• Journal of the Korea Society of Computer and Information
• /
• v.29 no.4
• /
• pp.9-21
• /
• 2024

Deep learning has been developing rapidly in recent years, with many researchers working to utilize large language models in various domains. However, there are practical difficulties that developing and utilizing language models require massive data and high-performance computing resources. Therefore, in-context learning, which utilizes prompts to learn efficiently, has been introduced, but there needs to be clear criteria for effective prompts for learning. In this study, we propose a methodology for enhancing prompt-based learning performance by improving the PET technique, which is one of the contextual learning methods, to select PVPs that are similar to the context of existing data. To evaluate the performance of the proposed methodology, we conducted experiments with 30,100 restaurant review datasets collected from Yelp, an online business review platform. We found that the proposed methodology outperforms traditional PET in all aspects of accuracy, stability, and learning efficiency.

• Journal of Korean Academy of Nursing
• /
• v.29 no.5
• /
• pp.1123-1133
• /
• 1999

The concept of empathy was examined In the professional caregiving relationship and its application was extended to the context of informal caregiving. Using the Lazarus and Folkman model, the influence of empathy on the caregiver's experience in the caregiving relationship was illustrated. The effects of the caregiver's empathy on his/her own caregiving outcomes were investigated by examining the relationship between empathy and burnout experience and life satisfaction. Empathy increased emotional exhaustion while increased personal accomplishment and personalization, indicating conflicting relationship between empathy and burnout. This conflict relationship between empathy and burnout can be explained by suggesting the distinct roles of two dimensions of empathy emotional and cognitive. The needs for more research to support the critical roles of empathy in informal caregiving context and to examine the definite roles of two dimensions of empathy were suggested.

• Proceedings of the KSPS conference
• /
• 2002.11a
• /
• pp.197-200
• /
• 2002

In this paper, we propose a phoneme duration modeling in a speech recognition system based on disicion tree state tying. We assume that phone duration has a Gamma distribution. In a training mode, we model mean and variance of each state duration in context-independent phone model based on decision tree state tying. In a recognition mode, we get mean and variance of each context-dependent phone duration form state duration information obtaind during training mode. We make a comparative study of the proposed meth with conventinal methods. Our method results in good performance compared with conventional methods.

• Journal of the Korean Home Economics Association
• /
• v.36 no.11
• /
• pp.19-41
• /
• 1998

본 연구는 개인-환경 상호작용 가정을 바탕으로 한국의 두 가지 결혼 유형에 따른 배우자 선택과 결혼 적응의 메커니즘을 이해하기 위하여 개인, 과정, 맥락의 역할을 포함안 연구 모델 (person-process-context model)을 적용시켰다. 배우사 선택, 이에 따른 부부간 유사성과 결혼적응에 있어 개인과 사회적 특성의 상대적 기여도와 그 매커니즘을 조사하기 위해 154쌍의 한국 부부를 대상으로 설문조사 하였다. 연구결과에 의하면, 중매결혼과 연애결혼의 결혼유형에 관계없이 모두 동질혼의 경향을 보였다. 개인적 특성이 결혼 적응도와 높은 상관을 보인 반면, 부부간 유사성은 결혼 적응도와 유의한 상관이 업었다. 이는 결혼 적응에 있어서 환경적 요소(e. g. dyadic similarity)보다 개인적 요소가 중요하다는 것을 시사한다. 결혼유형에 따른 동질혼의 정도에는 유의한 차이가 없으며, 또한 동질혼의 정도에 따른 결혼적응도에는 유의한 차이가 발견되지 않았다. 연애결혼과 중매결혼의 중요한 차이는 결혼전 교제기간 이었다. 연애 결혼한 부부는 비교적 오랜 교제 기간을 통해 더 만족한 결혼생활을 영위하는 것으로 나타났다. 결혼전 교제기간의 효과를 통제한 후 두 결혼 유형의 결혼적응도에 대한 차이는 사라졌다. 연구결과들을 결혼과 성격에 관한 이론을 비탕으로 논의하였다.

• Proceedings of the Korea Concrete Institute Conference
• /
• 2004.05a
• /
• pp.466-471
• /
• 2004

Efficient numerical finite element analysis of creeping concrete structures requires the use Kelvin or Maxwell chain model, which is most conveniently identified from a continuous retardation or relaxation spectrum, the spectrum in turn being determined from the given compliance or relaxation function. The method of doing that within the context of solidification theory for creep with aging was previously worked out by Bazant and Xi, but only for the case of a continuous retardation spectrum based on Kelvin chain. The present paper is motivated by the need to incorporate concrete creep into the recently published microplane model M4 for nonlinear triaxial behavior of concrete, including tensile fracturing and behavior under compression. In that context. the Maxwell chain is more effective than Kelvin chain. because of the kinematic constraint of the microplanes used in M4. Determination of the continuous relaxation spectrum for Maxwell chain. based on the solidification theory, is outlined and numerical examples are presented.

