• Journal of Broadcast Engineering
• /
• v.23 no.4
• /
• pp.511-518
• /
• 2018

This paper deals with improving speech intelligibility by applying binary mask to time-frequency units of speech in noise. The binary mask is set to "0" or "1" according to whether speech is dominant or noise is dominant by comparing signal-to-noise ratio with pre-defined threshold. Bayesian classifier trained with Gaussian mixture model is used to estimate the binary mask of each time-frequency signal. The binary mask based noise suppressor improves speech intelligibility only in noise condition which is included in the training data. In this paper, speaker adaptation techniques for speech recognition are applied to adapt the Gaussian mixture model to a new noise environment. Experiments with noise-corrupted speech are conducted to demonstrate the improvement of speech intelligibility by employing adaption techniques in a new noise environment.

• KIPS Transactions on Software and Data Engineering
• /
• v.6 no.9
• /
• pp.445-456
• /
• 2017

In this paper, we propose a method to detect spam tweets containing unhealthy information by using an n-gram dictionary under limited labeling. Spam tweets that contain unhealthy information have a tendency to use similar words and sentences. Based on this characteristic, we show that spam tweets can be effectively detected by applying a Naive Bayesian classifier using n-gram dictionaries which are constructed from spam tweets and normal tweets. On the other hand, constructing an initial training set requires very high cost because a large amount of data flows in real time in a twitter. Therefore, there is a need for a spam detection method that can be applied in an environment where the initial training set is very small or non exist. To solve the problem, we propose a method to generate pseudo-labels by utilizing twitter's retweet function and use them for the configuration of the initial training set and the n-gram dictionary update. The results from various experiments using 1.3 million korean tweets collected from December 1, 2016 to December 7, 2016 prove that the proposed method has superior performance than the compared spam detection methods.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.9 no.3
• /
• pp.73-84
• /
• 2010

This study presents a novel method to identify hazardous segments of freeway using global positioning system(GPS) based probe vehicle data. A variety of candidate contributing factors leading to higher potential of accident occurrence were extracted from the probe vehicle dataset. The research problem was defined as a classification problem, then a well-known classifier, bayesian neural network was adopted to solve the problem. A binary logistic regression technique was also used for selecting salient input variables. Test results showed that the proposed method is promising in extracting hazardous freeway sections. The outcome of this study will be effectively used for evaluating the safety of freeway sections and deriving countermeasures to prevent accidents.

• Journal of IKEEE
• /
• v.23 no.4
• /
• pp.1393-1399
• /
• 2019

With the rapid growth of the Internet, millions of indiscriminate advertising SMS are sent every day because of the convenience of sending and receiving data. Although we still use methods to block spam words manually, we have been actively researching how to filter spam in a various ways as machine learning emerged. However, spam words and patterns are constantly changing to avoid being filtered, so existing machine learning mechanisms cannot detect or adapt to new words and patterns. Recently, the concept of Lifelong Learning emerged to overcome these limitations, using existing knowledge to keep learning new knowledge continuously. In this paper, we propose a method of spam filtering system using ensemble techniques of naive bayesian which is most commonly used in document classification and LLML(Lifelong Machine Learning). We validate the performance of lifelong learning by applying the model ELLA and the Naive Bayes most commonly used in existing spam filters.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.21 no.4
• /
• pp.512-517
• /
• 2011

In this paper, we propose an Active Shape Model (ASM) and Lucas-Kanade (LK) optical flow-based feature extraction and analysis method for analyzing the emotional features from facial images. Considering the facial emotion feature regions are described by Facial Action Coding System, we construct the feature-related shape models based on the combination of landmarks and extract the LK optical flow vectors at each landmarks based on the centre pixels of motion vector window. The facial emotion features are modelled by the combination of the optical flow vectors and the emotional states of facial image can be estimated by the probabilistic estimation technique, such as Bayesian classifier. Also, we extract the optimal emotional features that are considered the high correlation between feature points and emotional states by using common spatial pattern (CSP) analysis in order to improvise the operational efficiency and accuracy of emotional feature extraction process.

• Journal of KIISE:Databases
• /
• v.33 no.2
• /
• pp.163-174
• /
• 2006

In distributed wireless sensor network, it is difficult to transmit and analyze the entire stream data depending on limited networks, power and processor. Therefore it is suitable to use alternative stream data processing after classifying the continuous stream data. We propose a classification framework for continuous multivariate stream data. The proposed approach works in two steps. In the preprocessing step, it takes input as a sliding window of multivariate stream data and discretizes the data in the window into a string of symbols that characterize the signal changes. In the classification step, it uses a standard text classification algorithm to classify the discretized data in the window. We evaluated both supervised and unsupervised classification algorithms. For supervised, we tested Bayesian classifier and SVM, and for unsupervised, we tested Jaccard, TFIDF Jaro and Jaro Winkler. In our experiments, SVM and TFIDF outperformed other classification methods. In particular, we observed that classification accuracy is improved when the correlation of attributes is also considered along with the n-gram tokens of symbols.

