• Annual Conference on Human and Language Technology
• /
• 2022.10a
• /
• pp.77-81
• /
• 2022

최근 사전학습 언어모델에 내재된 지식을 최대한으로 활용하고자 태스크에 대한 설명을 입력으로 주는 manual prompt tuning 방법과 자연어 대신 학습가능한 파라미터로 태스크에 대한 이해를 돕는 soft prompt tuning 방법론이 자연어처리 분야에서 활발히 연구가 진행되고 있다. 이에 본 연구에서는 페르소나 대화 생성 태스크에서 encoder-decoder 구조 기반의 사전학습 언어모델 BART를 활용하여 manual prompt tuning 및 soft prompt tuning 방법을 고안하고, 파인튜닝과의 성능을 비교한다. 전체 학습 데이터에 대한 실험 뿐 아니라, few-shot 세팅에서의 성능을 확인한다.

• Annual Conference on Human and Language Technology
• /
• 2023.10a
• /
• pp.539-545
• /
• 2023

초거대 언어모델은 심화된 언어적 이해를 요구하는 여러 분야에 높은 영향력을 미치고 있으나, 그에 수반되는 편향성과 윤리성에 대한 우려 또한 함께 증대되었다. 특히 편향된 언어모델은 인종, 성적 지향 등과 같은 다양한 속성을 가진 개인들에 대한 편견을 강화시킬 수 있다. 그러나 이러한 편향성에 관한 연구는 대부분 영어 문화권에 한정적이며 한국어에 관한 연구 또한 한국에서 발생하는 지역 갈등, 젠더 갈등 등의 사회적 문제를 반영하지 못한다. 이에 본 연구에서는 ChatGPT의 내재된 편향성을 도출하기 위해 의도적으로 다양한 페르소나를 부여하고 한국의 사회적 쟁점들을 기반으로 프롬프트 집합을 구성하여 생성된 문장의 독성을 분석하였다. 실험 결과, 특정 페르소나 또는 프롬프트에 관해서는 지속적으로 유해한 문장을 생성하는 경향성이 나타났다. 또한 각 페르소나-쟁점에 대해 사회가 갖는 편향된 시각이 모델에 그대로 반영되어, 각 조합에 따라 생성된 문장의 독성 분포에 유의미한 차이를 보이는 것을 확인했다.

• Journal of The Korean Association of Information Education
• /
• v.19 no.1
• /
• pp.77-86
• /
• 2015

The purpose of this paper is to give the opportunity to explore the career in UX(User Experience) designer and presents how to teach for UX design that is stressed in IT industry recently. We develop the textbook which is for 3 class periods for STEAM career exploration of UX designer for K-3 and K-4 students considering their students' intellectual level, prior knowledge and interests. We will offer this program to K-3 and K-4 students, soon. It will be expected that students can accept the knowledge about UX design and explore the career for UX designer.

• Annual Conference on Human and Language Technology
• /
• 2020.10a
• /
• pp.188-191
• /
• 2020

본 연구는 COVID-19 질의 응답 태스크를 위한 Poly-encoder 기반의 태스크를 제안하였다. COVID-19 질의 응답 시스템은 사람들에게 최신 정보에 대해 빠르고 신뢰성이 높은 정보를 전달하는 특성을 가져야한다. 검색 기반 질의 응답 시스템은 pairwise 연산을 기반으로 수행되는데, Poly-encoder는 사전 학습된 트랜스포머(transformer)기반의 pairwise 연산 방법론 중 기존 Cross-encoder와 Bi-encoder보다 실사용 및 성능이 뛰어남을 보였다 [1]. 특히, Poly-encoder는 정확도가 높으면서도 빠른 응답속도를 가지며 검색기반의 각종 태스크에서 좋은 성능을 보였다. 따라서 본 연구는 COVID-19를 위한 Poly-encoder기반의 질의 응답 태스크를 위하여 기존 질의 응답 태스크와 페르소나 기반의 질의 응답 태스크로 두 가지 유형의 태스크를 생성하여 모델을 학습하였다. 또한 신뢰성 있는 리소스정보로부터 모델에 최신 정보 반영을 위하여 자동 크롤러를 구축하여 데이터를 수집하였다. 마지막으로 전문가를 통한 데이터셋을 구축하여 질문-응답과 질의어-질문에 대한 모델 검증을 수행하였다.

• Journal of Korea Entertainment Industry Association
• /
• v.13 no.4
• /
• pp.93-106
• /
• 2019

Persona is an actor's external ego constructed by playing various roles, and his/her another self-portrait in the eyes of the audience. This study was conducted to analyze persona identity containing core persona(CP) and to gain implications for the growth strategy of the actress Son Ye-jin called "melo queen" by verifying consistency between the CP and audience responses to her starring works of the past. According to the related theories and models, the persona was firstly set as image, visuality, personality and consistency, and it was used to extract and sort descriptive texts about Son related news articles in the last 5 years of the six major Korean newspapers using the content analysis method. After that, we analyzed the number of viewers of her movies and the audience share of her dramas by genre. As a result, Son's persona components were found to have a proportion for 54.2% images (34.0% for melo and romance images, 20.2% for non-melo and romance images), 25.6% for visibility, 13.8% for consistency, and 6.4% for personality. Her CP was derived from a melo and romance image. Comparing this with the audience reaction, the melo romance genre dominated and showed consistency in the drama, but in the case of the film, the non-melo romance was dominant and did not match each other. The results were attributed to a wide gap between dramas and movies in terms of key viewers, box office factors, degree of genre hybridity and experimentality. Therefore, Son should actively use her CP in the drama and challenge the various roles in order to expand her persona spectrum in the film.

