• The Transactions of the Korea Information Processing Society
• /
• v.3 no.3
• /
• pp.429-438
• /
• 1996

An incremental learning algorithm is presented that constructs a multilayer perceptron whose size is optimal for solving a given problem. Unlike conventional algorithms in which a fixed size training set is processed repeat-edly, the method uses an increasing number of critical examples to find a necessary and sufficient number of hidden units for learning the entire data. Experimental results in hand- writtern digit recognition shows that the network size optimization combined with incremental pattern selection generalizes significantly better and converges faster than conventional methods.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2000.11a
• /
• pp.281-286
• /
• 2000

This paper describes a lateral guidance system of an unmanned vehicle, using a neural network model of magneto-resistive sensor and magnetic fields. The model equation was compared with experimental sensing data. We found that the experimental result has a negligible difference from the modeling equation result. We verified that the modeling equation can be used in the unmanned vehicle simulations. As the neural network controller acquires magnetic field values(B$\_$x/, B$\_$y/, B$\_$z/) from the three-axis, the controller outputs a steering angle. The controller uses the back-propagation algorithms of neural network. The learning pattern acquisition was obtained using computer simulation, which is more exact than human driving. The simulation program was developed in order to verify the acquisition of the learning pattern, learning itself, and the adequacy of the design controller. A computer simulation of the vehicle (including vehicle dynamics and steering) was used to verify the steering performance of the vehicle controller using the neural network. Good results were obtained. Also, the real unmanned electrical vehicle using neural network controller verified good results.

• The Journal of the Acoustical Society of Korea
• /
• v.14 no.1
• /
• pp.13-21
• /
• 1995

In this paper, we improve an image classifier algorithm based on neural network learning. It consists of two steps. The first is input pattern generation and the second, the global neural network implementation using an improved back-propagation algorithm. The feature vector for pattern recognition consists of the codebook data obtained from self-organization feature map learning. It decreases the input neuron number as well as the computational cost. The global neural network algorithm which is used in classifier inserts a control part and an address memory part to the back-propagation algorithm to control weights and unit-offsets. The simulation results show that it does not fall into the local minima and can implement easily the large-scale neural network. And it decreases largely the learning time.

• The Transactions of the Korean Institute of Electrical Engineers
• /
• v.40 no.2
• /
• pp.224-233
• /
• 1991

The multilayer expanson of single layer NN (Neural Network) was needed to solve the linear seperability problem as shown by the classic example using the XOR function. The EBP (Error Back Propagation ) learning rule is often used in multilayer Neural Networks, but it is not without its faults: 1)D.Rimmelhart expanded the Delta Rule but there is a problem in obtaining Ca from the linear combination of the Weight matrix N between the hidden layer and the output layer and H, wich is the result of another linear combination between the input pattern and the Weight matrix M between the input layer and the hidden layer. 2) Even if using the difference between Ca and Da to adjust the values of the Weight matrix N between the hidden layer and the output layer may be valid is correct, but using the same value to adjust the Weight matrixd M between the input layer and the hidden layer is wrong. Recognitron III was proposed to solve these faults. According to simulation results, since Recognitron III does not learn the three layer NN itself, but divides it into several single layer NNs and learns these with learning patterns, the learning time is 32.5 to 72.2 time faster than EBP NN one. The number of patterns learned in a EBP NN with n input and output cells and n+1 hidden cells are 2**n, but n in Recognitron III of the same size. [5] In the case of pattern generalization, however, EBP NN is less than Recognitron III.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.19 no.39
• /
• pp.47-61
• /
• 1996

We propose a parallel neural network model in which patterns are clustered and patterns in a cluster are studied in a parallel neural network. The learning algorithm used in this paper is based on LVQ algorithm of Kohonen(1990) for clustering and ADALINE(Adaptive Linear Neuron) network of Widrow and Hoff(1990) for parallel learning. The proposed algorithm consists of two parts. First, N patterns to be learned are categorized into C clusters by LVQ clustering algorithm. Second, C patterns that was selected from each cluster of C are learned as input pattern of ADALINE(Adaptive Linear Neuron). Data used in this paper consists of 250 patterns of ASCII characters normalized into $8\times16$ and 1124. The proposed algorithm consists of two parts. First, N patterns to be learned are categorized into C clusters by LVQ clustering algorithm. Second, C patterns that was selected from each cluster of C are learned as input pattern of ADALINE(Adaptive Linear Neuron). Data used in this paper consists 250 patterns of ASCII characters normalized into $8\times16$ and 1124 samples acquired from signals generated from 9 car models that passed Inductive Loop Detector(ILD) at 10 points. In ASCII character experiment, 191(179) out of 250 patterns are recognized with 3%(5%) noise and with 1124 car model data. 807 car models were recognized showing 71.8% recognition ratio. This result is 10.2% improvement over backpropagation algorithm.

