• Proceedings of the IEEK Conference
• /
• 2002.06d
• /
• pp.203-206
• /
• 2002

In this paper, we present a sentence speech recognizer for twenty questions game. The proposed approaches for speaker-independent sentence speech recognition can be divided into two steps. One is extraction of the number of syllables in eojeol for candidate reduction, and the other is knowledge based language model for sentence recognition. For twenty questions game, we implemented speech recognizer using 956 sentences and 1095 eojeols. The results obtained in our experiments were 87% sentence recognition rate and 90.15% eojeol recognition rate.

• Journal of the Korean Institute of Navigation
• /
• v.22 no.4
• /
• pp.37-44
• /
• 1998

This paper presents a procedure for automated extraction of ridge shadows in noisy gray images. This procedure mainly consists of 1) a deformable model which is designed basing upon the knowledge about the shape of shadows and is expected to be useful in extracting ridge shadows especially located in low signal to noise ratio background, and 2) the scale space scheme which is also useful even if there is less information about the size and the positions of ridge shadows in advance. This procedure is applied to artificial images and its performance is evaluated experimentally.

• Proceedings of the KSRS Conference
• /
• 2003.11a
• /
• pp.210-212
• /
• 2003

Active sensors are rapidly conquering a share on the remote sensing market and offer among others new possibilities toward automatically acquiring 3D building data. Better dissemination of information about new technological developments can possibly be achieved by short distance-learning courses. The paper describes the didactic and technical aspects of a course we have designed and conducted on airborne laser scanning and interferometric SAR. The building extraction application is a good example to illustrated the added value of short electronic-learning courses above simply publishing (digital) papers.

• Journal of KIISE:Software and Applications
• /
• v.27 no.11
• /
• pp.1082-1087
• /
• 2000

신경회로망 구조의 정제(精製)는 회로망의 일반화능력이나 효율성의 관점에서 중요한 문제이다. 본 논문에서는 feed-forward neural networks로부터 은닉지식을 추출하는 방법을 사용하여 네트워크 재구성을 통한 정제방법을 제안한다. 먼저, 효율적인 if-then rule 추출방법을 제시하고 그 추출된 룰들을 사용하여 룰기반 네트워크로 변환하는 과정을 보여준다. 생성된 룰기반 네트워크 fully connected network에 비하여 상당히 축소된 연결 복잡도를 가지게 되며 일반적으로 더 우수한 일반화능력을 가지게 된다. 본 연구는 도메인 지식이 없이 데이타만 사용하여 어떻게 정제된 룰기반 신경망회로를 생성하고 있는가를 보여준다. 도메인 데이타들에 대한 실험결과도 제시하였다.

• The Transactions of the Korea Information Processing Society
• /
• v.5 no.11
• /
• pp.2786-2796
• /
• 1998

Extraction of interesting and general knowledge from large spatial database is an important task in the development of geographical information system and knowledge-base systems. In this paper, we propose a spatial data mining system using generalization method; In this system, we extend an existing generalization mining and design a rulebase to support deriving new spatial knowledge. For this purpose, we propose an interleaved method which integrates spatial data dominated and nonspatial data dominated mining and construct a rulebase to extract topological relationship between spatial objects.

• Journal of Korean Medical classics
• /
• v.25 no.4
• /
• pp.105-113
• /
• 2012

Objectives : Acquiring information from symptoms is one of the important method to gain clinically available information in korean medicine. Therefore, up to now, study of symptom terms was frequently implemented in promotion of various information project. In data extraction methods using symptom information from DB, information search using synonym and method using ontology is studied and utilized. However, considering concept of symptom has essential information of appeared body area and phenomenon we think that extending synonym and ontology relationship in symptom terms can be useful for search and set to this study. Methods : We collect terms relevant to human body area and structure described in Donguibogam. Synonymous relationship between collected terms is organized. Relationship between collected terms is build to human-body-knowledge table which has form of Concept+Relation+Concept. Type of relationship is limited on a range of expressing content about parts of human body. Result & Conclusion : Search condition is generated automatically using relationship of the upper area in knowledge table contents. Information of next and previous acupuncture point, upper and lower acupuncture point, left and right acupuncture point can be searched using information of acupuncture point location, order, relative position in area, direction in knowledge table contents.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.31B no.8
• /
• pp.137-147
• /
• 1994

