• 대한전자공학회논문지SP
• /
• 제40권3호
• /
• pp.194-202
• /
• 2003

인쇄문서를 전자문서로 자동변환하기 위해서는 문서영상 영역해석과 문자인식 기술이 필요하다. 이들 중 영역해석은 문서영상을 세부 영역으로 분할하고, 분할한 영역을 문자, 그림, 표 등의 형태로 분류한파. 그러나 문자와 그림의 일부는 크기, 밀도, 화소분포의 복잡도가 비슷하여 정확한 분류가 어렵다. 따라서 영역해석에서의 오 분류는 자동변환을 어렵게 만드는 주된 원인이 된다. 본 논문에서는 분서영상을 문자와 그림영역으로 분할하는 영역해석 방법을 제안한다. 문자와 그림의 분류는 1차원 메디안 필터링을 기반으로 한 방법을 이용하여 언급한 문제점을 해결한다. 또한 메디안 필터링에 의해 발생하는 볼드체 문자와 그래프나 표와 같은 그림영역의 오 분류 문제를 표피 제거 필터와 문자의 최대크기를 이용하여 해결한다. 따라서 상용제품을 포함한 기존의 영역해석 방법보다 그 성능이 우수하다.

• 정보관리연구
• /
• 제41권3호
• /
• pp.27-44
• /
• 2010

지금까지 특허분석에 관한 연구는 특허지도(Patent Map; PM)분석을 수행하여 기술동향을 분석하고 있으나, 주제에 대한 검색결과를 엑셀 등을 이용하여 그래프화하여 출원빈도에 대한 추이를 나타내는 것이 대부분이다. 이러한 분석은 특정주제에 대한 것으로 국가의 산업이나 국제적인 산업의 동향을 직접 비교 분석하기에는 미흡한 실정이다. 본 연구에서는 PM분석의 신뢰성을 높이기 위해 통계분석의 필요성을 제시하고, 1990년부터 2004년에 출원된(출원번호 기준) 한국특허 전체를 대상으로 연도별, 분류별(IPC Section 분류)로 각 국가(한국, 미국, 일본, 독일)의 출원빈도를 통계적으로 유의한 수준을 분석하였다. 또한 국가 및 기술 분류에 대한 출원빈도에 대해 포아송 회귀분석을 수행하고, 통계 적으로 유의한 수준에 따른 평가를 R&D에 활용할 수 있도록 하였다.

• 대한토목학회논문집
• /
• 제41권3호
• /
• pp.277-288
• /
• 2021

IFC 정보의 시멘틱 무결성 확보를 위해 BIM 부재와 IFC 엔티티 간 매핑 검증이 필요하다. 이와 관련된 기존 연구들은 기하정보 기반으로 학습시킨 기계학습 알고리즘을 활용하여 BIM 부재 인식 및 분류를 통해 매핑 검증을 실시하였으나, 유사한 기하특성을 가진 부재를 구분하지 못한다는 한계점이 존재하였다. 이에 본 연구는 BIM 모델의 주요 부재를 인공신경망 기반으로 자동 분류하되, 부재 간 관계정보를 삽입하여 분류성능을 향상시키는 것을 목적으로 하였다. 이를 위해 기존 특성 외에 구조화된 신호를 함께 학습하는 NSL 프레임워크를 활용하여 8개의 BIM 부재를 분류하는 모델을 구축하였으며, 그 결과 기하정보 기반 인공신경망 모델과 대비하여 부재 간 관계정보를 삽입한 NSL 모델의 분류정확도가 현저히 상승한 것을 확인하였다.

