• Journal of Intelligence and Information Systems
• /
• 제18권4호
• /
• pp.79-95
• /
• 2012

As credit loan products significantly increase in most financial institutions, the number of fraudulent transactions is also growing rapidly. Therefore, to manage the financial risks successfully, the financial institutions should reinforce the qualifications for a loan and augment the ability to detect a credit loan fraud proactively. In the process of building a classification model to detect credit loan frauds, utility from classification results (i.e., benefits from correct prediction and costs from incorrect prediction) is more important than the accuracy rate of classification. The objective of this paper is to propose a new approach to building a classification model for detecting credit loan fraud based on an individual-level utility. Experimental results show that the model comes up with higher utility than the fraud detection models which do not take into account the individual-level utility concept. Also, it is shown that the individual-level utility computed by the model is more accurate than the mean-level utility computed by other models, in both opportunity utility and cash flow perspectives. We provide diverse views on the experimental results from both perspectives.

• Journal of Internet Technology
• /
• 제22권2호
• /
• pp.423-439
• /
• 2021

Android is one of the most popular platforms for the mobile and Internet of Things (IoT) devices. This popularity has made Android-based devices a valuable target of malicious apps. Thus, it is essential to devise automatic and portable malware detection approaches for the Android platform. There are many studies on detecting mobile malware using machine learning techniques. In these studies, however, the dataset is imbalanced or is not large enough to generalize the machine learning model, or the dimensionality of features is too high to apply nonlinear classifiers. In this article, we propose a machine learning-based Android malware detection scheme that uses API calls and permissions as features. To restrict the dimensionality of features, we propose minimal domain knowledge-based and Gini importance-based feature selection. We construct large and balanced real-world datasets to build a generalized and non-skewed model and verify our model through experiments. We achieve 96.51% classification accuracy using Random Forest classifier with low overhead. In addition, we also provide an analysis on falsely classified samples in detail. The analysis results show that API hiding can degrade the performance of API call information-based malware detection systems.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• 제23권12호
• /
• pp.1542-1550
• /
• 2019

Legacy studies for classifying arrhythmia have been studied to improve the accuracy of classification, Neural Network, Fuzzy, etc. Deep learning is most frequently used for arrhythmia classification using error backpropagation algorithm by solving the limit of hidden layer number, which is a problem of neural network. In order to apply a deep learning model to an ECG signal, it is necessary to select an optimal model and parameters. In this paper, we propose optimal parameter extraction method based on a deep learning. For this purpose, R-wave is detected in the ECG signal from which noise has been removed, QRS and RR interval segment is modelled. And then, the weights were learned by supervised learning method through deep learning and the model was evaluated by the verification data. The detection and classification rate of R wave and PVC is evaluated through MIT-BIH arrhythmia database. The performance results indicate the average of 99.77% in R wave detection and 97.84% in PVC classification.

• Database Research
• /
• 제34권3호
• /
• pp.45-57
• /
• 2018

The drowsiness that occurs in the driving is a very dangerous driver condition that can be directly linked to a major accident. In order to prevent drowsiness, there are traditional drowsiness detection methods to grasp the driver's condition, but there is a limit to the generalized driver's condition recognition that reflects the individual characteristics of drivers. In recent years, deep learning based state recognition studies have been proposed to recognize drivers' condition. Deep learning has the advantage of extracting features from a non-human machine and deriving a more generalized recognition model. In this study, we propose a more accurate state recognition model than the existing deep learning method by learning image and PPG at the same time to grasp driver's condition. This paper confirms the effect of driver's image and PPG data on drowsiness detection and experiment to see if it improves the performance of learning model when used together. We confirmed the accuracy improvement of around 3% when using image and PPG together than using image alone. In addition, the multimodal deep learning based model that classifies the driver's condition into three categories showed a classification accuracy of 96%.

• The Journal of the Korea institute of electronic communication sciences
• /
• 제15권3호
• /
• pp.513-520
• /
• 2020

Recently, the development of a vision inspection system using machine learning has become more active. This study seeks to develop a defect inspection model using machine learning. Defect detection problems for images correspond to classification problems, which are the method of supervised learning in machine learning. In this study, defect detection models are developed based on algorithms that automatically extract features and algorithms that do not extract features. One-dimensional CNN and two-dimensional CNN are used as algorithms for automatic extraction of features, and MLP and SVM are used as algorithms for non-extracting features. A defect detection model is developed based on four models and their accuracy and AUC compare based on AUC. Although image classification is common in the development of models using CNN, high accuracy and AUC is achieved when developing SVM models by converting pixels from images into RGB values in this study.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권11호
• /
• pp.5568-5587
• /
• 2018

