• Annual Conference of KIPS
• /
• 2024.05a
• /
• pp.570-573
• /
• 2024

본 연구는 대형 언어 모델 (LLM) 시대에 공감적 대화 생성을 위한 감정 인식의 필요성을 확인하고 소형 언어 모델 (SLM)을 통한 미세 조정 학습이 고비용 LLM, 특히 ChatGPT의 대안이 될 수 있는지를 탐구한다. 이를 위해 KoBERT 미세 조정 모델과 ChatGPT를 사용하여 사용자 감정을 인식하고, Polyglot-Ko 미세 조정 모델 및 ChatGPT를 활용하여 공감적 응답을 생성하는 비교 실험을 진행하였다. 실험 결과, KoBERT 기반의 감정 분류기는 ChatGPT의 zero-shot 접근 방식보다 뛰어난 성능을 보였으며, 정확한 감정 분류가 공감적 대화의 질을 개선하는 데 기여함을 확인하였다. 이는 공감적 대화 생성을 위해 감정 인식이 여전히 필요하며, SLM의 미세 조정이 고비용 LLM의 실용적 대체 수단이 될 수 있음을 시사한다.

• Journal of Internet Computing and Services
• /
• v.19 no.6
• /
• pp.125-132
• /
• 2018

In this paper, we design a programming education model that uses emotional intelligence and apply the model to programming education in elementary school. In our previous work, we found that there is a meaningful correlation between emotional intelligence and programming ability. In this paper, as a follow-up study, we design a programming education model based on a storytelling model and emotional intelligence. In order to test the performance of the proposed model, we applied our proposed model to the 5th grade elementary school students who have no programming experience. Based on extensive survey work and statistical analysis, we found that the experimental group by the programming education using the emotional intelligence got a statistically significant higher achievement than the comparative group by the traditional programming education. We hope that our model will be helpful in programming education in schools.

• Proceedings of the Korean Information Science Society Conference
• /
• 2006.10b
• /
• pp.426-430
• /
• 2006

TFT-LCD의 생산 과정에서 나오는 불량 제품의 검출은 자동화 과정에 의해 선택된 잠재적 불량 제품의 선택에 이은 인간의 목시검사에 의한 판단을 통해 이루어진다. 이러한 목시검사를 자동화하기 위해서는 불량의 식별성에 영향을 미치는 각 요소들에 대한 정량적인 분석, 그리고 각 요소들과 실제 최종적인 불량 판단 여부 사이의 체계적인 함수 관계의 파악이 필요하다. 본 논문에서는 TFT-LCD의 영역형 얼룩을 구성하는 특징적인 요소들을 정의하고, 이를 통해 불량 수준을 정의하는 수치화 함수를 유도하는 과정과, 이를 영역형 얼룩의 불량 수준 수치화에 적용한 결과를 보여 준다.

• Journal of Platform Technology
• /
• v.6 no.4
• /
• pp.11-17
• /
• 2018

In the smart intelligence information society, there is a possibility that the social dysfunction such as the personal information protection issue and the risk to the electronic surveillance society may be highlighted. In this paper, we refer to various categories and classify electronic surveillance into audio surveillance, visual surveillance, location surveillance, biometric information surveillance, and data surveillance. In order to respond to new electronic surveillance in the intelligent information society, it requires a change of perception that is different from that of the past. This starts with the importance of digital privacy and results in the right to self-determination of personal information. Therefore, in order to preemptively respond to the dysfunctions that may arise in the intelligent information society, it is necessary to further raise the awareness of the civil society to protect information human rights.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.19 no.3
• /
• pp.233-244
• /
• 2023

Although artificial intelligence and machine learning technologies have been used in various fields, problems with personal information protection have arisen based on centralized data collection and processing. Federated learning has been proposed to solve this problem. Federated learning is a process in which clients who own data in a distributed data environment learn a model using their own data and collectively create an artificial intelligence model by centrally collecting learning results. Unlike the centralized method, Federated learning has the advantage of not having to send the client's data to the central server. In this paper, we quantitatively present the performance improvement when federated learning is applied using the building change detection learning data. As a result, it has been confirmed that the performance when federated learning was applied was about 29% higher on average than the performance when it was not applied. As a future work, we plan to propose a method that can effectively reduce the number of federated learning rounds to improve the convergence time of federated learning.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1996.11a
• /
• pp.450-455
• /
• 1996

Emotional intelligence is investigated in terms of a composing machine as a modern abstract art form. Music has the longest tradition of being an art form which has an explicit formal foundation. Formal aspects of traditional and modern music theory are explained in terms of simple numerical relationship and illustrated with examples. The exploration of art in the view of intelligence, information and structure will restore the balanced sense of art and science which seeks happiness in life.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.8 no.2
• /
• pp.691-708
• /
• 2014

