• Proceedings of the Korean Society for Language and Information Conference
• /
• 2007.11a
• /
• pp.422-429
• /
• 2007

The rapid growth of the online information services causes the problem of information explosion. Automatic text summarization techniques are essential for dealing with this problem. There are different approaches to text summarization and different systems have used one or a combination of them. Considering the wide variety of summarization techniques there should be an evaluation mechanism to assess the process of summarization. The evaluation of automatic summarization is important and challenging, since in general it is difficult to agree on an ideal summary of a text. Currently evaluating summaries is a laborious task that could not be done simply by human so automatic evaluation techniques are appearing to help this matter. In this paper, we will take a look at summarization approaches and examine summarizers' general architecture. The importance of evaluation methods is discussed and the need to find better automatic systems to evaluate summaries is studied.

• Proceedings of the Korean Information Science Society Conference
• /
• 2011.06c
• /
• pp.168-171
• /
• 2011

One important issue in semantic web is identification and selection of domain concepts for domain ontology learning when several hundreds or even thousands of terms are extracted and available from relevant text documents shared among the members of a domain. We present a novel domain concept acquisition and selection approach for ontology learning that uses affinity propagation algorithm, which takes as input semantic and structural similarity between pairs of extracted terms called data points. Real-valued messages are passed between data points (terms) until high quality set of exemplars (concepts) and cluster iteratively emerges. All exemplars will be considered as domain concepts for learning domain ontologies. Our empirical results show that our approach achieves high precision and recall in selection of domain concepts using less number of iterations.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.10b
• /
• pp.277-279
• /
• 2001

최근 디지털 비디오 데이터의 사용이 급격히 증가하면서 저급 수준의 정보를 이용하여 고급 수준의 내용 정보를 자동으로 추출하는 기술이 필요하게 되었다. 축구와 같은 분야에서는 그 중에서도 골, 프리킥, 파울 장면 등의 고급 수준 내용 정보가 중요한 의미를 갖게 되는데 특히, 이러한 장면 중 중요하다고 여기는 장면은 재연 장면을 통하여 다시 시청자에게 보여주게 되며, 축구 비디오에 대한 요약에서는 이런 장면들이 꼭 포함되어야 한다. 본 논문에서는 이러한 축구 비디오 데이터에서 재연 장면을 자동으로 추출하는 방법을 제안한다. 기본적으로는 축구 고유의 특징들을 이용하는데 첫 번째 특징은 샷의 길이가 너무 짧거나 너무 길지 않다는 것이고, 두 번째 특징은 재연 장면이라는 것은 장면이 느리게 다시 재생되는 것이기 때문에 움직임 특징이 일반적인 장면과는 다르다는 것이다. 본 논문에서는 오브젝트의 움직임을 구분하기 위하여 재연 장면을 두 가지 종류로 나누었다. 하나는 확대 상태의 재연 장면이고 다른 하나는 축소 상태의 재연 장면이다. 본 논문의 알고리즘을 적용하여 실험한 결과 Recall과 precision 모두 77% 이상 나오는 것을 알 수 있었다.

• Journal of the Korea Society of Computer and Information
• /
• v.22 no.7
• /
• pp.125-131
• /
• 2017

It is not easy for the user to find the information that is appropriate for the user among the suddenly increasing information in recent years. One of the ways to help individuals make decisions in such a lot of information is the recommendation system. Although there are many recommendation methods for such recommendation systems, a representative method is collaborative filtering. In this paper, we design and implement the movie recommendation system on user-based collaborative filtering of apache mahout. In addition, Pearson correlation coefficient is used as a method of measuring the similarity between users. We evaluate Precision and Recall using the MovieLens 100k dataset for performance evaluation.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2016.04a
• /
• pp.818-819
• /
• 2016

