방송통신융합시대의 방송통신발전기금 부과원칙에 관한 연구

송민채 한국방송학회 2016 한국방송학회 학술대회 논문집 Vol.2016 No.04

한국어 자연어생성에 적합한 사전훈련 언어모델 특성 연구

송민채,신경식 한국지능정보시스템학회 2022 한국지능정보시스템학회 학술대회논문집 Vol.2022 No.11

감성어휘 임베딩과 어텐션 매커니즘에 기반한 LSTM을 이용한 감성분석

송민채,신경식 한국지능정보시스템학회 2018 한국지능정보시스템학회 학술대회논문집 Vol.2018 No.6

한국어 자연어생성에 적합한 사전훈련 언어모델 특성 연구

송민채,신경식 한국지능정보시스템학회 2022 지능정보연구 Vol.28 No.4

This study empirically analyzed a Korean pre-trained language models (PLMs) designed for natural language generation. The performance of two PLMs – BART and GPT – at the task of abstractive text summarization was compared. To investigate how performance depends on the characteristics of the inference data, ten different document types, containing six types of informational content and creation content, were considered. It was found that BART (which can both generate and understand natural language) performed better than GPT (which can only generate). Upon more detailed examination of the effect of inference data characteristics, the performance of GPT was found to be proportional to the length of the input text. However, even for the longest documents (with optimal GPT performance), BART still out-performed GPT, suggesting that the greatest influence on downstream performance is not the size of the training data or PLMs parameters but the structural suitability of the PLMs for the applied downstream task. The performance of different PLMs was also compared through analyzing parts of speech (POS) shares. BART’s performance was inversely related to the proportion of prefixes, adjectives, adverbs and verbs but positively related to that of nouns. This result emphasizes the importance of taking the inference data’s characteristics into account when fine-tuning a PLMs for its intended downstream task.

Customer sentiment classification

송민채,신경식 한국지능정보시스템학회 2016 한국지능정보시스템학회 학술대회논문집 Vol.2016 No.11

생강나무 꽃차 추출물의 항산화 및 LPS로 유도된 염증반응에 대한 Raw 264.7 대식세포 보호 효과

송민채,신승미,서원택,김현영,김지현 한국응용생명화학회 2024 Journal of Applied Biological Chemistry (J. Appl. Vol.67 No.-

We investigated the in vitro anti-oxidant activity and protective effects of Lindera obtusiloba flower tea extract (LFE) on lipopolysaccharide (LPS)-induced inflammatory response in Raw 264.7 macrophages. The polyphenol and flavonoid contents of LFE were 32.32 mg GAE/g and 213.83 mg QE/g, respectively. The LFE exhibited the highest 2,2-diphenyl-1-picrylhydrazyl and 2,2'-azino-bis (3-ethylbenzothiazoline-6-sulfonic acid) radical scavenging activities at a concentration of 100 μg/mL among stem and leaf of Lindera obtusiloba extracts. When treated with LFE at concentrations of 10-100 μg/mL, it dose-dependently reduced the production of nitric oxide (NO) in LPS-induced Raw 264.7 macrophages. The treatment of LFE significantly inhibited inflammatory cytokines such as interleukin (IL)-1β and IL-6 in Raw 264.7 macrophage induced by LPS. In particular, treatment of LFE showed a high suppression of IL-1β expression. To evaluate the anti-inflammatory mechanisms of LFE, we investigated protein expressions such as phospho-nuclear factor kappa-lightchain- enhancer of activated B cells (p-NF-B), inducible NO synthase (iNOS), and cyclooxygenase-2 (COX-2). The LFE inhibited the expression of p-NF-B, iNOS, and COX-2 compared with LPS-treated cells, resulting anti-inflammatory effects of LFE by inhibiting protein expression involved in NF-B signaling. Therefore, LFE could be a considered as a functional material with anti-oxidant and anti-inflammatory properties.

