• Annual Conference on Human and Language Technology
• /
• 2004.10d
• /
• pp.73-81
• /
• 2004

한국어에서의 품사 결정 문제는 형태론적 중의성 문제도 있지만, 영어에는 발생하지 않는 동품사 중의성 문제로 더 까다롭다. 이러한 문제들은 어휘 문맥을 고려하지 않고서는 해결하기 어렵다. 통계 자료 부족 문제에 쉽게 대처하는 모델이 필요하며 문맥에 따른 품사를 결정하고자 할 때 서로 다른 형태의 여러 가지 어휘 문맥 정보를 반영할 수 있는 모델이 필요하다. 본 논문에서는 이런 점에 가장 적합한 최대 엔트로피(maximum entropy : ME) 모델을 품사태깅 작업에 이용하는 문제에 대해 다룬다. 어휘 문맥 정보를 이용하기 위한 자질함수가 매우 많아지는 문제에 대처하기 위해 필요에 따라 어휘 문맥 정보를 사전화 한다. 본 시스템의 특징으로는 어절 단위 품사 태깅을 위한 처리 기법. 어절의 형태소 분석열에 대한 어절 내부 확률 계산. ME 모델의 정규화 과정 생략에 의한 성능 향상, 디코딩 경로의 확장과 같은 점들이 있다. 실험을 통하여 본 연구의 기법이 높은 성능의 시스템을 달성할 수 있음을 알게 되었다.

• Korean Institute of Interior Design Journal
• /
• v.13 no.2
• /
• pp.10-19
• /
• 2004

In order to attain unique aesthetic value, diversity of meanings of Fusion Design, 'openness', requires meaning interpreting process(communication) which harmonizes and combines new codes suggested by authors(senders) and codes interpreted by public(receivers). For this harmony, we should consider all the art aspects(integration of synchronic and diachronic, anti-rationalism ideology, negative concept) that are bases of creativity of Fusion Aesthetics. Moreover, we can approach open thought system of Fusion Design through theories of Entropy and Rhizome. This point of view which allowed a categorical means of natural science theory to be used in design context changed a methodic or interpretative perception. Not only in art interpreting process does Fusion Design take openness as one of its necessary factors but also in creating process, which enables a work of art to be opened at the maximum. In conclusion, the significance of fusion design is to suggest an ideal model of communication abstracted from open art works.

• Ocean and Polar Research
• /
• v.41 no.3
• /
• pp.183-191
• /
• 2019

This study sought to find which model is most appropriate for estimating potential yield in the East Sea, Republic of Korea. For comparison purposes, the Process-error model, ASPIC model, Maximum entropy model, Observation-error model, and Bayesian state-space model were applied using data from catch amounts and total efforts of the whole catchable fishes in the East Sea. Results showed that the Bayesian state-space model was estimated to be the most reliable among the models. Potential yield of catchable species was estimated to be 227,858 tons per year. In addition, it was analyzed that the amount of fishery resources in 2016 was about 63% of the biomass that enables a fish stock to deliver the maximum sustainable yield.

• 한국어정보학회:학술대회논문집
• /
• 2016.10a
• /
• pp.249-251
• /
• 2016

다중 화자 대화 시스템에서, 시스템의 입장에서 어느 시점에 참여해야하는지를 아는 것은 중요하다. 이러한 참여 모델을 구축함에 있어서 본 연구에서는 다수의 화자가 대화에 참여하는 영화 대본으로 구축된 MovieDic 말뭉치를 사용하였다. 구축에 필요한 자질로써 의문사, 호칭, 명사, 어휘 등을 사용하였고, 훈련 알고리즘으로는 Maximum Entropy Classifier를 사용하였다. 실험 결과 53.34%의 정확도를 기록하였으며, 맥락 자질의 추가로 정확도 개선을 기대할 수 있다.

• Journal of Korea Water Resources Association
• /
• v.42 no.7
• /
• pp.537-546
• /
• 2009

Determination of optimal pressure monitoring location is essential to manage water distribution system efficiently and safely. In this study, entropy theory is applied to overcome defects of previous researches about determining the optimal sensor location. The previous studies required the calibration using historical data, therefore, it was difficult to apply the proposed method in the place where the enough data were not available. Also, most researches have focused on the locations to minimize cost and maximize accuracy of the model, which is not appropriate for the purpose of maintenance of the water distribution system. The proposed method in this study quantify the entropy which is defined as the amount of information calculated from the pressure change due to the variation of discharge. When abnormal condition is occurred in a node, the effect on the entire network is presented by the entropy, and the emitter is used to reproduce actual pressure change pattern in EPANET. The optimal location to install pressure sensors in water distribution system is the nodes having the maximum information from other nodes. The looped and branched networks are evaluated using the proposed model. As a result, entropy theory provides general guideline to select the locations to install pressure sensors and the results can be used to help decision makers.

• Journal of Species Research
• /
• v.12 no.1
• /
• pp.109-113
• /
• 2023

To build a distribution model for Haemadipsa rjukjuana, we collected current occurrences of the species on Heuksando with adjacent islands. Based on current locations and 19 climate variables with DEM (digital elevation model), we built the MaxEnt (maximum entropy) species distribution model for H. rjukjuana in the islands. Then, we applied the MaxEnt model to the mainland of Korea with the current climate condition and topology. In addition to the current distribution scenario, we predicted the future distribution scenarios in Korea by Coupled Model Intercomparison Project Phase 6 (CMIP6) global climate models. Shared Socioeconomic Pathway (SSP) 585 of two CMIP6 models(GISS-E2-1 and INM-CM4-8) from 2040 to 2100 were used for the future projection.