지오코딩(Geocoding)은 자연어로 표현된 주소를 컴퓨터가 인지 가능한 (x,y) 좌표로 변환하는 과정이며, 지리정보 분석 등 다양한 영역의 필수적인 전처리 과정에서 사용된다. 현재 국내 주소를 지오코딩하는 API를 제공하는 서비스 프로바이더는 다수 존재하나, 성능 향상의 여지가 남아있는 현황이다. 본 연구에서는 지오코딩 위치정확도의 향상을 위해 Euclidean/Edit distance 기반 앙상블(Ensemble) 지오코딩 알고리즘(EEE-Geocoding)을 제안하였다. 화학물질 보유 업체 5569개소의 주소를 토대로 제안 기법에 대한 성능평가 실험을 진행하였으며, 평가결과는 0.99 precision, 0.87 recall, 0.92 F1 score 이었다.

• Journal of the Semiconductor & Display Technology
• /
• v.7 no.4
• /
• pp.57-62
• /
• 2008

Image retrieval is one of the most exciting and rapidly growing research issues in the field of multimedia technology. This paper proposes a new method that performs search the relevant images by using query-by-example. The proposed method for search and retrieval of images utilizes the location information where the image had been taken. The system associates the photo images with their corresponding GPS coordinates that are used as metadata for searching. Experimental results show that the proposed method demonstrates better performance improving up to 59% of average recall and 49% of average precision. Moreover, we learned from the experimental results geo-location information embedded within the image header is more effective and positive on the search and storage.

• Journal of the Korean Society for information Management
• /
• v.10 no.2
• /
• pp.43-67
• /
• 1993

The primary purpose of the study is to improve the low preclslon often found In full-text retrleval systems. In order to enhance the low precision of full-text retrleval wh~le retaining ~ t s hgh recall, relevance feedback mechanisms based on probabilistic retrieval models (binary independence and two-Polsson Independence models) were employed. Thls paper investigates the effect of relevance feedback on the performance of full-text retrieval systems.

• Journal of the Korean Society for information Management
• /
• v.1 no.1
• /
• pp.43-62
• /
• 1984

The present study has two main objectives. The first objective is to devise a new term weighting technique which can be used to weight the significance value of each word stem in a test collection of documents on the subject of "enteral hyperalimentation." The next objective is to evaluate retrieval performance of proposed term weighting technique, together with four other term weighting techniques, by conducting a set of experiments. The experimental results have shown that the performance of Sparck Jones's inverse document frequency weighting and the proposed term significance weighting techniques produced better recall and precision ratios than the other three complex weighting techniques.

• International Journal of Contents
• /
• v.11 no.2
• /
• pp.1-8
• /
• 2015

In this paper, we describe a novel method for document layout analysis that is based on a Fuzzy Energy Matrix (FEM). A FEM is a two-dimensional matrix that contains the likelihood of text and non-text and is generated through the use of Fuzzy theory. The key idea is to define an Energy map for the document to categorize text and non-text. The proposed mechanism is designed for execution with a low-resolution document image, and hence our method has a fast processing speed. The proposed method has been tested on public ICDAR 2009 datasets to conduct a comparison against other state-of-the-art methods, and it was also tested with Korean documents. The results of the experiment indicate that this scheme achieves superior segmentation accuracy, in terms of both precision and recall, and also requires less time for computation than other state-of-the-art document image analysis methods.

• ETRI Journal
• /
• v.32 no.3
• /
• pp.490-492
• /
• 2010

This letter proposes a robust feature extraction method using a spatially normalized mean for temporal ordinal measurement. Before computing a rank matrix from the mean values of non-overlapped blocks, each block mean is normalized so that it obeys the invariance property against linear additive and subtractive noise effects and is insensitive against multiplied and divided noise effects. Then, the temporal ordinal measures of spatially normalized mean values are computed for the feature matching. The performance of the proposed method showed about 95% accuracy in both precision and recall rates on various distortion environments, which represents the 2.7% higher performance on average compared to the temporal ordinal measurement.