• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권7호
• /
• pp.2633-2648
• /
• 2015

The problem of visual words' synonymy and ambiguity always exist in the conventional bag of visual words (BoVW) model based object category methods. Besides, the noisy visual words, so-called "visual stop-words" will degrade the semantic resolution of visual dictionary. In view of this, a novel bag of visual words method based on PLSA and chi-square model for object category is proposed. Firstly, Probabilistic Latent Semantic Analysis (PLSA) is used to analyze the semantic co-occurrence probability of visual words, infer the latent semantic topics in images, and get the latent topic distributions induced by the words. Secondly, the KL divergence is adopt to measure the semantic distance between visual words, which can get semantically related homoionym. Then, adaptive soft-assignment strategy is combined to realize the soft mapping between SIFT features and some homoionym. Finally, the chi-square model is introduced to eliminate the "visual stop-words" and reconstruct the visual vocabulary histograms. Moreover, SVM (Support Vector Machine) is applied to accomplish object classification. Experimental results indicated that the synonymy and ambiguity problems of visual words can be overcome effectively. The distinguish ability of visual semantic resolution as well as the object classification performance are substantially boosted compared with the traditional methods.

• 한국통신학회논문지
• /
• 제38B권12호
• /
• pp.954-961
• /
• 2013

본 논문은 퍼지ART의 학습 방법의 하나인 FCSR(Fast Commit Slow Recode)에서 패턴 인식을 향상시키기 위해 가변 학습을 이용하는 새로운 학습방법을 제안하였다. 기존의 학습 방법은 연결 강도(대표패턴)의 갱신에 고정된 학습률이 사용된다. 이 방법은 같은 카테고리 내의 입력패턴과 대표패턴의 유사성의 정도와 관계없이 고정된 학습률로 연결 강도를 갱신한다. 이 경우 카테고리 경계에 있는 유사성이 낮은 입력패턴이 연결강도의 갱신에 크게 영향을 주게 된다. 따라서 잡음 환경에서 이것은 불필요한 카테고리 증식의 원인이 되고, 패턴 인식 능력을 낮추는 문제가 된다. 제안된 방법에서는 대표 패턴과 입력 패턴 사이에 유사성이 적을수록 연결강도의 갱신에 입력패턴의 기여를 낮추어간다. 그 결과 잡음환경에서 퍼지 ART의 불필요한 카테고리 증식을 억제하였고, 패턴 인식 능력을 향상시켰다.

• 한국도서관정보학회지
• /
• 제48권1호
• /
• pp.161-189
• /
• 2017

이 연구는 시각장애 연구자의 학술정보 획득 및 활용에 대한 경험을 시각장애 연구자의 관점에서 기술하고, 그 경험의 내재된 본질을 탐색해 보는데 목적이 있다. 이를 위해 시각장애 연구자 12명을 심층 면담하고 Colaizzi의 현상학적 연구방법으로 분석하였다. 107개의 의미 있는 진술과 이를 통해 구성된 의미에서 44개의 주제, 12개의 주제묶음을 확인하여 구분한 결과, 총 4개의 범주가 최종적으로 도출되었다. '학술정보 확보 및 활용' 범주는 학술정보 제공 사이트의 웹 접근성 및 원문 접근성 수준이 낮아 시각장애 연구자들이 자료 검색-확보-활용과정에서 겪는 어려움을 보여준다. '인식 및 감정의 변화' 범주는 시간의 흐름에 따른 시각장애 연구자의 인식 및 감정을, '지원인력과의 관계' 범주는 지원인력 확보의 어려움, 지원인력의 역량 문제, 지원인력 관리와 관련하여 시각장애 연구자가 겪는 고충을 보여준다. '개선요구사항 범주'는 연구 참여자들이 학술정보 획득 및 활용과 관련하여 개선되기를 희망하는 내용이 무엇인지를 기술하고 있다.