뉴스기사를 이용한 소비자의 경기심리지수 생성

송민채(Minchae Song),신경식(Kyung-shik Shin) 한국지능정보시스템학회 2017 지능정보연구 Vol.23 No.3

경제주체들의 경기상황에 대한 판단 및 전망은 경기변동에 영향을 미치므로 경기심리지수와 거시경제지표들 간에는 밀접한 관련성을 나타내는 것으로 알려져 있다. 경기선행지표로 국내에서 많이 사용되는 경기심리지수에는 소비자동향조사, 기업경기조사, 경제심리지수가 있다. 그러나 설문조사를 통해 생성된 지수는 자료의 성격상 속보성이 떨어지는 문제가 있다. 본 연구에서는 이러한 정형데이터의 한계를 보완할 수 있도록 비정형데이터에서 정보를 추출해 경기심리지수를 생성하고, 경제분석에서의 활용 가능성을 검토하였다. 민간소비와 관련된 실물지표에는 소매판매업지수와 서비스업생산지수를 사용하였고, 고용지표에는 고용률과 실업률을, 가격지표에는 소비자물가상승률과 가계의 대출금리를 사용하여 지표들 간의 추이 분석 및 시차구조 파악을 위한 교차상관분석을 수행하였다. 마지막으로 이들 지표들에 대한 예측 가능성을 점검하였다. 분석결과, 다른 지표들의 선행지수로 많이 사용되는 소비자심리지수와 비교해 선택 지표들과 높은 상관관계를 보이며, 1~2개월 선행한 것으로 나타났다. 예측력 또한 향상되어 텍스트데이터에서 생성한 소비자 경기심리지수의 유용성이 확인되었다. 온라인에서 생성되는 뉴스기사나 소셜 SNS 등의 텍스트 데이터는 속보성이 뛰어나고, 커버리지가 넓어 특정 경제적 이슈가 발생할 경우 이것이 경제에 미치는 영향을 빠르게 파악할 수 있다는 점에서 경기판단지표로써의 잠재적 가능성이 클 것으로 보인다. 경제분석에서 비정형데이터를 활용한 국내연구는 초기 단계지만 데이터의 유용성이 확인되면 그 활용도가 크게 높아질 것으로 기대한다. It is known that the economic sentiment index and macroeconomic indicators are closely related because economic agent’s judgment and forecast of the business conditions affect economic fluctuations. For this reason, consumer sentiment or confidence provides steady fodder for business and is treated as an important piece of economic information. In Korea, private consumption accounts and consumer sentiment index highly relevant for both, which is a very important economic indicator for evaluating and forecasting the domestic economic situation. However, despite offering relevant insights into private consumption and GDP, the traditional approach to measuring the consumer confidence based on the survey has several limits. One possible weakness is that it takes considerable time to research, collect, and aggregate the data. If certain urgent issues arise, timely information will not be announced until the end of each month. In addition, the survey only contains information derived from questionnaire items, which means it can be difficult to catch up to the direct effects of newly arising issues. The survey also faces potential declines in response rates and erroneous responses. Therefore, it is necessary to find a way to complement it. For this purpose, we construct and assess an index designed to measure consumer economic sentiment index using sentiment analysis. Unlike the survey-based measures, our index relies on textual analysis to extract sentiment from economic and financial news articles. In particular, text data such as news articles and SNS are timely and cover a wide range of issues; because such sources can quickly capture the economic impact of specific economic issues, they have great potential as economic indicators. There exist two main approaches to the automatic extraction of sentiment from a text, we apply the lexicon-based approach, using sentiment lexicon dictionaries of words annotated with the semantic orientations. In creating the sentiment lexicon dictionaries, we enter the semantic orientation of individual words manually, though we do not attempt a full linguistic analysis (one that involves analysis of word senses or argument structure); this is the limitation of our research and further work in that direction remains possible. In this study, we generate a time series index of economic sentiment in the news. The construction of the index consists of three broad steps: (1) Collecting a large corpus of economic news articles on the web, (2) Applying lexicon-based methods for sentiment analysis of each article to score the article in terms of sentiment orientation (positive, negative and neutral), and (3) Constructing an economic sentiment index of consumers by aggregating monthly time series for each sentiment word. In line with existing scholarly assessments of the relationship between the consumer confidence index and macroeconomic indicators, any new index should be assessed for its usefulness. We examine the new index’s usefulness by comparing other economic indicators to the CSI. To check the usefulness of the newly index based on sentiment analysis, trend and cross - correlation analysis are carried out to analyze the relations and lagged structure. Finally, we analyze the forecasting power using the one step ahead of out of sample prediction. As a result, the news sentiment index correlates strongly with related contemporaneous key indicators in almost all experiments. We also find that news sentiment shocks predict future economic activity in most cases. In almost all experiments, the news sentiment index strongly correlates with related contemporaneous key indicators. Furthermore, in most cases, news sentiment shocks predict future economic activity; in head-to-head comparisons, the news sentiment measures outperform survey-based sentiment index as CSI. Policy makers want to understand consumer or public opinions about existing or proposed policies. Such opinions enab