• Journal of Physiology & Pathology in Korean Medicine
• /
• v.20 no.1
• /
• pp.224-228
• /
• 2006

The pattern identification of exterior-interior syndrome and cold-heat syndrome is one of the diagnostic methods using most frequently in Oriental medicine. There was no systematic studies analyzing the characteristics of the 'exterior-interior and cold-heat' between healthy and disease group. In this study, cold-heat pattern, blood pressure, pulse rate, height and weight are recorded from 100 healthy subjects and 196 disease subjects with age ranging from 30 to 59 years. To analyze the differences between healthy and disease group, we used the descriptive statistics. And linear regression function, linear support vector machine and bayesian classifier were used for distinguishing healthy group from disease group. The score of both exterior-heat and interior-cold in healthy group is higher than the score in disease group. This means that if one belongs to the disease group, his(or her) exterior gets cold and his interior gets hot. And also, these result have no relevance to age. But, the attempt to classify healthy group from disease group with a exterior-interior and cold-heat and other vital signs did not have good performance. It mean that even though they have a different trend each other, only these kinds of information couldn't classify healthy group and disease group.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.8
• /
• pp.3984-4005
• /
• 2017

Image recognition has become an increasingly important topic for its wide application. It is highly challenging when facing to large-scale database with large variance. The recognition systems rely on a key component, i.e. the low-level feature or the learned mid-level feature. The recognition performance can be potentially improved if the data distribution information is exploited using a more sophisticated way, which usually a function over hidden variable, model parameter and observed data. These methods are called generative score space. In this paper, we propose a discriminative extension for the existing generative score space methods, which exploits class label when deriving score functions for image recognition task. Specifically, we first extend the regular generative models to class conditional models over both observed variable and class label. Then, we derive the mid-level feature mapping from the extended models. At last, the derived feature mapping is embedded into a discriminative classifier for image recognition. The advantages of our proposed approach are two folds. First, the resulted methods take simple and intuitive forms which are weighted versions of existing methods, benefitting from the Bayesian inference of class label. Second, the probabilistic generative modeling allows us to exploit hidden information and is well adapt to data distribution. To validate the effectiveness of the proposed method, we cooperate our discriminative extension with three generative models for image recognition task. The experimental results validate the effectiveness of our proposed approach.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.20 no.4
• /
• pp.63-73
• /
• 2010

Malicious bot programs, the source of distributed denial of service attack, are widespread and the number of PCs which were infected by malicious bot program are increasing geometrically thesedays. The continuous distributed denial of service attacks are happened constantly through these bot PCs and some financial incident cases have found lately. Therefore researches to response distributed denial of service attack are necessary so we propose an effective feature generation method for distributed denial of service attack detection using entropy. In this paper, we apply our method to both the DARPA 2000 datasets and also the distributed denial of service attack datasets that we composed and generated ourself in general university. And then we evaluate how the proposed method is useful through classification using bayesian network classifier.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.193-193
• /
• 2023

여러 기후변화 시나리오에 의하면 기상재해의 발생빈도 및 강도가 증가할 것으로 예상된다. 그중 가뭄은 강수량 부족, 하천유량 감소, 토양 함수량 감소, 용수 수요량 증가 등의 다양한 요인으로 인해 발생하며, 한 가지 형태뿐만 아니라 복합적인 형태로 발생할 수 있다. 또한, 우리나라는 지역마다 기후 특성의 편차가 있어 기후변화에 따른 가뭄 취약성과 대응 능력이 지역마다 다르게 나타난다. 따라서 가뭄에 대응하기 위해서는 다양한 요인을 고려한 통합가뭄지수를 활용해야 하며, 미래의 기후변화를 고려하여 종합적으로 가뭄을 평가해야 한다. 본 연구에서는 동적 베이지안 분류기(DNBC) 기반의 통합가뭄지수를 활용하여 우리나라 전국에 대해 수문학적 위험도를 분석하고 미래 가뭄을 전망하였다. 기상학적, 수문학적, 농업적 및 사회경제적 요인을 고려한 통합가뭄지수를 산정하기 위하여 DNBC 분류기의 인자로 기후변화 시나리오 기반의 기상학적 가뭄지수 SPI, 수문학적 가뭄지수 SDI, 농업적 가뭄지수 ESI와 사회경제적 가뭄지수 WSCI를 활용하였다. 산정된 통합가뭄지수의 시계열을 기반으로 심도와 지속기간을 추출하고, 코플라 함수를 활용한 이변량 가뭄빈도분석을 수행하였다. 이후, 이변량 가뭄빈도분석에 의해 산정된 재현기간을 활용하여 수문학적 위험도를 산정하였다. 그 결과, P1(2021~2040) 기간이 수문학적 위험도 R=0.588로 가장 높은 위험도를 나타냈으며, 이후 P2(2041~2070) 기간까지 감소하였다가 P3(2071~2099) 기간에 다시 증가하는 추세를 보였다. P1(2021~2040) 기간과 P3(2071~2099) 기간은 영산강 유역이 각각 R=0.625(P1), R=0.550(P3)으로 가장 높은 위험도를 나타냈으나, P2(2041~2070) 기간은 금강 유역이 수문학적 위험도 R=0.482로 가장 높게 나타났다. 본 연구결과를 통해 향후 미래 가뭄에 대한 가뭄계획 수립 시에 기초자료로서 활용성이 높을 것으로 기대된다.