• Proceedings of the Korean Information Science Society Conference
• /
• 1999.10b
• /
• pp.57-59
• /
• 1999

본 논문에서는 최대 엔트로피 원리에 기반한 문서 분류기의 학습을 제안한다. 최대 엔트로피 기법은 자연언어 처리에서 언어 모델링(Language Modeling), 품사 태깅 (Part-of-Speech Tagging) 등에 널리 사용되는 방법중의 하나이다. 최대 엔트로피 모델의 효율성을 위해서는 자질 선정이 중요한데, 본 논문에서는 자질 집합의 선택을 위한 기준으로 chi-square test, log-likelihood ratio, information gain, mutual information 등의 방법을 이용하여 실험하고, 전체 후보 자질에 대한 실험 결과와 비교해 보았다. 데이터 집합으로는 Reuters-21578을 사용하였으며, 각 클래스에 대한 이진 분류 실험을 수행하였다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2006.10a
• /
• pp.22-27
• /
• 2006

압타머칩은 혈청(serum) 내의 지정된 단백질의 상대적 양을 직접 측정할 수 있는 바이오칩으로서, 의학적 질병 진단에 유용하게 사용할 수 있는 툴이다. 압타머칩 데이터 분석에는 기존의 마이크로어레이 분석기법을 그대로 적용할 수 있다. 본 논문에서는 Potential SVM(PSVM)을 이용하여, 심혈관질환 샘플 기반의 압타머칩 데이터에서 바이오마커 후보 단백질을 선정한 결과를 정리한다. PSVM은 분류 알고리즘으로서 뿐만 아니라 자질 선택(feature selection)에서도 우수한 성능을 보이는 알고리즘으로 알려져 있다. 심혈관 질환의 단계에 따라 구분한 4개 클래스, 135개 샘플로 구성된 3K 압타머칩 데이터에 대해 PSVM을 적용하여 자질을 선택하고 분류성능을 측정한 결과, 마이크로어레이에서의 자질 선택에 많이 사용되는 Gain Ratio 기법과 비교하여 보다 적은 수의 단백질 정보로 보다 나은 분류 성능을 보임을 확인하였다. 더불어, PSVM을 이용해 선택한 단백질군을 심혈관 질환 진단을 위한 바이오마커 후보로 제시한다.

• Journal of the Korean Society for information Management
• /
• v.36 no.2
• /
• pp.57-77
• /
• 2019

Random Forest (RF), a representative ensemble technique, was applied to automatic classification of journal articles in the field of library and information science. Especially, I performed various experiments on the main factors such as tree number, feature selection, and learning set size in terms of classification performance that automatically assigns class labels to domestic journals. Through this, I explored ways to optimize the performance of random forests (RF) for imbalanced datasets in real environments. Consequently, for the automatic classification of domestic journal articles, Random Forest (RF) can be expected to have the best classification performance when using tree number interval 100~1000(C), small feature set (10%) based on chi-square statistic (CHI), and most learning sets (9-10 years).

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.07b
• /
• pp.727-729
• /
• 2005

나이브베이즈분류기($na\ddot{i}ve$ Bayes classifier)는 학습, 적용 및 계산자원 이용의 측면에서 매우 효율적인 모델이다. 또한, 그 분류 성능 역시 다른 기법에 비해 크게 떨어지지 않음이 다양한 실험을 통해 보여져 왔다. 특히, 데이터를 생성한 실제 확률분포를 나이브베이즈분류기가 정확하게 표현할 수 있는 경우에는 최대의 효과를 볼 수 있다. 하지만, 실제 확률분포에 존재하는 조건부독립성(conditional independence)이 나이브베이즈분류기의 구조와 일치하지 않는 경우에는 성능이 하락할 수 있다. 보다 구체적으로, 각 자질변수(feature variable)들 사이에 확률적 의존관계(probabilistic dependency)가 존재하는 경우 성능 하락은 심화된다. 본 논문에서는 이러한 나이브베이즈분류기의 약점을 효율적으로 해결할 수 있는 자질변수의 통합기법을 제시한다. 자질변수의 통합은 각 변수들 사이의 관계를 명시적으로 표현해 주는 방법이며, 특히 상호정보량(mutual information)에 기반한 통합 변수의 선정이 성능 향상에 크게 기여함을 실험을 통해 보인다.

• Journal of the Korean Society for information Management
• /
• v.19 no.3
• /
• pp.5-30
• /
• 2002

In this paper, classification feature is selected with focus of musical content, note sequences pattern, and measures similarity between note sequences followed by constructing clusters by similar note sequences, which is easier for users to search by showing the similar note sequences with the search result in the CBMR system. Experimental document was $\ulcorner$A Dictionary of Musical Themes$\lrcorner$, the index of theme bar focused on classical music and obtained kern-type file. Humdrum Toolkit version 1.0 was used as note sequences treat tool. The hierarchical clustering method is by stages focused on four-type similarity matrices by whether the note sequences segmentation or not and where the starting point is. For the measurement of the result, WACS standard is used in the case of being manual classification and in the case of the note sequences starling from any point in the note sequences, there is used common feature pattern distribution in the cluster obtained from the clustering result. According to the result, clustering with segmented feature unconnected with the starting point Is higher with distinct difference compared with clustering with non-segmented feature.