자연어처리기법을 활용한 코로나19 전후 지속가능한 글로벌 공급망 연구 동향 변화와 시사점

이소연(So-Yeon Lee),송민채(Min-Chae Song) 한국산학기술학회 2022 한국산학기술학회논문지 Vol.23 No.10

코로나19 사태로 기존 글로벌 공급망의 취약점이 드러나면서 위기 발생에 대한 선제적 리스크 관리 및 신속대응을 위한 국가와 기업들의 새로운 공급망 구축에 대한 논의가 활발히 진행 중이다. 그 중 ‘지속가능한 공급망(Sustainable Global Value Chains)’ 전략이 가장 큰 주목을 받고 있다. 그러나 어떠한 배경에서 코로나19가 글로벌 공급망 연구와 관련 논의에 영향을 미쳤는지를 실증 분석한 연구는 찾아보기 어렵다. 본 연구는 코로나19가 어떻게 글로벌 공급망 전략에 영향을 미쳤고, 지속가능한 글로벌 공급망 연구가 2019년 전후로 어떠한 양상으로 변화하고 있는지 텍스트 분석기법을 이용해 살펴보았다. 결론으로 분석에서 도출된 결과를 토대로 지속가능한 글로벌 공급망 연구의 영역 및 연구대상(주체)의 확대, 실행가능한 제도 정비의 필요성 등을 미래 연구의제로 제시하였다. 본 연구는 국가 간 경제의 상호의존성 심화로 코로나19 충격이 개별 기업의 대응 역량과 범위를 벗어나 경제 구조 전반을 변화시킴에 따라 글로벌 공급망의 패러다임 전환이 진행 중에 있음을 실증분석으로 확인하고, 어떠한 대응이 필요한지 제안하였다는 점에서 그 의의가 있다. The vulnerabilities of present global value chains were revealed by the COVID-19 pandemic. Discussions are underway in various countries and companies to establish new global value chains for risk management purposes and improve response to global risks, and the sustainable global value chain strategy has received the most research attention. Although research on sustainable global value chains has increased significantly since 2019, few studies have empirically analyzed how COVID-19 affected global value chain research and discussion. In this study, we analyzed how the COVID-19 pandemic affected global value chain strategies and why sustainable global value chain research changed before and after 2019 using natural language preprocessing techniques. Based on the implications derived from empirical results, we suggest sustainable global value chain research scope and research subjects be expanded and that the need for system preparation be explored. This study shows, using natural language preprocessing, that a paradigm shift in global value chains is in progress due to the COVID-19 pandemic.

한국 재정수지와 경상수지 간의 인과관계 검정

전주성(Joo-Sung Jun),송민채(Min-Chae Song) 한국무역학회 2013 貿易學會誌 Vol.38 No.5

CNN을 적용한 한국어 상품평 감성분석

박현정(Hyun-jung Park),송민채(Min-chae Song),신경식(Kyung-shik Shin) 한국지능정보시스템학회 2018 지능정보연구 Vol.24 No.2