In this paper,we propose a knowledge-based system for locating Destination Address Block(DAB) by analyzing the structure of Korean envelope images. In the proposed system the preprocessing steps such as adaptive binarization connected component extraction and deskewing are carried out first for the effective structure analysis of the envelope image. Then DAB containing address name and zipcode parts of the input envelope image is extracted by an iterative procedure based on the knowledge acquired from the statistical feature analysis of the various envelope images. Most of the system for slocating address blocks on envelopes have extracted DAB by segmenting an envelope image into several candidate blocks followed by selecting one among the candidate blocks. Because it is very difficult to segment a Korean envelope image into several blocks due to the specific writing habits that the addresses on the envelope are written in close proximity to each other the proposed iterative procedure determines DAB by splitting or merging the connected components and verifies the determined DAB without segmentation and selection. Experiments with a great number of the live envelopes provided from Seoul Mail Center in Koorea were carried out. The results reveal that the proposed system is very effective for address block location on Korean envelopes.

• Journal of KIISE
• /
• v.42 no.6
• /
• pp.817-822
• /
• 2015

Relation extraction is an important information extraction technique that can be widely used in areas such as question-answering and knowledge population. Previous studies on relation extraction have been based on supervised machine learning models that need a large amount of training data manually annotated with relation categories. Recently, to reduce the manual annotation efforts for constructing training data, distant supervision methods have been proposed. However, these methods suffer from a drawback: it is difficult to use these methods for collecting negative training data that are necessary for resolving classification problems. To overcome this drawback, we propose a one-class classification model that can be trained without using negative data. The proposed model determines whether an input data item is included in an inner category by using a similarity measure based on lexical information and syntactic patterns in a vector space. In the experiments conducted in this study, the proposed model showed higher performance (an F1-score of 0.6509 and an accuracy of 0.6833) than a representative one-class classification model, one-class SVM(Support Vector Machine).

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.40 no.1
• /
• pp.10-19
• /
• 2003

Caption information in news videos can be useful for video indexing and retrieval since it usually suggests or implies the contents of the video very well. In this paper, a new algorithm for extracting and recognizing characters from news video is proposed, without a priori knowledge such as font type, color, size of character. In the process of text region extraction, in order to improve the recognition rate for videos with complex background at low resolution, continuous frames with identical text regions are automatically detected to compose an average frame. The image of the averaged frame is projected to horizontal and vertical direction, and we apply region filling to remove backgrounds to produce the character. Then, K-means color clustering is applied to remove remaining backgrounds to produce the final text image. In the process of character recognition, simple features such as white run and zero-one transition from the center, are extracted from unknown characters. These feature are compared with the pre-composed character feature set to recognize the characters. Experimental results tested on various news videos show that the proposed method is superior in terms of caption extraction ability and character recognition rate.

• Journal of Korea Spatial Information System Society
• /
• v.8 no.2 s.17
• /
• pp.39-52
• /
• 2006

With the recent the use of spatio-temporal data mining which can extract various knowledge such as movement patterns of moving objects in history data of moving object gets increasing. However, the existing movement pattern extraction methods create lots of candidate movement patterns when the minimum support is low. Therefore, in this paper, we suggest the STMPE(Spatio-Temporal Movement Pattern Extraction) algorithm in order to efficiently extract movement patterns of moving objects from the large capacity of spatio-temporal data. The STMPE algorithm generalizes spatio-temporal and minimizes the use of memory. Because it produces and keeps short-term movement patterns, the frequency of database scan can be minimized. The STMPE algorithm shows more excellent performance than other movement pattern extraction algorithms with time information when the minimum support decreases, the number of moving objects increases, and the number of time division increases.