• 복식문화연구
• /
• 제20권4호
• /
• pp.594-603
• /
• 2012

This study aimed to investigate a bias pattern making method with geometrical approach. The bias-cut dress has soft silhouette of drape in the garment. However, the bias cut dress has problem of satisfying the intended garment size spec. This problem occurs from various sources. The main reason is that the bias-cut fabric tends to stretch on longitudinal direction and to shrink horizontal direction when it was hung on the body. The goal of this study was to develop a bias-cut dress pattern making method satisfying the intended garment size spec. The researchers developed the geometrical method of measuring dimensional change by calculating the compensation ratio of the fabric in true bias direction. The compensation ratio was calculated by applying draping ratio of the fabric. Three types of fabrics were used in the experiment. The warp and weft crossing angle of fabric was ranged from $78^{\circ}$ to $82^{\circ}$. The fabrics stretched longitudinally 6.9~9.9% and shrank horizontally 7.2~11.0%. The compensation ratio of the bias-cut pattern for sample dress was calculated for each fabric type. Two types of experimental bias-cut dress patterns were developed for each fabric. One pattern was made with applying full compensation ratio and the other one made with applying partial ratio of the fabric. Experimental dresses were made with these patterns. The results of the evaluation showed that the bias-cut dress pattern applying the partial compensation ratio was more appropriate than the pattern applying the full compensation ratio.

• International journal of advanced smart convergence
• /
• 제8권3호
• /
• pp.46-53
• /
• 2019

When a partitioned structure is derived from a data set using a clustering algorithm, it is not unusual to have a different set of outcomes when it runs with a different order of data. This problem is known as the order bias problem. Many algorithms in machine learning fields try to achieve optimized result from available training and test data. Optimization is determined by an evaluation function which has also a tendency toward a certain goal. It is inevitable to have a tendency in the evaluation function both for efficiency and for consistency in the result. But its preference for a specific goal in the evaluation function may sometimes lead to unfavorable consequences in the final result of the clustering. To overcome this bias problems, the first clustering process proceeds to construct an initial partition. The initial partition is expected to imply the possible range in the number of final clusters. We apply the data centric sorting to the data objects in the clusters of the partition to rearrange them in a new order. The same clustering procedure is reapplied to the newly arranged data set to build a new partition. We have developed an algorithm that reduces bias effect resulting from how data is fed into the algorithm. Experiment results have been presented to show that the algorithm helps minimize the order bias effects. We have also shown that the current evaluation measure used for the clustering algorithm is biased toward favoring a smaller number of clusters and a larger size of clusters as a result.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 2001년도 춘계학술대회 학술발표 논문집
• /
• pp.69-72
• /
• 2001

In this paper, we propose a new design method for bidirectional associative memories model with high error correction ratio. We extend the conventional GBAM model using bias terms and formulate a design procedure in the form of a constrained optimization problem. The constrained optimization problem is then transformed into a GEVP(generalized eigenvalue problem), which can be efficiently solved by recently developed interior point methods. The effectiveness of the proposed approach is illustrated by a example.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2000년도 추계종합학술대회 논문집(1)
• /
• pp.353-356
• /
• 2000

Multi-bias parameter extraction technique for HEMT small signa] equivalent circuits is presented in this paper. The technique in this paper uses S-parameters measured at various bias points in the active region to construct one optimization problem, of which the vector of unknowns contains only a set of bias-independent elements. Tests are peformed on measured S-parameters of a pHEMT at 30 bias points. Results indicate that the calculated S-parameters is similar to the measured data.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2020년도 춘계학술발표대회
• /
• pp.455-457
• /
• 2020

Recent studies demonstrate that deep learning model is easily biased by trained with unbalanced datasets. For example, the deep network can be trained to make a prediction by background feature instead the real target's feature. For those problem, a measurement called leakage was introduced to digitize this tendency. In this paper, we propose augmentation strategy which are used generally in computer vision problem to remedy this bias problem and we showed a simple augmentation methods have a effect to this task with experiments.

• Journal of Information Processing Systems
• /
• 제19권5호
• /
• pp.641-651
• /
• 2023

Malicious hate speech and gender bias comments are common in online communities, causing social problems in our society. Gender bias and hate speech detection has been investigated. However, it is difficult because there are diverse ways to express them in words. To solve this problem, we attempted to detect malicious comments in a Korean hate speech dataset constructed in 2020. We explored bidirectional encoder representations from transformers (BERT)-based deep learning models utilizing hyperparameter tuning, data sampling, and logits ensembles with a label distribution. We evaluated our model in Kaggle competitions for gender bias, general bias, and hate speech detection. For gender bias detection, an F1-score of 0.7711 was achieved using an ensemble of the Soongsil-BERT and KcELECTRA models. The general bias task included the gender bias task, and the ensemble model achieved the best F1-score of 0.7166.

