• Proceedings of the Korea Information Processing Society Conference
• /
• 2013.11a
• /
• pp.1510-1512
• /
• 2013

3D 영화 산업 발전과 함께 4D 영화에 관한 수요가 증가했지만, 4D 영화 제작은 복잡한 시스템과 오랜 제작 시간을 요구한다. 본 논문에서는 기존의 4D 영화 제작 방식의 문제점을 해결하기 위해 자막과 시나리오 정보를 이용한 4D 이벤트 추출 방법을 제안한다. 자막과 시나리오의 문장으로 부터 4D 이벤트와 관련된 특징을 추출하고 4D 사전과 비교하여 4D 이벤트를 분류한다. 또한, 자막과 시나리오 각각의 4D 이벤트 추출 성능을 비교한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.11a
• /
• pp.626-628
• /
• 2021

사건 관계 추출 태스크는 구조화되지 않은 텍스트 데이터에서 사건의 구조화된 표현을 얻는 것이다. 하나의 문장에서도 많은 정보를 얻을 수 있는 중요한 태스크임에도 불구하고, 다양한 사전 학습 모델을 적용한 연구는 아직 활발하게 연구되지 않고 있다. 따라서 본 연구에서 사전 학습된 모델의 임베딩 기법 중 BERT, RoBERTa, SpanBERT에 각각 base, large 아키텍처를 적용하여 실험하였다. 사건을 식별하기 위한 trigger와 해당 trigger의 세부 argument를 식별하기 위한 분류기를 상위레이어로 각각 설계하였고, 다양한 배치 크기를 적용하여 실험하였다. 성능평가는 trigger/argument 각각 F1 score를 적용하였고, 결과는 RoBERTa large 모델에서 좋은 성능을 보인 것을 확인하였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.11a
• /
• pp.490-492
• /
• 2022

최근 사법분야에 지능형 법률 서비스를 제공하게 되면서 학습데이터로서 판결문의 중요성이 높아지고 있다. 그중 범죄사실은 수사자료와 유사하여 범죄수사에 귀중한 자료역할을 하고 있지만, 주체가 생략되거나 긴 문장의 형태로 인해 구성요건을 추출하고 사건의 인과관계 파악이 어려울 수 있어 이를 분석하는데 적지 않은 시간과 인력이 소비될 수밖에 없다. 따라서, 본 논문에서는 사전학습모델을 활용한 개체명 인식과 형태소 분석기반 이벤트 추출기법을 범죄사건 재구성에 적용하여 핵심 사건추출을 간편화하고 시각적으로 표현해 전체적인 사건 흐름 이해도를 향상할 수 있는 방법론을 제안하고자 한다.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.15 no.1
• /
• pp.1-8
• /
• 2020

If a dangerous situation arises in a place where out of reach from the human, UAVs can be used to determine the size and location of the situation to reduce the further damage. With this in mind, this paper sets the minimum value of the roll, pitch, and yaw using beta flight to detect the UAV's smooth hovering, integration, and derivative (PID) values to ensure that the UAV stays horizontal, minimizing errors for safe hovering, and the camera uses Open CV to install the Raspberry Pi program and then HSV (color, saturation, Brightness) using the color palette, the filter is black and white except for the red color, which is the closest to the fire we want, so that the UAV detects the image in the air in real time. Finally, it was confirmed that hovering was possible at a height of 0.5 to 5m, and red color recognition was possible at a distance of 5cm and at a distance of 5m.

• Journal of the Korean Society for information Management
• /
• v.23 no.1 s.59
• /
• pp.83-98
• /
• 2006

This study proposes a template-based method of automatic summarization of multiple news articles using the semantic categories of sentences. First, the semantic categories for core information to be included in a summary are identified from training set of documents and their summaries. Then, cue words for each slot of the template are selected for later classification of news sentences into relevant slots. When a news article is input, its event/accident category is identified, and key sentences are extracted from the news article and filled in the relevant slots. The template filled with simple sentences rather than original long sentences is used to generate a summary for an event/accident. In the user evaluation of the generated summaries, the results showed the 54.l% recall ratio and the 58.l% precision ratio in essential information extraction and 11.6% redundancy ratio.