This research uses artificial intelligence methods for computer network intrusion detection system modeling. Primary classification is done using self-organized maps (SOM) in two levels, while the secondary classification of ambiguous data is done using Sugeno type Fuzzy Inference System (FIS). FIS is created by using Adaptive Neuro-Fuzzy Inference System (ANFIS). The main challenge for this system was to successfully detect attacks that are either unknown or that are represented by very small percentage of samples in training dataset. Improved algorithm for SOMs in second layer and for the FIS creation is developed for this purpose. Number of clusters in the second SOM layer is optimized by using our improved algorithm to minimize amount of ambiguous data forwarded to FIS. FIS is created using ANFIS that was built on ambiguous training dataset clustered by another SOM (which size is determined dynamically). Proposed hybrid model is created and tested using NSL KDD dataset. For our research, NSL KDD is especially interesting in terms of class distribution (overlapping). Objectives of this research were: to successfully detect intrusions represented in data with small percentage of the total traffic during early detection stages, to successfully deal with overlapping data (separate ambiguous data), to maximize detection rate (DR) and minimize false alarm rate (FAR). Proposed hybrid model with test data achieved acceptable DR value 0.8883 and FAR value 0.2415. The objectives were successfully achieved as it is presented (compared with the similar researches on NSL KDD dataset). Proposed model can be used not only in further research related to this domain, but also in other research areas.

• The Journal of Korea Robotics Society
• /
• 제15권2호
• /
• pp.169-176
• /
• 2020

This paper studied the collision detection of robot manipulators for safe collaboration in human-robot interaction. Based on sensor-based collision detection, external torque is detached from subtracting robot dynamics. To detect collision using joint torque sensor data, a comparative study was conducted using data-based machine learning algorithm. Data was collected from the actual 3 degree-of-freedom (DOF) robot manipulator, and the data was labeled by threshold and handwork. Using support vector machine (SVM), decision tree and k-nearest neighbors KNN method, we derive the optimal parameters of each algorithm and compare the collision classification performance. The simulation results are analyzed for each method, and we confirmed that by an optimal collision status detection model with high prediction accuracy.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권2호
• /
• pp.632-657
• /
• 2022

Detecting web attacks is a major challenge, and it is observed that the use of simple models leads to low sensitivity or high false positive problems. In this study, we aim to develop a robust two-stage deep learning based stacked ensemble web application firewall. Normal and abnormal classification is carried out in the first stage of the proposed WAF model. The classification process of the types of abnormal traffics is postponed to the second stage and carried out using an integrated stacked ensemble model. By this way, clients' requests can be served without time delay, and attack types can be detected with high sensitivity. In addition to the high accuracy of the proposed model, by using the statistical similarity and diversity analyses in the study, high generalization for the ensemble model is achieved. Within the study, a comprehensive, up-to-date, and robust multi-class web anomaly dataset named GAZI-HTTP is created in accordance with the real-world situations. The performance of the proposed WAF model is compared to state-of-the-art deep learning models and previous studies using the benchmark dataset. The proposed two-stage model achieved multi-class detection rates of 97.43% and 94.77% for GAZI-HTTP and ECML-PKDD, respectively.

• KIPS Transactions on Software and Data Engineering
• /
• 제4권11호
• /
• pp.521-528
• /
• 2015

Road-view object classification methods are mostly influenced by weather and illumination conditions, thus the most of the research activities are based on dataset in clean weathers. In this paper, we present a road-view object classification method based on color segmentation that works for all kinds of weathers. The proposed method first classifies the weather and illumination conditions and then applies the weather-specified color models to find the road traffic signs. Using 5 different features of the road-view images, we classify the weather and light conditions as sunny, cloudy, rainy, night, and backlight. Based on the classified weather and illuminations, our model selects the weather-specific color ranges to generate Gaussian Mixture Model for each colors, Green, Yellow, and Blue. The proposed method successfully detects the traffic signs regardless of the weather and illumination conditions.

• The Journal of Korea Robotics Society
• /
• 제17권4호
• /
• pp.425-430
• /
• 2022

As robots are considered one of the mainstream digital transformations, robots with machine vision becomes a main area of study providing the ability to check what robots watch and make decisions based on it. However, it is difficult to find a small object in the image mainly due to the flaw of the most of visual recognition networks. Because visual recognition networks are mostly convolution neural network which usually consider local features. So, we make a model considering not only local feature, but also global feature. In this paper, we propose a detection method of a small marker on the object using deep learning and an algorithm that considers global features by combining Transformer's self-attention technique with a convolutional neural network. We suggest a self-attention model with new definition of Query, Key and Value for model to learn global feature and simplified equation by getting rid of position vector and classification token which cause the model to be heavy and slow. Finally, we show that our model achieves higher mAP than state of the art model YOLOr.