• Asia Marketing Journal
• /
• 제16권3호
• /
• pp.23-33
• /
• 2014

The selection of keywords for bidding is a critical component of paid search advertising. When the number of possible keywords is enormous, it becomes difficult to choose the best keywords for advertising and then subsequently to assess their effect. To this end, we propose an ultrahigh dimensional keyword selection approach that not only reduces the dimension for selections, but also generates the top listed keywords for profits. An empirical analysis using a unique panel dataset from a large online clothes retailer that advertises on the largest search engine in China (i.e., Baidu) is presented to illustrate the usefulness of our approach.

• 대한음성학회:학술대회논문집
• /
• 대한음성학회 2002년도 11월 학술대회지
• /
• pp.183-186
• /
• 2002

The speech detection is one of the important problems in real-time speech recognition. The accurate detection of speech boundaries is crucial to the performance of speech recognizer. In this paper, we propose a speech detector based on Mel-band selection through training. In order to show the excellence of the proposed algorithm, we compare it with a conventional one, so called, EPD-VAA (EndPoint Detector based on Voice Activity Detection). The proposed speech detector is trained in order to better extract keyword speech than other speech. EPD-VAA usually works well in high SNR but it doesn't work well any more in low SNR. But the proposed algorithm pre-selects useful bands through keyword training and decides the speech boundary according to the energy level of the sub-bands that is previously selected. The experimental result shows that the proposed algorithm outperforms the EPD-VAA.

• 한국멀티미디어학회논문지
• /
• 제21권8호
• /
• pp.960-968
• /
• 2018

The mobile visual search service uses a query image to acquire linkage information through pre-constructed DB search. From the standpoint of this purpose, it would be more useful if you could perform a search on a web-based keyword search system instead of a pre-built DB search. In this paper, we propose a representative query extraction algorithm to be used as a keyword on a web-based search system. To do this, we use image classification labels generated by the CNN (Convolutional Neural Network) algorithm based on Deep Learning, which has a remarkable performance in image recognition. In the query extraction algorithm, dictionary meaningful words are extracted using Wikipedia, and hierarchical categories are constructed using WordNet. The performance of the proposed algorithm is evaluated by measuring the system response time.

• 한국전자거래학회지
• /
• 제23권2호
• /
• pp.83-96
• /
• 2018

문서를 대표하는 키워드를 추출하는 것은 문서의 정보를 빠르게 전달할 수 있을 뿐만 아니라 문서의 검색, 분류, 추천시스템 등의 자동화서비스에 유용하게 사용 될 수 있어 매우 중요하다. 그러나 웹사이트 문서에서 출현하는 단어의 빈도수, 단어의 동시출현관계를 통한 그래프 알고리즘 등의 기반으로 키워드를 추출할 경우 웹페이지 구조상 잠재적으로 주제와 관련이 없는 다양한 단어를 포함하고 있는 문제점과 한국어 형태소 분석의 정확성이 떨어지는 형태소 분석기 성능의 한계점 때문에 의미적인 키워드를 추출하는데 어려움이 존재한다. 따라서 본 논문에서는 의미적 단어 위주로 구축된 후보키워드들의 집합과 의미적 유사도 기반의 후보 키워드를 선택하는 방법으로써 의미적 키워드를 추출하지 못하는 문제점과 형태소 분석의 정확성이 떨어지는 문제점을 해결하고 일관성 없는 키워드를 제거하는 필터링 과정을 통해 최종 의미적 키워드를 추출하는 기법을 제안한다. 실 중소기업 웹페이지를 통한 실험 결과, 본 연구에서 제안한 기법의 성능이 통계적 유사도 기반의 키워드 선택기법보다 34.52% 향상된 것을 확인하였다. 따라서 단어 간의 의미적 유사성을 고려하고 일관성 없는 키워드를 제거함으로써 문서에서 키워드를 추출하는 성능을 향상시켰음을 확인하였다.

• 한국치위생학회지
• /
• 제21권5호
• /
• pp.639-644
• /
• 2021

Objectives: The purpose of this study was to analyze the factors that affected the number of citations for articles published in the Journal of Korean Society of Dental Hygiene based on previous studies. Methods: Information on papers including the number of citations was collected using a web crawling technique. The effect of the number of author keywords, the number of Medical Subject Headings (MeSH) keywords, MeSH match rate, abstract word count and keyword-abstract ratio on the number of citations was analyzed by multiple regression analysis. Results: The use of the MeSH keyword did not have a significant effect on the number of citations. Among the other factors, only the keyword-abstract ratio was statistically significant. Conclusions: Select a topic of constant interest in the field, write the title in detail using colons or asterisks if necessary, and do not repeat the words used in the title in keywords. Select specific keywords deeply related to the topic. In particular, choice words or phrases that are frequently used in the abstract. If the MeSH keyword selection contradicts the previous strategies, boldly give up the MeSH keyword.