• Journal of the Korean Society of Environmental Restoration Technology
• /
• v.23 no.5
• /
• pp.45-58
• /
• 2020

This study was conducted to comprehend the spatial distribution characteristics, habitats and appearances of Hydropotes inermis by using the biotope mapping in Daebudo Island, Ansan-si. The result is base data to understand status and manage potential inhabitation of Hydropotes inermis in Daebudo Island through the Maximum Entropy model. The study used 105 traces from the primary investigation and 452 traces in the secondary investigation. The biotope types were distinquished Hydropotes inermis habitats largest from the order of natural forest (15.1%), natural coast (13.7%), marshy cultivated land (12.6%), and dry cultivated land (11.7%), and from the inhabitation trace results. Hydropotes inermis appearanced biotope types were the greatest in the order of cultivated land (49.73%) > forest (18.85%) > coast (7.00%) > grassland (6.28%). Since forests in Daebudo Island have low slope and altitude, it was concluded that Hydropotes inermis would live in most of the forests. A high number of Hydropotes inermis was found to appear in areas where the grassland is formed including cultivated lands (include unused paddies and fields) and marshy grasslands, which would result in direct damage of crops. According to the Maxent modeling analysis that used location information of Hydropotes inermis, the AUC value was 0.635 based on the ROC curve. In Daebudo Island, areas with over 0.635 potential inhabitation value are distributed all over the place, and it was concluded that each population would have a different scope of influence and home range. Hydropotes inermis living in Daebudo Island have high habitat suitability mainly around the cultivated lands near the roads, but due to the bare lands and roads, it is expected that their habitats would be fragmented and damaged, which would have a direct and indirect effect in maintaining the Hydropotes inermis population. Also, considering habitat disturbance, diverse methods for reducing damage including capturing some individuals within the limit that does not disperse Hydropotes inermis population in Daebudo Island must be carried out.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.54 no.3
• /
• pp.217-223
• /
• 2018

Overfishing capacity has become a global issue due to over-exploitation of fisheries resources, which result from excessive fishing intensity since the 1980s. In the case of Korea, the fishing effort has been quantified and used as an quantified index of fishing intensity. Fisheries resources of coastal fisheries in the Korean waters of the East Sea tend to decrease productivity due to deterioration in the quality of ecosystem, which result from the excessive overfishing activities according to the development of fishing gear and engine performance of vessels. In order to manage sustainable and reasonable fisheries resources, it is important to understand the fluctuation of biomass and predict the future biomass. Therefore, in this study, we forecasted biomass in the Korean waters of the East Sea for the next two decades (2017~2036) according to the changes in fishing intensity using four fishing effort scenarios; $f_{current}$, $f_{PY}$, $0.5{\times}f_{current}$ and $1.5{\times}f_{current}$. For forecasting biomass in the Korean waters of the East Sea, parameters such as exploitable carrying capacity (ECC), intrinsic rate of natural increase (r) and catchability (q) estimated by maximum entropy (ME) model was utilized and logistic function was used. In addition, coefficient of variation (CV) by the Jackknife re-sampling method was used for estimation of coefficient of variation about exploitable carrying capacity ($CV_{ECC}$). As a result, future biomass can be fluctuated below the $B_{PY}$ level when the current level of fishing effort in 2016 maintains. The results of this study are expected to be utilized as useful data to suggest direction of establishment of fisheries resources management plan for sustainable use of fisheries resources in the future.

• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.1
• /
• pp.110-114
• /
• 2010

We describe a new method for text categorization using TextRank algorithm. Text categorization is a problem that over one pre-defined categories are assigned to a text document. TextRank algorithm is a graph-based ranking algorithm. If we consider that each word is a vertex, and co-occurrence of two adjacent words is a edge, we can get a graph from a document. After that, we find important words using TextRank algorithm from the graph and make feature which are pairs of words which are each important word and a word adjacent to the important word. We use classifiers: SVM, Na$\ddot{i}$ve Bayesian classifier, Maximum Entropy Model, and k-NN classifier. We use non-cross-posted version of 20 Newsgroups data set. In consequence, we had an improved performance in whole classifiers, and the result tells that is a possibility of TextRank algorithm in text categorization.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.31 no.1 s.256
• /
• pp.55-61
• /
• 2007

Recently simulation model becomes an essential tool for analysis and design of a system but it is often expensive and time consuming as it becomes complicate to achieve reliable results. Therefore, high-fidelity simulation model needs to be replaced by an approximate model, the so-called metamodel. Metamodeling techniques include 3 components of sampling, metamodel and validation. Cross-validation approach has been proposed to provide sequnatially new sample point based on cross-validation error but it is very expensive because cross-validation must be evaluated at each stage. To enhance the cross-validation of metamodel, sequential sampling method using candidate points and representative cross-validation is proposed in this paper. The candidate and representative cross-validation approach of sequential sampling is illustrated for two-dimensional domain. To verify the performance of the suggested sampling technique, we compare the accuracy of the metamodels for various mathematical functions with that obtained by conventional sequential sampling strategies such as maximum distance, mean squared error, and maximum entropy sequential samplings. Through this research we team that the proposed approach is computationally inexpensive and provides good prediction performance.