• International Journal of Computer Science & Network Security
• /
• v.21 no.9
• /
• pp.31-40
• /
• 2021

The increasing number of credit card fraud cases has become a considerable problem since the past decades. This phenomenon is due to the expansion of new technologies, including the increased popularity and volume of online banking transactions and e-commerce. In order to address the problem of credit card fraud detection, a rule-based approach has been widely utilized to detect and guard against fraudulent activities. However, it requires huge computational power and high complexity in defining and building the rule base for pattern matching, in order to precisely identifying the fraud patterns. In addition, it does not come with intelligence and ability in predicting or analysing transaction data in looking for new fraud patterns and strategies. As such, Data Mining and Machine Learning algorithms are proposed to overcome the shortcomings in this paper. The aim of this paper is to highlight the important techniques and methodologies that are employed in fraud detection, while at the same time focusing on the existing literature. Methods such as Artificial Neural Networks (ANNs), Support Vector Machines (SVMs), naïve Bayesian, k-Nearest Neighbour (k-NN), Decision Tree and Frequent Pattern Mining algorithms are reviewed and evaluated for their performance in detecting fraudulent transaction.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.15 no.2
• /
• pp.13-22
• /
• 2005

To improve the anomaly IDS using system calls, this study focuses on Neural Networks Learning using the Soundex algorithm which is designed to change feature selection and variable length data into a fixed length learning pattern. That is, by changing variable length sequential system call data into a fixed length behavior pattern using the Soundex algorithm, this study conducted neural networks learning by using a backpropagation algorithm with fuzzy membership function. The back-propagation neural networks and Neuro-Fuzzy technique are applied for anomaly intrusion detection of system calls using Sendmail Data of UNM to demonstrate its aspect of he complexity of time, space and MDL performance.

• Smart Structures and Systems
• /
• v.29 no.1
• /
• pp.105-116
• /
• 2022

Stay cables play an essential role in cable-stayed bridges. Severe vibrations and/or harsh environment may result in cable failures. Therefore, an efficient structural health monitoring (SHM) solution for cable damage detection is necessary. This study proposes a data-driven method for immediately detecting cable damage from measured cable forces by recognizing pattern transition from the intact condition when damage occurs. In the proposed method, pattern recognition for cable damage detection is realized by time series classification (TSC) using a deep learning (DL) model, namely, the long short term memory fully convolutional network (LSTM-FCN). First, a TSC classifier is trained and validated using the cable forces (or cable force ratios) collected from intact stay cables, setting the segmented data series as input and the cable (or cable pair) ID as class labels. Subsequently, the classifier is tested using the data collected under possible damaged conditions. Finally, the cable or cable pair corresponding to the least classification accuracy is recommended as the most probable damaged cable or cable pair. A case study using measured cable forces from an in-service cable-stayed bridge shows that the cable with damage can be correctly identified using the proposed DL-TSC method. Compared with existing cable damage detection methods in the literature, the DL-TSC method requires minor data preprocessing and feature engineering and thus enables fast and convenient early detection in real applications.

• Journal of Korea Multimedia Society
• /
• v.21 no.2
• /
• pp.211-223
• /
• 2018

To understand the meaning of data is a common goal of research on unstructured data. Among these unstructured data, there are difficulties in analyzing the meaning of unstructured data related to corpus and sentences. In the existing researches, the researchers used LSA to select sentences with the most similar meaning to specific words of the sentences. However, it is problematic to examine many sentences continuously. In order to solve unstructured data classification problem, several search sites are available to classify the frequency of words and to serve to users. In this paper, we propose a method of classifying documents by using the frequency of similar words, and the frequency of non-relevant words to be applied as weights, and storing them in terms of a multi-pattern storage. We use Tensorflow's Softmax to the nearby sentences for machine learning, and utilize it for unstructured data analysis and the inference of traffic information.

• Archives of design research
• /
• v.15 no.4
• /
• pp.337-346
• /
• 2002

Among many design processes such as learning, storing, retrieving and applying, the process that learns design knowledge is very important for producing creative results that solve design purposes in design computations. The computer should have the ability similar to human in learning design knowledge. It should recognize not only physical properties but also high level design knowledge constructed from the first level physical properties. The high level design knowledge are recognised in terms of isometric translation relationships. This paper explains properties of isometric translation and methods how the computer can recognize high level shape design knowledge using shape pattern representation.