• Journal of Teachers' Pension
• /
• v.1
• /
• pp.157-185
• /
• 2016

본 연구에서는 비정상적 사건을 정의하고 이에 따른 비정상적 위험의 구체적인 유형을 파악하며, 이와 관련된 사학연금의 위험관리 체계에 대한 검토와 함께 비정상적 위험에 효과적으로 대응할 수 있는 자산운용방안을 제시하였다. 우선 비정상적 사건을 '과거 자료를 이용한 발생확률의 추정이나 발생여부에 대한 예측이 불가능하며 따라서 이의 발생 가능성을 사전에 고려하고 대비하는 사전적인 대처가 어려운 사건으로서 자산운용과 위험관리에 무시할 수 없는 영향을 미치는 사건'으로 정의하였으며, 이의 구체적인 형태로서 금융위기를 포함하는 9가지 사건 유형을 파악하였다. 동비정상적 사건들은 포트폴리오 투자를 통한 자산운용에서 개별자산군의 기대수익률과 위험 및 자산군 사이의 상관관계에 영향을 미쳐, 기존의 자산배분안의 최적성을 상실시키고 위험수준의 측정치인 VaR값을 과소 또는 과대추정하게 할 수 있는 것으로 분석되었다. 한편 비정상적 사건의 해외 사례에 대한 분석에서는 비정상적 사건의 영향이 개별 사건마다 다양한 양태로 발현되는 것이 관측되었다. 본 연구에서는 사학연금의 현행 자산배분 체계가 이와 같은 비정상적 사건의 영향에 적절하게 대응하기 어려운 상황이라고 진단하였으며, 비정상적 사건에 적절히 대응하기 위한 자산관리방안의 일환으로서 일별 수익률 자료를 사용한 비정상적 사건의 영향 평가방안을 제시하였다. 한편, 사학연금의 현행 위험관리 체계는 비정상적 사건의 발생에 적절하게 대응할 수 있는 것으로 평가되었다

• The KIPS Transactions:PartB
• /
• v.12B no.7 s.103
• /
• pp.829-840
• /
• 2005

Word Sense Disambiguation(WSD) is one of the most difficult problem in Korean information processing. We propose a WSD model with the capability to filter semantic information using the specific characteristics in dictionary dictions, and nth added information, useful to sense determination, such as statistical, distance and case information. we propose a model, which can resolve the issues resulting from the scarcity of semantic information data based on the word hierarchy system (thesaurus) developed by Ulsan University's UOU Word Intelligent Network, a dictionary-based toxicological database. Among the WSD models elaborated by this study, the one using statistical information, distance and case information along with the thesaurus (hereinafter referred to as 'SDJ-X model') performed the best. In an experiment conducted on the sense-tagged corpus consisting of 1,500,000 eojeols, provided by the Sejong project, the SDJ-X model recorded improvements over the maximum frequency word sense determination (maximum frequency determination, MFC, accuracy baseline) of $18.87\%$ ($21.73\%$ for nouns and inter-eojeot distance weights by $10.49\%$ ($8.84\%$ for nouns, $11.51\%$ for verbs). Finally, the accuracy level of the SDJ-X model was higher than that recorded by the model using only statistical information, distance and case information, without the thesaurus by a margin of $6.12\%$ ($5.29\%$ for nouns, $6.64\%$ for verbs).

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.8
• /
• pp.371-380
• /
• 2023

Relation extraction is to extract relationships between named entities from text. Traditionally, relation extraction methods only extract relations between predetermined subject and object entities. However, in end-to-end relation extraction, all possible relations must be extracted by considering the positions of the subject and object for each pair of entities, and so this method uses time and resources inefficiently. To alleviate this problem, this paper proposes a method that sets directions based on the positions of the subject and object, and extracts relations according to the directions. The proposed method utilizes existing relation extraction data to generate direction labels indicating the direction in which the subject points to the object in the sentence, adds entity position tokens and entity type to sentences to predict the directions using a pre-trained language model (KLUE-RoBERTa-base, RoBERTa-base), and generates representations of subject and object entities through probabilistic crossover operation. Then, we make use of these representations to extract relations. Experimental results show that the proposed model performs about 3 ~ 4%p better than a method for predicting integrated labels. In addition, when learning Korean and English data using the proposed model, the performance was 1.7%p higher in English than in Korean due to the number of data and language disorder and the values of the parameters that produce the best performance were different. By excluding the number of directional cases, the proposed model can reduce the waste of resources in end-to-end relation extraction.

• Korean Journal of Culture and Social Issue
• /
• v.11 no.1
• /
• pp.59-84
• /
• 2005

Three studies were conducted to examine the degree to which three common heuristics, anchoring heuristic, framing effect and representative-ness heuristic, influence the decision-making precesses of actual judges and 5-persons mock juries. With scenarios regarding various issues that are commonly raised in actual criminal and civil trials, study 1 examined the 158 actual judges' decisions. In study 2, the decisions of 80 mock jury groups that consisted of college students were examined with similar scenarios. And individual decisions were examined in study 3 to compare with the group decisions in study 2. The decision processes of the actual judges and the mock jury groups alike were found to be influenced by "anchors". But the biases by the anchoring heuristic were more pronounced in the group decisions than in the decisions of the actual judges. With respect to framing effect, the actual judges were found to be resistant, while a small effect was found in the decisions of mock jury groups. Representative-ness biases weren't found in the decisions of both the actual judges and mock juries. The implications of the results for judicial systems were discussed.

