several most common trainers of the MLP revealed the superiority of the Levenberg-Marquardt (LM) algorithm. It, therefore, is considered as the classical training approach. Next, the optimum configurations of two metaheuristic algorithms, namely cuttlefish optimization algorithm (CFOA) and teaching-learning-based optimization (TLBO) are incorporated to optimally train the MLP. In these two models, the LM is replaced with metaheuristic strategies. Overall, the results demonstrated the high competency of the MLP (correlations above 0.997) in the presence of all three strategies. It was also observed that the TLBO enhances the learning and prediction accuracy of the classical MLP (by nearly 7.7% and 9.2%, respectively), while the CFOA performed weaker than LM. Moreover, a comparison between the efficiency of the used metaheuristic optimizers showed that the TLBO is a more time-effective technique for predicting the PE. Hence, it can serve as a promising approach for indirect PE analysis.

• Korean Journal of Remote Sensing
• /
• v.40 no.2
• /
• pp.167-177
• /
• 2024

With advancements in satellite technology, interest in target detection and identification is increasing quantitatively and qualitatively. Synthetic Aperture Radar(SAR) images, which can be acquired regardless of weather conditions, have been applied to various areas combined with machine learning based detection algorithms. However, conventional studies primarily focused on the detection of stationary targets. In this study, we proposed a method to identify moving targets using an algorithm that integrates sub-aperture SAR images and cosine similarity calculations. Utilizing a transformer-based deep learning target detection model, we extracted the bounding box of each target, designated the area as a region of interest (ROI), estimated the similarity between sub-aperture SAR images, and determined movement based on a predefined similarity threshold. Through the proposed algorithm, the quantitative evaluation of target identification capability enhanced its accuracy compared to when training with the targets with two different classes. It signified the effectiveness of our approach in maintaining accuracy while reliably discerning whether a target is in motion.

• Advances in materials Research
• /
• v.11 no.1
• /
• pp.75-90
• /
• 2022

In this paper, functionality of soft computing algorithms such as Group method of data handling (GMDH), Random forest (RF), Random tree (RT), Linear regression (LR), M5P, and artificial neural network (ANN) have been looked out to predict the compressive strength of concrete mixed with marble powder. Assessment of result suggests that, the overall performance of ANN based model gives preferable results over the different applied algorithms for the estimate of compressive strength of concrete. The results of coefficient of correlation were maximum in ANN model (0.9139) accompanied through RT with coefficient of correlation (CC) value 0.8241 and minimum root mean square error (RMSE) value of ANN (4.5611) followed by RT with RMSE (5.4246). Similarly, other evaluating parameters like, Willmott's index and Nash-sutcliffe coefficient value of ANN was 0.9458 and 0.7502 followed by RT model (0.8763 and 0.6628). The end result showed that, for both subsets i.e., training and testing subset, ANN has the potential to estimate the compressive strength of concrete. Also, the results of sensitivity suggest that the water-cement ratio has a massive impact in estimating the compressive strength of concrete with marble powder with ANN based model in evaluation with the different parameters for this data set.