• 한국문헌정보학회지
• /
• 제29권
• /
• pp.205-230
• /
• 1995

The publication of books on modern agriculture started under the influence of Silhak Thought developed from the late Chosun Dynasty. The common intention found in these publications is the objective to enrich the country through increased agricultural productivity and thus to secure national independence from the surrounding powers. The study reviews three different categories in this area; five general books on agriculture, three on agriculture related legal regulations, and ten books introducing techniques of commercial agriculture. The first category is comprehensive treatment of general agricultural content and the second is legal regulations which affected the contemporary agriculture. The third category is introductory books on commercial techniques aimed at accumulating wealth through agriculture. Silkworm cultivation occupies an important place in these publications owing to the active encouragement given by the Section of Silkworm Cultivation in the Ministry of Agriculture, Commerce and Industry. We can recognise the extensive involvement of Suh Pyung Sook, who served as the Section Chief of Silkworm Cultivation in 1904, by examining introductions and prefaces of many books in this category. The examples of general books on agriculture are 'New Agricultural Administration' (농정신편), 'General Introduction on Agiculture' (농업대요), 'New Textbook on Agricuture' (신찬농업교과서), 'Pragmatic Agriculture' (실리농방신편), Of these, 'New Agricultural Administration' (농정신편) is evaluated as the first publication on modern agriculture. It was written in 1881, and the first edition was published was published in 1901 with the second edition following in 1905. Examples of the second category are; 'The regulation on utilizing uncultivated state owned land' (국유미간지리용법) legislated and declared with the intention of expanding cultivated area, 'Mining and Forestry Regulation' (광임법규), 'Forest Land Regulation' (임야법령) to administer mining and forestry. Books on commercial agriculture take up the highest proportion in the agricultural publication of this period. These cover silkworm cultivation, chicken farming, vegetables and fruits. The books introduce Western techniques with more scientific and rational approach especially on Silkworm cultivation and chichen farming which had become increasingly significant as commercial agriculture from the late Chosun Dynasty.

• 한국비블리아학회지
• /
• 제26권2호
• /
• pp.217-239
• /
• 2015

올바른 독서치료를 위해서 이용자는 자신의 상황과 증상에 대한 자가판단과 함께 치유서의 올바른 선정및 접근이 필요하다. 이용자는 독서치료 독서목록의 상황별 카테고리를 이용해서 치유서에 접근하지만, 독서치료를 원하는 이용자가 자신의 복잡한 상황 및 증상을 간략하게 정리하여 하나의 단어 혹은 키워드로 정의하는 것은 어렵다. 국내 독서치료 독서목록은 연령별 카테고리를 제외한 1개 혹은 최대 2개의 상황별 카테고리를 제공하며, 이로 치유서를 분류하고 있다. 이용자가 복잡한 상황을 가지고 카테고리에 접근하는 만큼, 이용자의 접근 효율성과 용이성을 높이기 위해서는 1개의 카테고리 보다는 복수의 카테고리 제공이 필요하다. 이에 본 연구에서는 독서치료를 원하는 이용자가 원하는 치유서에 보다 효율적으로 접근할 수 있도록, 기초 연구로서 국내의 독서치료 독서목록과 치유서의 관계에 대해 분석하였다. 국내 대표 독서치료 독서목록을 이용하여 카테고리간의 매핑과 크로스워크를 실시하였고, 독서치료 독서목록의 전반적인 조사를 통해서 국내 독서치료 독서목록 카테고리를 분석하였다.