This paper formalizes a special type of social networking knowledge, which is called "workflow performer-role affiliation networking knowledge." A workflow model specifies execution sequences of the associated activities and their affiliated relationships with roles, performers, invoked-applications, and relevant data. In Particular, these affiliated relationships exhibit a stream of organizational work-sharing knowledge and utilize business process intelligence to explore resources allotting and planning knowledge concealed in the corresponding workflow model. In this paper, we particularly focus on the performer-role affiliation relationships and their implications as organizational and business process intelligence in workflow-driven organizations. We elaborate a series of theoretical formalisms and practical implementation for modeling, discovering, and visualizing workflow performer-role affiliation networking knowledge, and practical details as workflow performer-role affiliation knowledge representation, discovery, and visualization techniques. These theoretical concepts and practical algorithms are based upon information control net methodology for formally describing workflow models, and the affiliated knowledge eventually represents the various degrees of involvements and participations between a group of performers and a group of roles in a corresponding workflow model. Finally, we summarily describe the implications of the proposed affiliation networking knowledge as business process intelligence, and how worthwhile it is in discovering and visualizing the knowledge in workflow-driven organizations and enterprises that produce massively parallel interactions and large-scaled operational data collections through deploying and enacting massively parallel and large-scale workflow models.

• Asian Journal of Innovation and Policy
• /
• v.5 no.3
• /
• pp.275-292
• /
• 2016

Many countries set growth engine technologies and industries for economic growth and job creation. Each country always wants to know their technological or industrial position in the world in these industries. This study aims at identifying the worldwide position of 19 growth engine industries defined in Korean government. The methods are quantitative by counting the number of startup companies in the world. The ORBIS database was used to extract the number. Therefore, this article may be the first research for the world appearance of growth engine industries and its comparison between world and G7, and between G7 countries. Also, this may be the first study using the ORBIS database on the analysis of certain technology industries. Further, we showed a method to identify world features of technology industries.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.6
• /
• pp.3121-3143
• /
• 2019

Segmentation plays an important role in the field of image processing and computer vision. Intuitionistic fuzzy C-means (IFCM) clustering algorithm emerged as an effective technique for image segmentation in recent years. However, standard fuzzy C-means (FCM) and IFCM algorithms are sensitive to noise and initial cluster centers, and they ignore the spatial relationship of pixels. In view of these shortcomings, an improved algorithm based on IFCM is proposed in this paper. Firstly, we propose a modified non-membership function to generate intuitionistic fuzzy set and a method of determining initial clustering centers based on grayscale features, they highlight the effect of uncertainty in intuitionistic fuzzy set and improve the robustness to noise. Secondly, an improved nonlinear kernel function is proposed to map data into kernel space to measure the distance between data and the cluster centers more accurately. Thirdly, the local spatial-gray information measure is introduced, which considers membership degree, gray features and spatial position information at the same time. Finally, we propose a new measure of intuitionistic fuzzy entropy, it takes into account fuzziness and intuition of intuitionistic fuzzy set. The experimental results show that compared with other IFCM based algorithms, the proposed algorithm has better segmentation and clustering performance.

• Asia pacific journal of information systems
• /
• v.32 no.1
• /
• pp.125-150
• /
• 2022

Technology companies launched their intelligent personal assistants (IPAs). IPAs not only provide individuals with a convenient way to interact with technology but also offer them the opportunity to interact with AI in a useful and meaningful form. Therefore, the global IPAs have experienced tremendous growth over the past decade. But maintaining continuous usage intention is still a massive challenge for developers and marketers and previous technology adoption models are not enough to explain continuous usage intention of IPAs. Thus, we adopted the bi-dimensional perspectives of utilitarian and hedonic value in this research model, and investigated how three characteristics of IPAs - anthropomorphism, autonomy, and intelligence - affect utilitarian value and hedonic value, which in turn continuous usage intentions. 227 data were collected from IPA users. The results showed that IPAs' continuous usage intention is significantly determined by both utilitarian and hedonic value, with the hedonic value being more prominent. In addition, the results showed that anthropomorphism and intelligence are the most important antecedents of utilitarian and hedonistic value. The results also illustrated that autonomy is a crucial predictor of utilitarian value rather than hedonistic value. Our work contributes to current research by widening the theoretical understanding of the effect of IPA characteristics on continuous usage intention through bi-dimensional values. Our paper also provides IPAs' developer and marketer guidelines for enhancing continuous usage intention.