• Journal of KIISE:Databases
• /
• v.32 no.6
• /
• pp.603-615
• /
• 2005

As recent business environments are changed and become complex, a more efficient and effective business process management are needed. This paper proposes a new approach to the automatic execution of business processes using Event-Condition-Action (ECA) rules that can be automatically triggered by an active database. First of all, we propose the concept of blocks that can classify process flows into several patterns. A block is a minimal unit that can specify the behaviors represented in a process model. An algorithm is developed to detect blocks from a process definition network and transform it into a hierarchical tree model. The behaviors in each block type are modeled using ACTA formalism. This provides a theoretical basis from which ECA rules are identified. The proposed ECA rule-based approach shows that it is possible to execute the workflow using the active capability of database without users' intervention.

• KIPS Transactions on Software and Data Engineering
• /
• v.1 no.1
• /
• pp.69-74
• /
• 2012

Many internet users attempt to focus on the issues which have posted on social network services in a very short time. When some social big issue or event occurred, it will affect the number of comments and retweet on that day in twitter. In this paper, we propose the method of extracting core events based on timeline analysis, sentiment feature and retweet information in twitter data. To validate our method, we have compared the methods using only the frequency of words, word frequency with sentiment analysis, using only chi-square method and using sentiment analysis with chi-square method. For justification of the proposed approach, we have evaluated accuracy of correct answers in top 10 results. The proposed method achieved 94.9% performance. The experimental results show that the proposed method is effective for extracting core events in twitter corpus.

• Journal of the Korea Society of Computer and Information
• /
• v.17 no.2
• /
• pp.139-147
• /
• 2012

In this paper, we discuss the structure of biological knowledge discovery system based on text mining and automatic inference. Given a set of biology documents, the system produces a new hypothesis in an integrated manner. The text mining module of the system first extracts the 'event' information of predefined types from the documents. The inference module then produces a new hypothesis based on the extracted results. Such an integrated system can use information more up-to-date and diverse than other automatic knowledge discovery systems use. However, for the success of such an integrated system, the precision of the text mining module becomes crucial, as any hypothesis based on a single piece of false positive information would highly likely be erroneous. In this paper, we propose a probabilistic filtering method that filters out false positives from the extraction results. Our proposed method shows higher performance over an occurrence-based baseline method.

• Journal of the Economic Geographical Society of Korea
• /
• v.18 no.1
• /
• pp.76-86
• /
• 2015

In this study, we presented a method for estimating the potential population damage of the Seoul Nowon-gu area in the event of a terrorist using a vehicle improvised explosive devices (IED). Using the object-based building extraction method with orthophoto image, the area of the apartment has been determined, and the apartment's height and level were estimated based on the elevation data. Using the population estimation method based on total floor area of building, each apartment resident population was estimated, and then potential population damage at the time of terrorist attacks was estimated around the subway station through a scenario analysis. Terrorism damage using IED depends on the type of vehicle greatly because of the amount loadable explosives. Therefore, potential population damage was calculated based on the type of vehicle. In the results, the maximum potential damage population during terrorist attacks has been estimated to occur around Madeul station, Nowon-gu. The method used in this study can be used various population estimation research and disaster damage estimation.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.23 no.6
• /
• pp.711-718
• /
• 2019

BDD is a well-known alternative to the conventional Boolean logic method in fault tree analysis. As the size of fault tree increases, the calculation time and computer resources for BDD dramatically increase. A new failure path search and path restructure method is proposed for efficient calculation of CS and MCS from BDD. Failure path grouping and bottom-up path search is proved to be efficient in failure path search in BDD and path restructure is also proved to be used in order to reduce the number of CS comparisons for MCS extraction. With these newly proposed methods, the top event probability can be calculated using the probability by ASDMP(Approximate Sum of Disjoint MCS Products), which is shown to be equivalent to the result by the conventional MCUB(Minimal Cut Upper Bound) probability.