• Agribusiness and Information Management
• /
• 제8권2호
• /
• pp.1-8
• /
• 2016

The present study has aimed at analyzing the technical and scale efficiencies of credit utilization by the farmer-borrowers in Chittoor district of Andhra Pradesh, India. DEA approach was followed to analyze the credit utilization efficiency and to analyze the factors influencing the credit utilization efficiency, log-linear regression analysis was attempted. DEA analysis revealed that, the number of farmers operating at CRS are more in number in marginal farms (40%) followed by other (35%) and small (17.5%) farms. Regarding the number of farmers operating at VRS, small farmers dominate the scenario with 72.5 per cent followed by other (67.5%) and marginal (42.5%) farmers. With reference to scale efficiency, marginal farmers are in majority (52.5%) followed by other (47.5%) and small (25%) farmers. At the pooled level, 26.7 per cent of the farmers are being operated at CRS, 63 per cent at VRS and 32.5 per cent of the farmers are either performed at the optimum scale or were close to the optimum scale (farms having scale efficiency values equal to or more than 0.90). Nearly 58, 15 and 28 percents of the farmers in the marginal farms category were found operating in the region of increasing, decreasing and constant returns respectively. Compared to marginal farmers category, there are less number of farmers operating at CRS both in small farmers category (15%) and other farmers category (22.5%). At the pooled level, only 5 per cent of the farmers are operating at DRS, majority of the farmers (73%) are operating at IRS and only 22 per cent of the farmers are operating at CRS indicating efficient utilization of credit. The log-linear regression model fitted to analyze the major determinants of credit utilization (technical) efficiency of farmer-borrowers revealed that, the three variables viz., cost of cultivation and family expenditure (both negatively influencing at 1% significant level) and family income (positively influencing at 1% significant level) are the major determinants of credit utilization efficiency across all the selected farmers categories and at pooled level. The analysis further indicate that, escalation in the cost of cultivation of crop enterprises in the region, rise in family expenditure and prior indebtedness of the farmers are showing adverse influence on the credit utilization efficiency of the farmer-borrowers.

• 한국정보과학회논문지:소프트웨어및응용
• /
• 제32권7호
• /
• pp.674-682
• /
• 2005

본 논문에서는 다중분광 영상의 분류를 위하여 퍼지 G-K(Gustafson- Kessel) 알고리즘과 PCM 알고리즘을 융합한 분류방법을 제안하였다. 제안된 방법은 학습데이타를 이용하여 퍼지 G-K 알고리즘을 수행한 후 그 결과를 이용하여 PCM 알고리즘을 수행한다 PCM 알고리즘과 퍼지 G-K 알고리즘 분류결과를 비교하여 그 결과가 일치하면 해당 항목으로 분류항목을 결정한다. 일치하지 않는 화소는 PCM 알고리즘의 평균내부거리 안쪽에 있는 화소들을 새로운 학습데이타로 하여 베이시안 최대우도 분류를 수행하여 분류항목을 결정한다. 평균내부거리 안쪽에 있는 화소 데이타는 정규분포형태를 보여준다. 다차원 다중분광 영상인 IKONOS와 LANDSAT TM 위성영상을 이용하여 제안된 알고리즘의 효율성을 검증한 결과 퍼지 G-K 알고리즘과 PCM 알고리즘 그리고 전통적인 분류 방법인 최대우도 분류 알고리즘보다 전체 정확도가 더 높은 결과를 얻을 수 있었다

• 한국지리정보학회지
• /
• 제24권3호
• /
• pp.10-31
• /
• 2021

본 연구를 통해 정의된 용도복합 개념은 세 가지 이상의 주요 도시기능의 복합을 의미하며 도시공간의 경제적, 사회적, 환경적 가치를 동시에 실현하기 위한 수단을 의미한다. 이러한 정의 하에서 서울의 용도복합 현상을 탐구하였다. 상대적 엔트로피 개념을 사용하여 단위 지역인 법정동을 기준으로 복합도(R.E.)를 측정하였으며 도시기능의 유형은 용도복합의 정의에서 도출한 개념을 통해 건축물 층별 용도를 유형화하여 계량화하였다. 서울의 용도복합 현상은 복합도(R.E.) 0.631 이하를 '복합아님', 복합도(R.E.) 0.631 초과 0.884 미만을 '불균형 복합', 복합도(R.E.) 0.884 이상 0.991 미만을 '균형복합', 복합도(R.E.) 0.991 이상을 '완전균형복합'의 단계로 구분하여 용도복합 현상을 해석하는 기준으로 지표화하였다.