• Journal of KIISE:Databases
• /
• v.37 no.4
• /
• pp.177-184
• /
• 2010

In these day, many data including sensor, delivery, credit and stock data are generated continuously in massive quantity. It is difficult to learn from these data because they are large in volume and changing fast in their concepts. To handle these problems, learning methods based in sliding window methods over time have been used. But these approaches have a problem of rebuilding models every time new data arrive, which requires a lot of time and cost. Therefore we need very simple incremental learning methods. Bayesian method is an example of these methods but it has a disadvantage which it requries the prior knowledge(probabiltiy) of data. In this study, we propose a learning method based on attribute values. In the proposed method, even though we don't know the prior knowledge(probability) of data, we can apply our new method to data. The main concept of this method is that each attribute value is regarded as an expert learner, summing up the expert learners lead to better results. Experimental results show our learning method learns from data very fast and performs well when compared to current learning methods(decision tree and bayesian).

• Journal of Korea Water Resources Association
• /
• v.47 no.5
• /
• pp.435-446
• /
• 2014

Various drought indices developed from previous studies can not consider the inherent uncertainty of drought because they assess droughts using a pre-defined threshold. In this study, to consider inherent uncertainty embedded in monthly streamflow data, Hidden Markov Model (HMM) based drought index (HMDI) was proposed and then probabilistic assessment of hydrologic drought was performed using HMDI instead of using pre-defined threshold. Using monthly streamflow data (1966~2009) of Pyeongchang river and Upper Namhan river provided by Water Management Information System (WAMIS), applying the HMM after moving-averaging the data with 3, 6, 12 month windows, this study calculated the posterior probability of hidden state that becomes the HMDI. For verifying the method, this study compared the HMDI and Standardized Streamflow Index (SSI) which is one of drought indices using a pre-defined threshold. When using the SSI, only one value can be used as a criterion to determine the drought severity. However, the HMDI can classify the drought condition considering inherent uncertainty in observations and show the probability of each drought condition at a particular point in time. In addition, the comparison results based on actual drought events occurred near the basin indicated that the HMDI outperformed the SSI to represent the drought events.

• Journal of Korea Multimedia Society
• /
• v.13 no.2
• /
• pp.279-286
• /
• 2010

Unbalanced traffic load offered to the nodes making up a path in the network guaranteeing quality-of-service has been known as a main cause deteriorating the capability of the path in admitting traffic flows. Several path load balancing methods have been developed to resolve this problem which used a feedback control scheme that adjusts the delay budget of a flow allocated to each node according to the conditions of available resource in the path. Because of no consideration about the probabilistic characteristics of the service quality, it is impossible for them to prevent in advance a bottleneck on the path which leads to a native restriction in the improvement of the capability being deteriorated. This paper proposes a preventive intra-path load balancing method applicable to the RSVP system which is based on the probabilistic characteristics of the quality-of-service of the offered load. The results of the simulation of the proposed method on a simple evaluation network showed that it provides the gain of 4~22% compared to the legacy one in terms of the number of admitted flows.

• Journal of the Korean Society of Hazard Mitigation
• /
• v.10 no.4
• /
• pp.81-88
• /
• 2010

As there occurs recently and frequently a flash flood due to the climate change, a sudden local flood of great volume and short duration caused by heavy or excessive rainfall in a short period of time over a small area, it is increasing that significant danger and loss of life and property in Korea as well as the whole world. Since a flash flood usually occurs as the result of intense rainfall over small steep slope regions and has rapid runoff and debris flow, a flood rises quite quickly with little or no advance warning to prevent flood damage. The aim of this study is to quantify the severity of flash food by estimation of a flash flood index(FFI) from probability rainfall data in a study basin. FFI-D-F(FFI-Duration-Frequency) curves that present the relative severity of flash flood are developed for a study basin to provide regional basic information for the local flood forecasting and warning system particularly in ungauged catchments. It is also expected that FFI-D-F curves can be utilized for evaluation on flash flood mitigation ability and residual flood risk of both existing and planned flood control facilities.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.17 no.4
• /
• pp.965-970
• /
• 2013

It is important work to extract saliency regions from an image as preprocessing for various image processing methods. In this paper, we introduce an improved method to estimate saliency value of each pixel from an image. The proposed method is an improved work of the previously studied method using color and statistical framework to estimate saliency values. At first, saliency value of each pixel is calculated using the local contrast of an image region at various scales and the most significant saliency pixel is determined using saliency value of each pixel. Then, saliency value of each pixel is again estimated using gauss weight with respect to the most significant saliency pixel and the saliency of each pixel is determined to calculate initial probability. At last, the saliency value of each pixel is calculated by Bayes' rule. The experiments show that our approach outperforms the current statistical based method.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.16 no.4
• /
• pp.2401-2406
• /
• 2015

Pipelines are used by large heavy industries to deliver various types of fluids. Since this is important to maintain the performance of large systems, it is necessary to accurately predict remaining life of the corroded pipeline. However, predicting the remaining life is difficult due to uncertainties in the associated variables, such as geometries, material properties, corrosion rate, etc. In this paper, a statistical method for predicting corrosion remaining life is proposed using Bayesian inference. To accomplish this, pipeline failure probability was calculated using prior information about pipeline failure pressure according to elapsed time, and the given experimental data based on Bayes' rule. The corrosion remaining life was calculated as the elapsed time with 10 % failure probability. Using 10 and 50 samples generated from random variables affecting the corrosion of the pipe, the pipeline failure probability was estimated, after which the estimated remaining useful life was compared with the assumed true remaining useful life.