• Journal of Korean Library and Information Science Society
• /
• v.54 no.2
• /
• pp.23-42
• /
• 2023

The library tour program is a new type of cultural program that was first introduced and operated by J City, and library tourists travel to specialized libraries in the city according to a set course and experience various experiences. This study aims to build a customized course recommendation model that considers the characteristics of individual participants in addition to the existing fixed group travel format so that more users can enjoy the opportunity to participate in library tours. To this end, the characteristics of library travelers were categorized to establish traveler personas, and library evaluation items and evaluation criteria were established accordingly. We selected 22 libraries targeted by the library travel program and measured library data through actual visits. Based on the collected data, we derived the characteristics of suitable libraries and developed a persona-based library tour course recommendation model using a decision tree algorithm. To demonstrate the feasibility of the proposed recommendation model, we build a mobile application mockup, and conducted user evaluations with actual library users to identify satisfaction and improvements to the developed model.

• Journal of Digital Contents Society
• /
• v.19 no.6
• /
• pp.1169-1176
• /
• 2018

This study aims to create a video-commerce website that uses video for Vietnamese consumers. First of all, we completed the persona by conducting understanding questionnaires, in - depth interviews, and analysis of prior research to understand the content interest and understanding of Vietnamese consumers, purpose of website use. Persona results: First, make it easy for Vietnamese and Korean people to interact with each other by creating a contact menu. Secondly, it would be provided various information about K-pop, Korean fashion, Korean makeup(cosmetics), Korean food, Korean travel, Korean, etc., which Vietnamese people are interested in. Third, it makes it easy to purchase the products that are shown in the video that they watch.

• Annual Conference on Human and Language Technology
• /
• 2020.10a
• /
• pp.572-577
• /
• 2020

일관된 발화를 생성함에 있어 인격데이터(persona)의 도입을 이용한 연구가 활발히 진행되고 있지만, 한국어 데이터셋의 부재와 데이터셋 생성의 어려움이 문제점으로 지적된다. 본 연구에서는 인격데이터를 포함하지 않고 일관된 발화를 생성할 수 있는 방법으로 다중 대화 시스템에서 사전 학습된 자연어 추론(NLI) 모델을 도입하는 방법을 제안한다. 자연어 추론 모델을 이용한 관계 분석을 통해 과거 대화 내용 중 발화 생성에 이용할 대화를 선택하고, 자가 참조 모델(self-attention)과 다중 어텐션(multi-head attention) 모델을 활용하여 과거 대화 내용을 반영한 발화를 생성한다. 일관성 있는 발화 생성을 위해 기존 NLI데이터셋으로 수행할 수 있는 새로운 학습모델 nMLM을 제안하고, 이 방법이 일관성 있는 발화를 만드는데 기여할 수 있는 방법에 대해 연구한다.

• The Journal of the Korea Contents Association
• /
• v.18 no.6
• /
• pp.333-341
• /
• 2018

Recently, there has been a growing interest in user-centered smart city services. This study examines the spatial information type to derive a smart city service scenario through space - based communication and aims to establish a direction for user - centered experience factor scenario. The results of this study are meaningful for presenting basic data to suggest a space scenario where citizens can experience smart city as walking map and experience route applicable to future smart city test bed. In particular, we simulated the walking scenarios in the smart city test bed through Walking Map, which is tested the service direction based on user requirements. Through this study, we confirmed that smart city service through existing infrastructure can be used multipurpose. The spatial information and the experience factor linkage model and the walking map through the persona model presented in this study can be utilized as a preliminary scenario applicable to the future smart city test bed.

• Journal of Internet Computing and Services
• /
• v.20 no.5
• /
• pp.87-94
• /
• 2019

In this paper, the method of analyzing the unformatted data of consumers accumulated on social networks in the era of the Fourth Industrial Revolution by utilizing data from the service design and social psychology aspects was proposed. First, the fandom phenomenon, which shows subjective and collective behavior in a space on a social network rather than physical space, was defined from a data service perspective. The fandom model has been transformed into a collective level of customer Persona that has been analyzed at a personal level in traditional service design, and social network analysis that analyzes consumers' big data has been presented as an efficient way to pattern and visually analyze it. Consumer data collected through social leasing were pre-processed by column based on correlation, stability, missing, and ID-ness. Based on the above data, the company's brand strategy was divided into active and passive interventions and the effect of this strategic attitude on the growth direction of the consumer's fandom community was analyzed. To this end, the fandom model of consumers was proposed by dividing it into four strategies that the brand strategy had: stand-alone, decentralized, integrated and centralized, and the fandom shape of consumers was proposed as a growth model analysis technique that analyzes changes over time.