고객과 대중의 니즈를 파악하기 위한 감성분석의 중요성이 커지면서 최근 영어 텍스트를 대상으로 다양한 딥러닝 모델들이 소개되고 있다. 본 연구는 영어와 한국어의 언어적인 차이에 주목하여 딥러닝 모델을 한국어 상품평 텍스트의 감성분석에 적용할 때 부딪히게 되는 기본적인 이슈들에 대하여 실증적으로 살펴본다. 즉, 딥러닝 모델의 입력으로 사용되는 단어 벡터(word vector)를 형태소 수준에서 도출하고, 여러 형태소 벡터(morpheme vector) 도출 대안에 따라 감성분석의 정확도가 어떻게 달라지는지를 비정태적(non-static) CNN(Convolutional Neural Network) 모델을 사용하여 검증한다. 형태소 벡터 도출 대안은 CBOW(Continuous Bag-Of-Words)를 기본적으로 적용하고, 입력 데이터의 종류, 문장 분리와 맞춤법 및 띄어쓰기 교정, 품사 선택, 품사 태그 부착, 고려형태소의 최소 빈도수 등과 같은 기준에 따라 달라진다. 형태소 벡터 도출 시, 문법 준수도가 낮더라도 감성분석 대상과 같은 도메인의 텍스트를 사용하고, 문장 분리 외에 맞춤법 및 띄어쓰기 전처리를 하며, 분석불능 범주를 포함한 모든 품사를 고려할 때 감성분석의 분류정확도가 향상되는 결과를 얻었다. 동음이의어 비율이 높은 한국어 특성 때문에 고려한 품사 태그 부착 방안과 포함할 형태소에 대한 최소 빈도수 기준은 뚜렷한 영향이 없는 것으로 나타났다. With the increasing importance of sentiment analysis to grasp the needs of customers and the public, various types of deep learning models have been actively applied to English texts. In the sentiment analysis of English texts by deep learning, natural language sentences included in training and test datasets are usually converted into sequences of word vectors before being entered into the deep learning models. In this case, word vectors generally refer to vector representations of words obtained through splitting a sentence by space characters. There are several ways to derive word vectors, one of which is Word2Vec used for producing the 300 dimensional Google word vectors from about 100 billion words of Google News data. They have been widely used in the studies of sentiment analysis of reviews from various fields such as restaurants, movies, laptops, cameras, etc. Unlike English, morpheme plays an essential role in sentiment analysis and sentence structure analysis in Korean, which is a typical agglutinative language with developed postpositions and endings. A morpheme can be defined as the smallest meaningful unit of a language, and a word consists of one or more morphemes. For example, for a word "예쁘고", the morphemes are "예쁘(= adjective)" and "고(=connective ending)". Reflecting the significance of Korean morphemes, it seems reasonable to adopt the morphemes as a basic unit in Korean sentiment analysis. Therefore, in this study, we use "morpheme vector" as an input to a deep learning model rather than "word vector" which is mainly used in English text. The morpheme vector refers to a vector representation for the morpheme and can be derived by applying an existent word vector derivation mechanism to the sentences divided into constituent morphemes. By the way, here come some questions as follows. What is the desirable range of POS(Part-Of-Speech) tags when deriving morpheme vectors for improving the classification accuracy of a deep learning model? Is it proper to apply a typical word vector model which primarily relies on the form of words to Korean with a high homonym ratio? Will the text preprocessing such as correcting spelling or spacing errors affect the classification accuracy, especially when drawing morpheme vectors from Korean product reviews with a lot of grammatical mistakes and variations? We seek to find empirical answers to these fundamental issues, which may be encountered first when applying various deep learning models to Korean texts. As a starting point, we summarized these issues as three central research questions as follows. First, which is better effective, to use morpheme vectors from grammatically correct texts of other domain than the analysis target, or to use morpheme vectors from considerably ungrammatical texts of the same domain, as the initial input of a deep learning model? Second, what is an appropriate morpheme vector derivation method for Korean regarding the range of POS tags, homonym, text preprocessing, minimum frequency? Third, can we get a satisfactory level of classification accuracy when applying deep learning to Korean sentiment analysis? As an approach to these research questions, we generate various types of morpheme vectors reflecting the research questions and then compare the classification accuracy through a non-static CNN(Convolutional Neural Network) model taking in the morpheme vectors. As for training and test datasets, Naver Shopping"s 17,260 cosmetics product reviews are used. To derive morpheme vectors, we use data from the same domain as the target one and data from other domain; Naver shopping"s about 2 million cosmetics product reviews and 520,000 Naver News data arguably corresponding to Google’s News data. The six primary sets of morpheme vectors constructed in this study differ in terms of the following three criteria. First, they come from two types of data source; Naver news of high grammatical correctness and Naver shopping’s cosme