• 한국언론정보학보
• /
• 제71권
• /
• pp.274-298
• /
• 2015

이 연구는 프레임 개념과 편향성 개념을 통합적으로 연결해 재난 사고 뉴스의 정치사회적 의미를 해석했다. 국내 언론이 세월호 침몰 사고라는 특수한 재난 사고의 문제 정의, 원인 해석, 도덕적 평가, 그리고 사후 처방을 제시하는 과정에 어떤 프레임을 더 편향되게 배치했는지를 이론적으로 검정해 보았다. 또한 프레임의 편향성이 정치적 이념을 달리하는 보수 신문과 진보 신문 간에 어떤 차이가 있는지 비교 분석해 보았다. 내용 분석 결과를 제시하면 다음과 같다. 첫째, 세월호 사고 진단 프레임에서는 전체적으로 파편화>개인화>권위무질서>극화의 순으로 편향되어 있었다. <조선일보>는 파편화 편향성이, 한겨레는 권위무질서 편향성이 상대적으로 컸다. 둘째, 사고 평가에서는 책임 프레임>도덕적 프레임>문제 해결 프레임>사고 원인 프레임의 순으로 편향되어 나타났다. <조선일보>는 책임 프레임, 도덕적 프레임 편향적으로 사고를 평가했다. <한겨레>는 책임 프레임, 문제 해결 프레임 편향성이 두드러졌다. 셋째, 책임 소재 프레임에서는 정부>개인>조직의 순으로 편향되어 제시됐다. <조선일보>는 정부와 개인의 책임 편향성을 드러낸 반면에, <한겨레>는 상대적으로 정부에 책임을 더 강조하면서 조직에 대한 책임 편향성도 보였다. 넷째, 문제 해결 프레임에서는 전체적으로 주제적 프레임과 일화적 프레임 편향성이 엇비슷한 수준으로 나타났다. <조선일보>는 일화적 프레임으로, <한겨레>는 주제적 프레임으로 더 편향화하는 차이를 보였다. 세월호 사고의 평가와 해석에 대한 언론의 프레임 편향성과 함께 이념적 차이에 따른 언론 간의 편향성 차이를 사회적 맥락 차원에서 토론했다.

• 대한기계학회논문집A
• /
• 제28권7호
• /
• pp.923-929
• /
• 2004

A magnetostrictive sensor is a sensor measuring elastic waves. Because of its unique non-contact measurement feature, the sensor receives more attentions in recent years. These sensors have been mainly used to measure longitudinal and torsional waves in ferromagnetic waveguides, but there increases an interest in using the sensor for flexural wave measurement. Since the performance of the sensor is strongly influenced by the applied bias magnetic field distribution, the design of the bias magnetic system providing the desired magnetic field is critical. The motivation of this investigation is to design a bias magnetic system consisting of electromagnets and yokes and the specific objective is to formulate the design problem as a bias yoke topology optimization. For the formulation, we employ linear magnetic behavior and examine the optimized results for electromagnets located at various locations. After completing the design optimization, we fabricate the prototype of the proposed bias magnetic system, and test its performance through flexural wave measurements.

• 지식경영연구
• /
• 제23권2호
• /
• pp.229-251
• /
• 2022

온라인 리뷰는 제품에 대한 이전 구매자들의 경험을 제공함으로써 다른 소비자들이 합리적인 구매 의사결정을 하는데 유용하게 활용되고 있다. 하지만 온라인 리뷰가 제품의 질과 특성을 정확히 반영하지 않고 편향되어 작성된다면 온라인 리뷰를 더이상 신뢰할 수 없는 문제가 발생한다. 따라서, 본 연구에서는 대표적인 온라인 리뷰의 편향 중 하나인 과소보고 편향의 특성을 실증 데이터를 통해 살펴보고자 한다. 구체적으로 온라인 게임 플랫폼인 스팀의 14,165개의 리뷰 데이터를 활용하여 과소보고하는 성향을 지니는 리뷰어의 특성을 살펴보고자 하였다. 분석결과, 과소보고하는 리뷰어는 주로 추천 의도를 담은 리뷰를 작성하고, 게임 출시일로부터 짧은 기간 안에 리뷰를 작성하나 다소 긴 시간동안 게임을 플레이한 후 리뷰를 작성하는 경향이 있으며, 높은 가격의 게임을 구매했을 때 리뷰를 작성하는 경향을 보였다. 본 연구는 과소보고하는 리뷰어의 특성을 탐색적으로 살펴보았기에 과소보고 편향에 대한 이해를 확장시키는 기초 연구로서 의미를 지닐 것이다.

• 한국통계학회:학술대회논문집
• /
• 한국통계학회 2003년도 추계 학술발표회 논문집
• /
• pp.63-68
• /
• 2003

One of the widely accepted assumptions in many statistical problem is that the underlying distribution is symmetric. Though a large number of nonparametric test are available in the literature for this problem, very few procedures focuses on the distributional structure when the symmetry assumption is rejected. Yanagimoto and Sibuya (1972) provided the various types of asymmetric distributional structure, positive biasedness, namely. In this paper we consider the test of symmetry against several new positive biasedness restrictions which are stronger than Yanagimoto and Sibuya's type II bias but weaker than type IV (III) bias.