• 디지털산업정보학회논문지
• /
• 제16권4호
• /
• pp.19-28
• /
• 2020

Vehicle license plate recognition system is not generalized in Malaysia due to the loose character layout rule and the varying number of characters as well as the mixed capital English characters and italic English words. Because the italic English word is hard to segmentation, a separate method is required to recognize in Malaysian license plate. In this paper, we propose a mixed character level and word level English license plate recognition algorithm using deep learning neural networks. The difference of Gaussian method is used to segment character and word by generating a black and white image with emphasized character strokes and separated touching characters. The proposed deep learning neural networks are implemented on the LPR system at the gate of a building in Kuala-Lumpur for the collection of database and the evaluation of algorithm performance. The evaluation results show that the proposed Malaysian English LPR can be used in commercial market with 98.01% accuracy.

• 대한임베디드공학회논문지
• /
• 제17권4호
• /
• pp.199-208
• /
• 2022

Korean is an agglutinative language, and one or more morphemes are combined to form a single word. Part-of-speech tagging method separates each morpheme from a word and attaches a part-of-speech tag. In this study, we propose a new Korean part-of-speech tagging method based on the Head-Tail tokenization technique that divides a word into a lexical morpheme part and a grammatical morpheme part without decomposing compound words. In this method, the Head-Tail is divided by the syllable boundary without restoring irregular deformation or abbreviated syllables. Korean part-of-speech tagger was implemented using the Head-Tail tokenization and deep learning technique. In order to solve the problem that a large number of complex tags are generated due to the segmented tags and the tagging accuracy is low, we reduced the number of tags to a complex tag composed of large classification tags, and as a result, we improved the tagging accuracy. The performance of the Head-Tail part-of-speech tagger was experimented by using BERT, syllable bigram, and subword bigram embedding, and both syllable bigram and subword bigram embedding showed improvement in performance compared to general BERT. Part-of-speech tagging was performed by integrating the Head-Tail tokenization model and the simplified part-of-speech tagging model, achieving 98.99% word unit accuracy and 99.08% token unit accuracy. As a result of the experiment, it was found that the performance of part-of-speech tagging improved when the maximum token length was limited to twice the number of words.

• 한국컴퓨터산업학회논문지
• /
• 제2권10호
• /
• pp.1301-1308
• /
• 2001

이 논문에서는 기존의 number-theoretic 순환함수와 연계된 word-theoretic 순환함수 및 술어(predicates)들의 Grzegorzyk 클래스를 논한다. 특히 small 클래스 $\xi\sum{n}$($n\leq2$)에서의 특성은 그에 대응하는 number-theoretic small 클래스 $\xi\sum{n}$과는 매우 틀린 특성을 보인다 [2]. 흥미 있는 문제 중의 하나인 $\xi\sum{0}$ 등급에서의 동치문제는 undecidable 임을 증명한다.

• Journal of Information Technology Applications and Management
• /
• 제21권4호
• /
• pp.157-171
• /
• 2014

As electronic word of mouth plays an important role in purchase behavior among consumers, the number of studies on the impact of electronic word of mouth is rapidly increasing. Nevertheless, it is difficult to discover comparative studies on the mass media which had a great impact on consumer's purchase behavior before the impact of electronic word of mouth becomes greater versus the social media where electronic words of mouth are created and distributed. It is considered that it seems to be necessary to find an appropriate mutual supplement point between the media designed for a successful marketing by comparing and analyzing the existing mass media versus the social media, major media for electronic word of mouth. Therefore, this study aims to compare and analyze the impact of comments on movie revenue in the representative forms of mass media such as online news and social media blogs. In particular, this study also considers an appropriate media for promoting movies by period by comparing and analyzing the two media before and after film release. For analysis, this study collects the information on the number of comments on online news and blogs in 70 Korean movies released in 2011 and 2012 from five weeks before film release to eight weeks after film release on a daily basis via Naver. This study also collects the information on the movie revenue using the statistical data of movie industry from Korean Film Commission. As a result of empirical data analysis, it is found that the two media showed no difference in movie revenue before film release, but after film release, the impact of blogs was more significant than that of online news.