• Journal of the Korean Society for information Management
• /
• v.40 no.3
• /
• pp.99-118
• /
• 2023

The objective of this study is to identify performance factors in the automatic classification of records by utilizing metadata that contains the contextual information of records. For this study, we collected 97,064 records of original textual information from Korean central administrative agencies in 2022. Various classification algorithms, data selection methods, and feature extraction techniques are applied and compared with the intent to discern the optimal performance-inducing technique. The study results demonstrated that among classification algorithms, Random Forest displayed higher performance, and among feature extraction techniques, the TF method proved to be the most effective. The minimum data quantity of unit tasks had a minimal influence on performance, and the addition of features positively affected performance, while their removal had a discernible negative impact.

• Annual Conference on Human and Language Technology
• /
• 2014.10a
• /
• pp.183-186
• /
• 2014

특허 분석에서 관심 있는 기술명, 서비스명, 제품명을 인식하도록 기계학습 기법을 사용해 개체명 인식기의 성능을 평가해 보았다. 개체인식을 위한 엔진은 스탠포드 대학의 NER과 CRF++을 사용하였다. 그 결과 F1값인 0.5612로 나타났다. 이것은 인명, 지역명, 조직명 개체를 인식하는 다른 연구에서 나타난 0.7857보다 0.2245 떨어지는 결과이다. 특허 개체명 인식에 대한 자질값 선정과 사전처리에 대한 더 많은 연구가 필요하다.

• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.4
• /
• pp.427-431
• /
• 2010

Multidimensional scaling (MDS) is a widely used method for dimensionality reduction, of which purpose is to represent high-dimensional data in a low-dimensional space while preserving distances among objects as much as possible. MDS has mainly been applied to data visualization and feature selection. Among various MDS methods, the classical MDS is not readily applicable to data which has large numbers of objects, on normal desktop computers due to its computational complexity. More precisely, it needs to solve eigenpair problems on dissimilarity matrices based on Euclidean distance. Thus, running time and required memory of the classical MDS highly increase as n (the number of objects) grows up, restricting its use in large-scale domains. In this paper, we propose an efficient approximation algorithm for the classical MDS based on divide-and-conquer and CUDA. Through a set of experiments, we show that our approach is highly efficient and effective for analysis and visualization of data consisting of several thousands of objects.

• Journal of Intelligence and Information Systems
• /
• v.21 no.2
• /
• pp.69-92
• /
• 2015

The explosion of social media data has led to apply text-mining techniques to analyze big social media data in a more rigorous manner. Even if social media text analysis algorithms were improved, previous approaches to social media text analysis have some limitations. In the field of sentiment analysis of social media written in Korean, there are two typical approaches. One is the linguistic approach using machine learning, which is the most common approach. Some studies have been conducted by adding grammatical factors to feature sets for training classification model. The other approach adopts the semantic analysis method to sentiment analysis, but this approach is mainly applied to English texts. To overcome these limitations, this study applies the Word2Vec algorithm which is an extension of the neural network algorithms to deal with more extensive semantic features that were underestimated in existing sentiment analysis. The result from adopting the Word2Vec algorithm is compared to the result from co-occurrence analysis to identify the difference between two approaches. The results show that the distribution related word extracted by Word2Vec algorithm in that the words represent some emotion about the keyword used are three times more than extracted by co-occurrence analysis. The reason of the difference between two results comes from Word2Vec's semantic features vectorization. Therefore, it is possible to say that Word2Vec algorithm is able to catch the hidden related words which have not been found in traditional analysis. In addition, Part Of Speech (POS) tagging for Korean is used to detect adjective as "emotional word" in Korean. In addition, the emotion words extracted from the text are converted into word vector by the Word2Vec algorithm to find related words. Among these related words, noun words are selected because each word of them would have causal relationship with "emotional word" in the sentence. The process of extracting these trigger factor of emotional word is named "Emotion Trigger" in this study. As a case study, the datasets used in the study are collected by searching using three keywords: professor, prosecutor, and doctor in that these keywords contain rich public emotion and opinion. Advanced data collecting was conducted to select secondary keywords for data gathering. The secondary keywords for each keyword used to gather the data to be used in actual analysis are followed: Professor (sexual assault, misappropriation of research money, recruitment irregularities, polifessor), Doctor (Shin hae-chul sky hospital, drinking and plastic surgery, rebate) Prosecutor (lewd behavior, sponsor). The size of the text data is about to 100,000(Professor: 25720, Doctor: 35110, Prosecutor: 43225) and the data are gathered from news, blog, and twitter to reflect various level of public emotion into text data analysis. As a visualization method, Gephi (http://gephi.github.io) was used and every program used in text processing and analysis are java coding. The contributions of this study are as follows: First, different approaches for sentiment analysis are integrated to overcome the limitations of existing approaches. Secondly, finding Emotion Trigger can detect the hidden connections to public emotion which existing method cannot detect. Finally, the approach used in this study could be generalized regardless of types of text data. The limitation of this study is that it is hard to say the word extracted by Emotion Trigger processing has significantly causal relationship with emotional word in a sentence. The future study will be conducted to clarify the causal relationship between emotional words and the words extracted by Emotion Trigger by comparing with the relationships manually tagged. Furthermore, the text data used in Emotion Trigger are twitter, so the data have a number of distinct features which we did not deal with in this study. These features will be considered in further study.