• 패션비즈니스
• /
• 제23권3호
• /
• pp.51-66
• /
• 2019

The aim of this study is to identify research trends of "Journal of Fashion Business" by analyzing the keyword network of the paper published between 2006 and 2017. The papers selected for analysis in the study were 287 fashion design articles and 281 fashion marketing articles published between February 2006 and December 2017 and titles, volumes, publishing years, authors, keywords, and abstracts of each paper were collected for data analysis. The research was carried out through selection, collection of article data, keyword extraction and coding, keywords refinement, formation of network matrix, and analysis and visualization process. First, based on the title of the paper used in the analysis, the fashion design/aesthetics, marketing/social psychology, clothing materials, clothing composition, and other fields were classified. Research analysis used the Netminer 4 (Ver.4.3.2) program. Results indicated showed that the intellectual structure of the "Fashion Business" research paper showed key word changes over time, and the degree centrality and between centrality of the keywords.

• Journal of Information Technology Applications and Management
• /
• 제12권4호
• /
• pp.45-57
• /
• 2005

In order to provide personalized knowledge recommendation services, most web portals for organizational knowledge management use category or keyword information that portal users explicitly express interests in. However, it is usually difficult to collect correct preference data for all users with this approach, and, moreover, users' preferences may easily change over time, which results In outdated user profiles and impaired recommendation qualify. In order to address this problem, this paper suggests knowledge recommendation methods for portals using user profiles that are automatically constructed from users' activities such as posting or uploading of articles and documents. The result of our experiment shows that the Proposed method can provide equivalent performance with the manual category or keyword selection method.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 2001년도 추계학술대회 학술발표 논문집
• /
• pp.117-120
• /
• 2001

In this work, we propose a new method of extracting and weighting representative keywords(RKs) from a few documents that might interest a user. In order to extract RKs, we first extract candidate terms and then choose a number of terms called initial representative keywords (IRKS) from them through fuzzy inference. Then, by expanding and reweighting IRKS using term co-occurrence similarity, the final RKs are obtained. Performance of our approach is heavily influenced by effectiveness of selection method of IRKS so that we choose fuzzy inference because it is more effective in handling the uncertainty inherent in selecting representative keywords of documents. The problem addressed in this paper can be viewed as the one of calculating center of document vectors. So, to show the usefulness of our approach, we compare with two famous methods - Rocchio and Widrow-Hoff - on a number of documents collections. The results show that our approach outperforms the other approaches.

• 기술혁신학회지
• /
• 제8권2호
• /
• pp.860-885
• /
• 2005

This study relates to develop a quantitative method through which promising technology-based business items can be discovered and selected. For this study, we utilized patent trend analysis, technology-industry concordance analysis, and keyword co-occurrence analysis of US patents. By analyzing patent trends and technology-industry concordance, we were able to find out the emerging industry trends : prevalence of bio industry, service industry, and B2C business. From the direct and co-occurrence analysis of newly discovered patent keywords in the year, 2000, 28 promising business item candidates were extracted. Finally, the promising item candidates were prioritized using 4 business attractiveness determinants; market size, product life cycle, degree of the technological innovation, and coincidence with the industry trends. This result implicates that reliable discovery and selection of promising technology-based business items can be performed by a quantitative, objective and low- cost process using knowledge discovery method from patent database instead of peer review.

• 디지털융복합연구
• /
• 제11권11호
• /
• pp.351-357
• /
• 2013

질의 키워드의 정보 검색은 키워드의 의미가 다양하여 검색된 문서들이 사용자의 요구에 부합되지 않는 문제점을 가지며, 사용자의 상황과 특성이 사용자 마다 달라 정보가 매우 적고 연관성을 찾기 어렵다. 또한, 일반 상관 계수의 사용은 정보에 대한 연관성을 나타내지 못하는 문제가 있다. 본 연구에서는 시맨틱 웹 기술을 기반으로 인선에 필요한 다양한 개념들과 지식으로 구성된 인선 온톨로지와 인선 규칙들을 구축하고 규칙들을 지원하는 인선 절차와 인선 결과의 적합성을 확인할 수 있는 지식검색 시스템을 제안한다. 제안된 시스템에서는 스피어만 상관계수를 이용하여 사용자의 상황과 특성에 적합한 정보를 제공하여 제한적인 정보 추천의 단점을 해결하였다. 키워드 기반 검색과 시맨틱 기반 검색 실험 결과 시맨틱 기반 검색이 키워드 검색에 비하여 정확도는 90.3%, 재현율은 71.8%의 성능을 보였다.