• 정보처리학회논문지B
• /
• 제9B권6호
• /
• pp.791-798
• /
• 2002

본 논문에서는 코퍼스를 이용한 확률 의존문법 자동 생성 기술을 다룬다. 한국어의 부분 자유 어순성질과 문장의 필수적 성분의 생략과 같은 특성으로 인하여 한국어 구문분석에 관한 연구들에서는 주로 의존문법을 선호하고 있다. 본 논문에서는 기존의 어절단위학습방법에서는 학습할 수 없었던 어절 내의 의존관계를 학습할 수 있는 형태소 단위의 학습 방법을 제안한다. KAIST의 트리 부착 코퍼스 약 3만 문장에서 추출한 25,000문장의Tagged Corpus을 가지고 한국어 확률 의존문법 학습을 시도하였다. 그 결과 초기문법 2,349개의 정확한 문법을 얻을 수 있었으며, 문법의 정확성을 실험하기 위해 350개의 실험문장을 parsing한 결과 69.77%의 파싱 정확도를 보였다. 이로서 한국어 어절 특성을 고려한 형태소 단위 학습으로 얻어진 의존문법이 어절 단위 학습으로 얻어진 문법보다 더 정확하다는 사실을 알 수 있었다.

• Asia Marketing Journal
• /
• 제21권2호
• /
• pp.73-98
• /
• 2019

As information and communication technology continue its remarkable development, the exchange of information online becomes as prevalent and frequent as face-to-face communication in daily life. Therefore, the management and application of WOM (word of mouth) practices will become more important than ever to companies. Currently, there are various types of communication channels for online WOM, and each channel has its own unique traits. Most of the previous research studies online WOM by examining the information inside a single communication channel, but this research chooses two different communication channels and analyzes the effects of online WOM with each channel's unique characteristics. More specifically, this research focuses on the expectation that the effects of information from Twitter and blogs on product sales may differ because Twitter and blogs, two different communication channels for online WOM, have their own unique traits. Our particular aim is to perform an in-depth examination on the effects of communication channel's volume and valence on product sales, two important attributes of online WOM. Furthermore, while most of the empirical research focuses on online WOM and analyzes its effect on markets of temporary experience goods, such as movies and books, this research highlights focuses on the automobile market, a durable goods market. The results of our analysis are as follows: First, regarding blogs, a positive valence significantly and positively affects the sales of products, and this result indicates that consumers are influenced more by the emotional aspect of a product presented in a post than by the number of blog posts. Second, regarding Twitter, the volume of online WOM significantly and positively affects sales, an indication that as the number of posts increase, the sales increase. Through this research, we suggest that even those firms that sell durable goods can increase sales through the management and application of online WOM. Moreover, according to the characteristics of communication channels, the effects of online WOM on sales differ. As a practical implication of this research, we suggest that companies can and should create marketing strategies appropriate to their targeted communication channels.

• 아태비즈니스연구
• /
• 제12권3호
• /
• pp.51-69
• /
• 2021

Purpose - Based on literature studies of movie reviews and movie ratings, this study raised two research questions on the contents of online word of mouth and the number of movie screens as mediator variables. Research question 1 wanted to figure out which topics of word groups had a positive or negative impact on movie ratings. Research question 2 tried to identify the role of the number of movie screens between movie ratings and box office outcomes. Design/methodology/approach - Through R program, this study collected about 82,000 movie reviews and movie ratings posted on Naver's movie website to examine the role of online word of mouths and movie screen counts in 10 movies that were considered commercially unsuccessful with fewer than 2 million viewers despite securing about 1,000 movie screens. To confirm research question 1, topic modeling, a text mining technique, was conducted on movie reviews. In addition, this study linked the movie ratings posted on Naver with information of KOBIS by date, to identify the research question 2. Findings - Through topic modeling, 5 topics were identified. Topics found in this study were largely organized into two groups, the content of the movie (topic 1, 2, 3) and the evaluation of the movie (topics 4, 5). When analyzing the relationship between movie reviews and movie ratings with 5 mediators identified in topic modeling to probe research question 1, the topic word groups related to topic 2, 3 and 5 appeared having a negative effect on the netizen's movie ratings. In addition, by connecting two secondary data by date, analysis for research question 2 was implemented. The outcomes showed that the causal relationship between movie ratings and audience numbers was mediated by the number of movie screens. Research implications or Originality - The results suggested that the information presented in text format was harder to quantify than the information provided in scores, but if content information could be digitalized through text mining techniques, it could become variable and be analyzed to identify causality with other variables. The outcomes in research question 2 showed that movie ratings had a direct impact on the number of viewers, but also had indirect effects through changes in the number of movie screens. An interesting point is that the direct effect of movie ratings on the number of viewers is found in most American films released in Korea.