• 한국조경학회지
• /
• 제25권2호
• /
• pp.73-81
• /
• 1997

The main purpose of this study was to develop a new method for evaluation of stream naturalness in order to appraise and prescribe for streams effectively in the process of ecological restoration of stream corridors. The results are as follows : 1) For this purpose six factors were selected on considering the spatial axes of stream corridor variation and total 20 descriptors about the physical structure were selected. 2) The calculation of S.N.I. for each segment was consisted of three steps, such as calculation of S.N.I.s of the individual descriptors, averaging all the descriptors's for each factor, and finally averaging the factors's for the Total S.N.I. 3) The evaluation unit was decided to be 100m size. The score system ranging 1~5 was adopted. Weighting parameters of factors were unified with each other. 4) A GIS model was adopted for classification, calculation, querying, analysing, and presenting S.N.I. information. And the format of S.N.I. maps including statistical graphs and other spatial watershed information was designed for the GIS odel. The naturalness of stream corridor was was investigated by the naturalness of habitat, and assessed by the descriptors focused on physical structure, therefore the S.N.I. can manifest prescriptions for restoration of the stream corridor. On the other hand because some evaluation factors such as water quality, water volume, fauna, flora, functions of stream exosystem has been excluded, S.N.I. could have some limits on representing the full aspects of stream naturalness. This evaluation method is hypothetical one, so it would be investigated through iterative applicatons.

• Asia pacific journal of information systems
• /
• 제29권4호
• /
• pp.838-855
• /
• 2019

Online product reviews are a vital source for companies in that they contain consumers' opinions of products. The earlier methods of opinion mining, which involve drawing semantic information from text, have been mostly applied in one dimension. This is not sufficient in itself to elicit reviewers' comprehensive views on products. In this paper, we propose a novel approach in opinion mining by projecting online consumers' reviews in a multidimensional framework to improve review interpretation of products. First of all, we set up a new framework consisting of six dimensions based on a marketing management theory. To calculate the distances of review sentences and each dimension, we embed words in reviews utilizing Google's pre-trained word2vector model. We classified each sentence of the reviews into the respective dimensions of our new framework. After the classification, we measured the sentiment degrees for each sentence. The results were plotted using a radar graph in which the axes are the dimensions of the framework. We tested the strategy on Amazon product reviews of the iPhone and Galaxy smartphone series with a total of around 21,000 sentences. The results showed that the radar graphs visually reflected several issues associated with the products. The proposed method is not for specific product categories. It can be generally applied for opinion mining on reviews of any product category.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권8호
• /
• pp.2948-2963
• /
• 2015

In the field of computer vision, visual surveillance systems have recently become an important research topic. Growth in this area is being driven by both the increase in the availability of inexpensive computing devices and image sensors as well as the general inefficiency of manual surveillance and monitoring. In particular, the ultimate goal for many visual surveillance systems is to provide automatic activity recognition for events at a given site. A higher level of understanding of these activities requires certain lower-level computer vision tasks to be performed. So in this paper, we propose an intelligent activity recognition model that uses a structure learning method and a classification method. The structure learning method is provided as a K2-learning algorithm that generates Bayesian networks of causal relationships between sensors for a given activity. The statistical characteristics of the sensor values and the topological characteristics of the generated graphs are learned for each activity, and then a neural network is designed to classify the current activity according to the features extracted from the multiple sensor values that have been collected. Finally, the proposed method is implemented and tested by using PETS2013 benchmark data.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제10권4호
• /
• pp.1643-1660
• /
• 2016

Connectivity and trust in social networks have been exploited to propose applications on top of these networks, including routing, Sybil defenses, and anonymous communication systems. In these networks, and for such applications, connectivity ensures good performance of applications while trust is assumed to always hold, so as collaboration and good behavior are always guaranteed. In this paper, we study the impact of differential behavior of users on performance in typical social network-enabled routing applications. We classify users into either collaborative or rational (probabilistically collaborative) and study the impact of this classification and the associated behavior of users on the performance of such applications, including random walk-based routing, shortest path based routing, breadth-first-search based routing, and Dijkstra routing. By experimenting with real-world social network traces, we make several interesting observations. First, we show that some of the existing social graphs have high routing costs, demonstrating poor structure that prevents their use in such applications. Second, we study the factors that make probabilistically collaborative nodes important for the performance of the routing protocol within the entire network and demonstrate that the importance of these nodes stems from their topological features rather than their percentage of all the nodes within the network.

