• Annual Conference on Human and Language Technology
• /
• 2023.10a
• /
• pp.66-71
• /
• 2023

ChatGPT가 등장한 이후 다양한 거대 언어 모델(Large Language Model, LLM)이 등장하였고, 이러한 LLM을 목적에 맞게 파인튜닝하여 사용할 수 있게 되었다. 하지만 LLM을 새로 학습하는 것은 물론이고, 단순 튜닝만 하더라도 일반인은 시도하기 어려울 정도의 많은 컴퓨팅 자원이 필요하다. 본 연구에서는 공개된 LLM을 별도의 학습 없이 사용하여 zero-shot 프롬프팅으로 이진 분류 태스크에 대한 성능을 확인하고자 했다. 학습이나 추가적인 튜닝 없이도 기존 선학습 언어 모델들에 준하는 이진 분류 성능을 확인할 수 있었고, 성능이 좋은 LLM의 경우 분류 실패율이 낮고 일관적인 성능을 보여 상당히 높은 활용성을 확인하였다.

• Journal of the Korea Convergence Society
• /
• v.12 no.6
• /
• pp.57-63
• /
• 2021

The purpose of the convergence study is to provide basic data for improving job satisfaction of speech and language therapists by identifying the relationship between personal characteristics, empathy ability, and job satisfaction of speech and language therapists, especially between cognitive and emotional empathy ability and job satisfaction. The subjects of the study were 111 people of speech and language therapists working in Daegu and Gyeongbuk. For data analysis, t-test, ANOVA, and correlation analysis were performed using the SPSS/PC 21.0 statistical program according to the purpose of the study. As a result of the study, the relationship between empathy ability and job satisfaction was found to be high, and there was a high correlation between cognitive and emotional empathy ability, which is a sub-factor of empathy ability, and job satisfaction. Therefore, follow-up research is needed to identify the empathy and job satisfaction about nationwide speech and language therapists for implementing various education programs and improving empathy skills in the future.

• Journal of the Korea Convergence Society
• /
• v.13 no.3
• /
• pp.77-89
• /
• 2022

It has been proven through many previous researches that the pretrained language model with a large corpus helps improve performance in various natural language processing tasks. However, there is a limit to building a large-capacity corpus for training in a language environment where resources are scarce. Using the Cross-lingual Post-Training (XPT) method, we analyze the method's efficiency in Korean, which is a low resource language. XPT selectively reuses the English pretrained language model parameters, which is a high resource and uses an adaptation layer to learn the relationship between the two languages. This confirmed that only a small amount of the target language dataset in the relationship extraction shows better performance than the target pretrained language model. In addition, we analyze the characteristics of each model on the Korean language model and the Korean multilingual model disclosed by domestic and foreign researchers and companies.

• Smart Media Journal
• /
• v.12 no.1
• /
• pp.32-46
• /
• 2023

Natural Language Processing (NLP) has grown tremendously in recent years. Typically, bilingual, and multilingual translation models have been deployed widely in machine translation and gained vast attention from the research community. On the contrary, few studies have focused on translating between spoken and sign languages, especially non-English languages. Prior works on Sign Language Translation (SLT) have shown that a mid-level sign gloss representation enhances translation performance. Therefore, this study presents a new large-scale Korean sign language dataset, the Museum-Commentary Korean Sign Gloss (MCKSG) dataset, including 3828 pairs of Korean sentences and their corresponding sign glosses used in Museum-Commentary contexts. In addition, we propose a translation framework based on self-supervised learning, where the pretext task is a text-to-text from a Korean sentence to its back-translation versions, then the pre-trained network will be fine-tuned on the MCKSG dataset. Using self-supervised learning help to overcome the drawback of a shortage of sign language data. Through experimental results, our proposed model outperforms a baseline BERT model by 6.22%.

• Journal of the Korea Convergence Society
• /
• v.3 no.3
• /
• pp.1-5
• /
• 2012

In this paper, we developed integrated billing system for multi-language. It is very easy to customize for other language and also It has functions in integrated billing system. Whenever administrator insert UI message data for other language, we can make billing packages for its own country to service integrated billing policy.

• Journal of Digital Convergence
• /
• v.17 no.10
• /
• pp.321-327
• /
• 2019

The purpose of this study is to develop a smart device-based Language Rehabilitation Program (LRP) to improve communication ability for the patients with language disability. The content of the LRP includes a variety of semantic categories and grammatical elements and consists of 17 semantic categories, 29 tasks and 3780 items to improve comprehension/production ability at word level, semantic category level, sentence level and discourse level. We developed LRP as a Windows-base management program and an Android-base language rehabilitation application. LRP was developed into an application for smart devices, providing real-time delivery of training contents, measurement and database of training task results, and patient progress and monitoring. A follow-up study will be conducted on the verification of the language rehabilitation effect using LRP by patients with language disability.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.8
• /
• pp.1987-1994
• /
• 2014

The purpose of this paper is to propose convergence education for creative manpower training between foreign language and Information technology department's students. Academia and industry, and government have attempted a number of changes and efforts for the employment of university graduated students and they have also promoted various projects and plans to suggest a better way. But the graduates employment rate is still left as serious social problem. To solve this problem, we seek to propose convergence education plan to enhance the employment rate of foreign language department that is relatively low unlike other departments. This proposal will make a new paradigm in education - Creativity, Difference, Locality, and Convergence. Based on these paradigms, it is expected to overcome the limitation of labor market.

• Journal of the Korea Convergence Society
• /
• v.6 no.5
• /
• pp.131-142
• /
• 2015

The aim of study is to present a better teaching strategy to reduce writing errors of Chinese learners of Korean language after finding out what reasons of errors were after analyzing of their writing errors in Korean language. Analyzed contents were writhing in Korean language of 'how I think Korean', 'about Chinese and Korean culture', 'friends' and analyzed what errors were occurred. The vocabulary errors frequencies were counted by the criteria which was set by a researcher. The results were as follows. The frequency of substitute error was the most and were followed by spelling error, wrong type error, omission error and adding error. It is suggested when we teach Korean Language to Chinese learners and develop text for them, the vocabularies should be presented with examples of how to be used in context instead of presenting only vocabulary on the text. It would be a better way to reduce writing errors of Chinese learners of Korean language.

• Journal of the Korea Convergence Society
• /
• v.7 no.4
• /
• pp.173-180
• /
• 2016

OpenCV is a programming language used in digital image processing and computer vision. In this study, look at media arts made using OpenCV programming language and find out about the utilization possibilities. To this end, the first, look at OpenCV functions that are frequently used in media art, the examples of utilizing the functions. The second, discuss media arts using OpenCV. focused on the OpenCV functions, programming language for an production of media art. The third, analyze features of media arts using OpenCV, mainly focused on the functions and programming languages. The study may provide guidance to the artists to produce a media art using the OpenCV or programming language.

• Journal of the Korea Convergence Society
• /
• v.12 no.1
• /
• pp.57-63
• /
• 2021

Spoken language understanding, which aims to understand utterance as naturally as human would, are mostly focused on English language. In this paper, we construct a Korean language dataset for spoken language understanding, which is based on a conversational corpus between reservation system and its user. The domain of conversation is limited to restaurant reservation. There are 7 types of slot tags and 5 types of intent tags in 6857 sentences. When a model proposed in English-based research is trained with our dataset, intent classification accuracy decreased a little, while slot filling F1 score decreased significantly.