• Asia pacific journal of information systems
• /
• v.12 no.4
• /
• pp.55-75
• /
• 2002

The purposes of this study are in threefold; to explain the interactivity in web site, to extend the Technology Acceptance Model(TAM) in the WWW context, and to examine its relationship with the existing TAM components. We proposed that the WWW is not just web systems but interactive spaces for virtual community. Web interactivity is represented by the two constructs: system interactivity and social interactivity. We extended and empirically examined the TAM for the web context. An individual's attitude toward the use of the web site is significantly affected by the individual's perception about ease of use, usefulness, and system interactivity. At the same time, behavioral intentions to use the web site is highly related to the social interactivity. The results showed that the system interactivity and social interactivity of a web site are important factors in affecting the user's beliefs and behavior intention of such a web site, So web page providers not only have to make the web site useful and easy to use, but also need to design an interactive web page by putting both systemic aspects and social aspects.

• Genomics & Informatics
• /
• v.9 no.4
• /
• pp.194-196
• /
• 2011

Promoter prediction is a very important problem and is closely related to the main problems of bioinformatics such as the construction of gene regulatory networks and gene function annotation. In this context, we developed an integrated promoter prediction program using hybrid methods, PromoterWizard, which can be employed to detect the core promoter region and the transcription start site (TSS) in vertebrate genomic DNA sequences, an issue of obvious importance for genome annotation efforts. PromoterWizard consists of three main modules and two auxiliary modules. The three main modules include CDRM (Composite Dependency Reflecting Model) module, SVM (Support Vector Machine) module, and ICM (Interpolated Context Model) module. The two auxiliary modules are CpG Island Detector and GCPlot that may contribute to improving the predictive accuracy of the three main modules and facilitating human curator to decide on the final annotation.

• Journal of the Korea Society of Computer and Information
• /
• v.21 no.2
• /
• pp.113-122
• /
• 2016

This study investigates how anticipated value(attainment value, intrinsic value, utility value, and reward value) and self-efficacy in providing information have an influence on continuance intention to provide information in the context of online community or social media. We propose a conceptual model for examining the framework for the effect of anticipated value on attitude toward providing information, and the effect of attitude toward providing information and self-efficacy in providing information on continuance intention to provide information. Proposed research model is empirically assessed using the survey data obtained from 221 respondents in Korea. Results show that attainment value and intrinsic value have a significant impact on attitude toward providing information, whereas utility value and reward value are found to be insignificant. In addition, attitude toward providing information and self-efficacy in providing information are positively and significantly related to continuance intention to provide information.

• Speech Sciences
• /
• v.10 no.1
• /
• pp.71-84
• /
• 2003

In the acoustic modeling for large vocabulary speech recognition, a sparse data problem caused by a huge number of context-dependent (CD) models usually leads the estimated models to being unreliable. In this paper, we develop a new clustering method based on the C45 decision-tree learning algorithm that effectively encapsulates the CD modeling. The proposed scheme essentially constructs a supervised decision rule and applies over the pre-clustered triphones using the C45 algorithm, which is known to effectively search through the attributes of the training instances and extract the attribute that best separates the given examples. In particular, the data driven method is used as a clustering algorithm while its result is used as the learning target of the C45 algorithm. This scheme has been shown to be effective particularly over the database of low unknown-context ratio in terms of recognition performance. For speaker-independent, task-independent continuous speech recognition task, the proposed method reduced the percent accuracy WER by 3.93% compared to the existing rule-based methods.

• Language and Information
• /
• v.18 no.2
• /
• pp.1-29
• /
• 2014

Relativization is one of the common syntactic operations to merge two different clauses into a single information unit. This operation plays a pivotal role to structuralize multiple clauses cohesively as well as serves to specify the property an individual has within the context. That implies that relativization contributes to information structure of multiclausal sentences. In this context, this paper delves into information structure of relative clauses in English with an eye toward creation of a computational model from a standpoint of machine translation. The current work employs Head-driven Phrase Structure Grammar (HPSG, Pollard and Sag (1994)) as a theory of grammar and Minimal Recursion Semantics (MRS, Copestake et al. (2005) as a meaning representation system. Building upon these formalisms, this paper addresses how information structure of relative clauses can be represented and constrained. The current work makes use of Individual CONStraints (ICONS) for modeling relative clauses with respect to information structure. The current work also investigates which relative clause involves which information structure constraint. The present study argues that non-restrictive relative clauses impose a more specific constraint on information structure than restrictive relative clauses.