• 한국농촌건축학회논문집
• /
• 제24권4호
• /
• pp.33-40
• /
• 2022

Sharehouse has been supplied as an alternative to solving the steadily increasing one-room housing problem of single-person households every year, and it is necessary to investigate the satisfaction of residents who actually live in sharehouse through P.O.E. Therefore, this study analyzed priror researches related to the existing one-person households, sharehouses, and satisfaction surveys, and summarized indicators with high relevance and frequency to derive satisfaction survey factors that can clearly evaluate the improvement architectural plan of sharehouses. As a result, it was classified into 4 items in the 'general information' category to investigate the status, housing rental type, and housing cost of the sharehouse, 15 factors in the 'peripheral environment' category to evaluate the safety, 3 factors in the 'community' category, and 17 factors in the space (facility) and service category. In the "General Information" section, the overall one-person housing satisfaction, desired sharehouse type, housing rental type, housing cost, and living expenses were reduced. In the "Surrounding Environment", accessibility to public office, accessibility to cultural facilities, accessibility to medical facilities, accessibility to work and school, convenience stores, noise pollution and safety. In addition, in the "community" section, it consists of interactions with various people, relationships with housemates and in the "space (facilities) and service" section, heating, waterproof, soundproof, ventilation, moisture and condensation blocking, facility management, interior, room size, built-in furniture, storage space, laundry, parking. Most of the scales for each factor were 5-point Likert scales, allowing evaluation of the degree of satisfaction, and some factors presented criteria to induce structured answers. Therfore, it is expected that the survey will be conducted on residents who actually live by deriving factors for the satisfaction survey of one-person households living in the sharehouse, and the current status of the sharehouse will be identified, and the degree of satisfaction will be analyzed to be used for research.

• 유통과학연구
• /
• 제15권3호
• /
• pp.61-72
• /
• 2017

Purpose - The ICT development is affecting the consumer behaviors in selecting channel or distribution system. This study aims to advance the theory on the influence and interaction with omni-channel behaviors. Specifically, analyzing moderating variable is category knowledge that effect between propensity of brand loyalty and its precedence factor which is perceived difference, perceived value, authenticity and consumer-brand relationship. Research design, data, and methodology - The subject of this research is consumers who purchase goods in omni-channel situation. The hypothesis of this research is derived from the literature of the preceding research analysis on brand loyalty, omni-channel and consumer behaviors. This study have constructs that were defined operationally with reference to previous studies, and the research model was designed to figure out the structural relationship among perceived difference, perceived value, authenticity, consumer-brand relationship and brand loyalty. From 2016 Sept. 1 to Dec. 31, a questionnaire survey was performed targeting customers using omni-channel. 327 questionnaire survey had conducted. 316 survey data were used for empirical analysis except data that had missing and wrong value. AMOS(structural equation) was used to confirm the hypothesis which developed by researcher. Results - The results of this study are as follows. First, an authenticity has significant effect on brand loyalty. Second, in the omni-channel situation, but perceived differentiation, perceived value, consumer-brand relationship does not affect brand loyalty. According to this result, it is judged that it is easy to search for information in the situation of omni-channel and integrated decision making is done without distinction between channels. Third, category knowledge has moderating effect between brand loyalty and precedence factors. When the category knowledge level is low, preceding factors have a significant effect on brand loyalty. when the category knowledge level is high, the preceding factors did not have a significant effect on brand loyalty except the authenticity. Conclusions - This study finds out omni-channel's phenomenon is different from other distribution channel phenomenon. In the situation of omni-channel, it is suggested that brand loyalty may be relatively low for a certain brand because it raises the knowledge level of the category. Then this study provides a managerial implications based on the role of the moderate effect on category knowledge, brand loyalty and omni-channel.