• 한국정보과학회논문지:소프트웨어및응용
• /
• 제31권8호
• /
• pp.1101-1112
• /
• 2004

본 논문에서는 웹 검색의 성능 향상을 위해 질의어 의미별 사용자 선호도를 이용한 웹 페이지의 가중치 부여 방식을 제안한다. 일반적으로 검색엔진들은 검색 질의어와 웹 페이지의 어휘 비교에 의한 관련도 측정만을 사용하여 웹 페이지의 가중치를 부여한다. 웹과 같이 방대한 자료를 대상으로 검색을 할 경우 유사한 관련도를 가진 검색 결과가 매우 많으므로 어휘 비교만으로는 중요한 웹 페이지를 선별하기 어렵다. 본 논문에서는 질의어의 의미를 구분하도록 워드넷(WordNet)을 이용한 사용자 인터페이스를 구축하고, 사용자의 클릭 수를 각 웹 페이지의 가중치에 누적함으로써 다수 사용자의 검색 행위에 의한 묵시적 평가가 웹 페이지의 검색 순위에 반영되는 검색 시스템을 구현하였다. 클릭수의 누적에 있어서 질의 어 의미별로 가중치를 구분하여 저장함으로써 일반적인 검색엔진보다 정확한 검색이 되었으며, 웹 페이지의 범주별 가중치와 질의어의 의미별 사용자 선호도를 이용함으로써 검색 시스템의 성능을 향상시킬 수 있다는 것을 20개의 어휘에 관련된 41개의 의미들을 대상으로 실험한 결과로 확인하였다.

• 대한한의학회지
• /
• 제41권3호
• /
• pp.1-8
• /
• 2020

Objectives: This paper aims to investigate the Donguibogam-based pattern diagnosis by applying natural language processing and machine learning. Methods: A database has been constructed by gathering symptoms and pattern diagnosis from Donguibogam. The symptom sentences were tokenized with nouns, verbs, and adjectives with natural language processing tool. To apply symptom sentences into machine learning, Word2Vec model has been established for converting words into numeric vectors. Using the pair of symptom's vector and pattern diagnosis, a pattern prediction model has been trained through Logistic Regression. Results: The Word2Vec model's maximum performance was obtained by optimizing Word2Vec's primary parameters -the number of iterations, the vector's dimensions, and window size. The obtained pattern diagnosis regression model showed 75% (chance level 16.7%) accuracy for the prediction of Six-Qi pattern diagnosis. Conclusions: In this study, we developed pattern diagnosis prediction model based on the symptom and pattern diagnosis from Donguibogam. The prediction accuracy could be increased by the collection of data through future expansions of oriental medicine classics.

• 대한의료기공학회지
• /
• 제20권1호
• /
• pp.148-162
• /
• 2020

Objective : The purpose of this study is not only to examine the negative functions of alcohol, but also to examine the positive functions and look at it with a balanced perspective. Methods : We investigated the number of times the word "酒"(alcohol) appears in 《Hwangjenaegyeong》 and 《Gyeongakjeonseo》. The meaning of alcohol was divided into seven categories. The number of positive and negative functions of alcohol was counted and the ratio was calculated. Results : 1. In the 《Hwangjenaegyeong》, the word alcohol appears 23 times, of which 9 times has positive functions, 10 times has negative functions, and the remaining 4 times does not correspond anywhere. The ratio of positive functions is 39.13% and negative functions 43.48%. 2. In the 《Gyeongakjeonseo》, the word alcohol appears 1,487 times, of which 1,140 times (76.66%) has positive functions, and 327 times (21.99%) has negative functions. Conclusions : 1. In 《Hwangjenaegyeong》, the number of comments about positive and negative functions of alcohol is similar. 2. 《Gyeongakjeonseo》 has commented a lot about the positive functions of alcohol. However, it has more mentions to negative functions of alcohol except for those related to herbs and prescriptions (48 to 64 chapter). 3. It is somewhat unreasonable to judge the emphasis on the positive and negative function of alcohol according to the number of references to alcohol in oriental medicine classics. But in these books, we can find a balanced approach between the two sides, not a one-sided biased view. 4. From a curative point of view, it is desirable to know and to use the positive and negative functions of alcohol, and drink alcohol as appropriate control.