• 지적과 국토정보
• /
• 제47권1호
• /
• pp.53-79
• /
• 2017

본 연구의 주된 목적은 토픽 모델링(topic modeling)을 이용하여 트위터 상에서 스모그 리스크(smog risks)에 관한 대중 인식(public perceptions)을 측정하고 분류하는 것이다. 선행연구에 있어서 연구 갭(research gap)을 확인하기 위하여 본 연구는 스모그 리스크와 토픽 모델링에 대한 선행연구를 검토하였다. 그 결과 본 저자는 기존의 연구에서 상당한 연구 갭이 존재하고 있음을 확인하였으며, 이러한 연구 갭을 메우기 위해 다섯 개의 연구 질문을 설정하였다. 연구 질문들에 답을 구하기 위하여 본 연구는 10,000개의 트위터 자료를 추출하였고, 이에 대하여 워드 클라우드 분석(word cloud analysis), 상관분석, LDA를 이용한 토픽 모델링, 스트림그래프(stream graph), 위계적 집락분석(hierarchical cluster analysis)을 실시하였다. 분석 결과 자주 언급되는 단어들(the most frequent terms), 단어네트워크(terms network)의 형태, 상관관계의 유형, 스모그 관련 주제의 변동패턴에 있어서 뉴욕과 런던 사이에 큰 차이가 있음을 확인하였다. 그리하여 본 저자는 다섯 개의 연구 질문 중 네 개에 대하여 긍정적인 답을 구할 수 있었고, 이를 토대로 몇 가지 정책적 시사점을 제시하고, 향후 연구를 위한 제안들을 하였다.

• International Journal of Computer Science & Network Security
• /
• 제22권9호
• /
• pp.43-50
• /
• 2022

For high-quality education of the modern generation of students, forms of organizing the educational process and the latest methods of obtaining knowledge that differ from traditional ones are necessary. The importance of multimedia teaching tools is shown, which are promising and highly effective tools that allow the teacher not only to present an array of information in a larger volume than traditional sources of information, but also to include text, graphs, diagrams, sound, animation, video, etc. in a visually integrated form. Approaches to the classification of multimedia learning tools are revealed. Special features, advantages of multimedia, expediency of use and their disadvantages are highlighted. A comprehensive analysis of the capabilities of multimedia teaching tools gave grounds for identifying the didactic functions that they perform. Several areas of multimedia application are described. Multimedia technologies make it possible to implement several basic methods of pedagogical activity, which are traditionally divided into active and passive principles of student interaction with the computer, which are revealed in the article. Important conditions for the implementation of multimedia technologies in the educational process are indicated. The feasibility of using multimedia in education is illustrated by examples. Of particular importance in education are game forms of learning, in the implementation of which educational elements based on media material play an important role. The influence of the game on the development of attention by means of works of media culture, which are very diverse in form and character, is shown. The importance of the role of multimedia in student education is indicated. In the educational process of multimedia students, a number of educational functions are implemented, which are presented in the article. Recommendations for using multimedia are given.

• 한국초등과학교육학회지:초등과학교육
• /
• 제33권3호
• /
• pp.491-509
• /
• 2014

The purpose of this study was to compare and analyze the elementary science curriculum of Finland, which ranked at the first place in the science domain of Programme for International Student Assessment (PISA), a comparative study on the academic achievements of OECD member nations, for three consecutive years and recorded a high national competitiveness index, and that of South Korea, thus providing information needed to develop curriculums and textbooks in the middle of occasional curriculum revisions and giving useful implications for the implementation of curriculum in the field of education in South Korea. The research findings were as follows: First, as for the science content areas based on the ratio of large area items according to the evaluation and analysis framework of TIMSS 2007, South Korea's ratio of life, chemistry, physics, and earth hardly showed fitness for TIMSS 2007 and exhibited equal distribution among the areas. In Finland, the ratio of life, chemistry, physical, and earth was similar to the fourth grade level of TIMSS 2007. The country showed differential distribution with life accounting for the highest percentage. Second, as for the cognitive domains, South Korea showed a high percentage in "Uses and Procedures of Tools" of "1. Knowing" and "Making Connections," "Comparison/Contrast/Classification," and "Uses of Models" of "2. Application." Finland recorded a high percentage in "Information Interpretation" of "2. Application." While South Korea focused on the uses and methods of scientific instruments during scientific activities, Finland made an approach with a focus on problems related to daily life such as the interpretation of information including reports and graphs from an interpretative perspective.