• The Journal of the Convergence on Culture Technology
• /
• v.8 no.1
• /
• pp.545-550
• /
• 2022

In Deep Learning method, it is well known that it requires large amount of data to train the deep neural network. And it also requires the labeling of each data to fully train the neural network, which means that experts should spend lots of time to provide the labeling. To alleviate the problem of time-consuming labeling process, some methods have been suggested such as weak-supervised method, one-shot learning, self-supervised, suggestive learning, and so on. In this manuscript, those methods are analyzed and its possible future direction of the research is suggested.

• Journal of Korean Society for Geospatial Information Science
• /
• v.14 no.4 s.38
• /
• pp.71-77
• /
• 2006

Support Vector Machine (SVM) which has roots in a statistical learning theory is a training algorithm based on structural risk minimization. Generally, SVM algorithm uses the kernel for determining a linearly non-separable boundary and classifying the data. But, classical kernels can not apply to effectively the hyperspectral image classification because it measures similarity using vector's dot-product or euclidian distance. So, This paper proposes the spectral similarity kernel to solve this problem. The spectral similariy kernel that calculate both vector's euclidian and angle distance is a local kernel, it can effectively consider a reflectance property of hyperspectral image. For validating our algorithm, SVM which used polynomial kernel, RBF kernel and proposed kernel was applied to land cover classification in Hyperion image. It appears that SVM classifier using spectral similarity kernel has the most outstanding result in qualitative and spatial estimation.

• The Journal of the Acoustical Society of Korea
• /
• v.41 no.6
• /
• pp.662-667
• /
• 2022

Computing music similarity is an indispensable component in developing music search service. This paper focuses on learning a music similarity function in order to boost cover song identification performance. By using the probabilistic linear discriminant analysis, we construct a latent music space where the distances between cover song pairs reduces while the distances between the non-cover song pairs increases. We derive a music similarity function by testing hypothesis, whether two songs share the same latent variable or not, using the probabilistic models with the assumption that observed music features are generated from the learned latent music space. Experimental results performed on two cover music datasets show that the proposed music similarity improves the cover song identification performance.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.29B no.8
• /
• pp.1-6
• /
• 1992

This paper proposed a neural computer architecture for the learning of script character pattern recognition categories. Oriented filter with complex cells preprocess about the input script character, abstracts contour from the character. This contour normalized and inputed to the ART. Top-down attentional and matching mechanisms are critical in self-stabilizing of the code learning process. The architecture embodies a parallel search scheme that updates itself adaptively as the learning process unfolds. After learning ART self-stabilizes, recognition time does not grow as a function of code complexity. Vigilance level shows the similarity between learned patterns and new input patterns. This character recognition system is designed to adaptable. The simulation of this system showed satisfied result in the recognition of the hand written characters.

• Proceedings of the Korean Information Science Society Conference
• /
• 2011.06c
• /
• pp.168-171
• /
• 2011

One important issue in semantic web is identification and selection of domain concepts for domain ontology learning when several hundreds or even thousands of terms are extracted and available from relevant text documents shared among the members of a domain. We present a novel domain concept acquisition and selection approach for ontology learning that uses affinity propagation algorithm, which takes as input semantic and structural similarity between pairs of extracted terms called data points. Real-valued messages are passed between data points (terms) until high quality set of exemplars (concepts) and cluster iteratively emerges. All exemplars will be considered as domain concepts for learning domain ontologies. Our empirical results show that our approach achieves high precision and recall in selection of domain concepts using less number of iterations.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.12 no.5
• /
• pp.385-392
• /
• 2002

Collaborative filtering method for personalization can suggest new items and information which a user hasn t expected. But there are some problems. Not only the steps for calculating similarity value between each user is complex but also it doesn t reflect user s interest dynamically when a user input a query. In this paper, classifying users by their interest makes calculating similarity simple. We propose the a1gorithm for readjusting user s interest dynamically using the profile and Bayesian learning. When a user input a keyword searching for a item, his new interest is readjusted. And the user s profile that consists of used key words and the presence frequency of key words is designed and used to reflect the recent interest of users. Our methods of adjusting user s interest using the profile and Bayesian learning can improve the real satisfaction of users through the experiment with data set, collected in University s library. It recommends a user items which he would be interested in.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.29 no.4
• /
• pp.775-784
• /
• 2019

Malware infringement attacks are continuously increasing in various environments such as mobile, IOT, windows and mac due to the emergence of new and variant malware, and signature-based countermeasures have limitations in detection of malware. In addition, analytical performance is deteriorating due to obfuscation, packing, and anti-VM technique. In this paper, we propose a system that can detect malware based on machine learning by using similarity hashing-based pattern detection technique and static analysis after file classification according to packing. This enables more efficient detection because it utilizes both pattern-based detection, which is well-known malware detection, and machine learning-based detection technology, which is advantageous for detecting new and variant malware. The results of this study were obtained by detecting accuracy of 95.79% or more for benign sample files and malware sample files provided by the AI-based malware detection track of the Information Security R&D Data Challenge 2018 competition. In the future, it is expected that it will be possible to build a system that improves detection performance by applying a feature vector and a detection method to the characteristics of a packed file.

• Journal of IKEEE
• /
• v.23 no.1
• /
• pp.242-248
• /
• 2019

Fire flame and smoke detection algorithm studies are challenging task in computer vision due to the variety of shapes, rapid spread and colors. The performance of a typical sensor based fire detection system is largely limited by environmental factors (indoor and fire locations). To solve this problem, a deep learning method is applied. Because it extracts the feature of the object using several methods, so that if a similar shape exists in the frame, it can be detected as false postive. This study proposes a new algorithm to reduce false positives by using frame similarity before using deep learning to decrease the false detection rate. Experimental results show that the fire detection performance is maintained and the false positives are reduced by applying the proposed method. It is confirmed that the proposed method has excellent false detection performance.

• The Journal of the Convergence on Culture Technology
• /
• v.7 no.3
• /
• pp.591-598
• /
• 2021

It's possible to get the word-vector by the statistical SVD or deep-learning CBOW and LSTM methods and theses ones learn the contexts of forward/backward words or the sequence of following words. It's used to analyze the poems by Ki Hyung-do with similar words recommended by the word-vector showing the core images of the poetry. It seems at first sight that the words don't go well with the images but they express the similar style described by the reference words once you look close the contexts of the specific poems. The word-vector can analogize the words having the same relations with the ones between the representative words for the core images of the poems. Therefore you can analyze the poems in depth and in variety with the similarity and analogy operations by the word-vector estimated with the statistical SVD or deep-learning CBOW and LSTM methods.

• Journal of the Korea Convergence Society
• /
• v.11 no.8
• /
• pp.99-106
• /
• 2020

The purpose of this study is to find out whether image similarity between celebrities and brands on social network service be a determinant to predict advertising effectiveness. To this end, an advertising effect prediction model for celebrity endorsed advertising was created and its validity was verified through a machine learning method which is a big data analysis technique. Firstly, the celebrity-brand image similarity, which was used as an independent variable, was quantified by the association network theory with social big data, and secondly a multiple regression model which used data representing advertising effects as a dependent variable was repeatedly conducted to generate an advertising effect prediction model. The accuracy of the prediction model was decided by comparing the prediction results with the survey outcomes. As for a result, it was proved that the validity of the predictive modeling of advertising effects was secured since the classification accuracy of 75%, which is a criterion for judging validity, was shown. This study suggested a new methodological alternative and direction for big data-based modeling research through celebrity-brand image similarity structure based on social network theory, and effect prediction modeling by machine learning.