• International Journal of Advanced Culture Technology
• /
• v.8 no.1
• /
• pp.218-225
• /
• 2020

Purpose Wearable devices are widely used in our daily life. The purpose of this study is to investigate the relationship between persuasive designs of fitness trackers and users' physical activity behavior. Methods To test the research model, data was collected from a web-based survey in China, resulting in an effective sample of 166 usable questionnaires. The survey was restricted only to respondents who wear a fitness tracker. Results The sample surveyed in this study indicated that half of the respondents had been wearing a smart fitness tracker shorter than one year, and only 27% were long-time users (longer than two years). Dialogue support and social support strategies were both proved to be effective in increasing users' workout behavior intention. Social support strategies had a greater effect on behavior change than dialogue support strategies. Conclusion The findings from this study make several contributions to the practice. Wearable devices developers can employ the result from this study to help them design devices, which can persuade people to do more exercises and preserve a healthier life.

• International Journal of Contents
• /
• v.13 no.3
• /
• pp.43-48
• /
• 2017

As a computational approach to literary criticism, the method of character network analysis has attracted attention. The character network is composed of nodes as characters and links as relationship between characters, and has been used to analyze literary works systematically. However, there were limitations in that relationships between characters were so superficial that they could not reflect intimate relationships and quantitative data from the network were not interpreted in depth regarding meaning of literary works. In this study, we propose an improved method of character network analysis through a case study on the play . First, we segmented the character network into a dialogue network focused on speaker-to-listener relationship and an opinion network focused on subject-to-object relationship. We analyzed these networks in various ways and discussed how analysis results could reflect structure and meaning of the work. Through these studies, we strived to find a way of organic and meaningful connection between literary criticism in humanities and network analysis in computer science.

• The Journal of the Korea Contents Association
• /
• v.22 no.1
• /
• pp.138-147
• /
• 2022

Recently, chatbots have been applied in various fields and have shown good results, and many attempts to use chatbots in shopping mall product recommendation services are being conducted on e-commerce platforms. In this paper, for a conversation system that recommends a fashion that a user wants based on conversation between the user and the system and fashion image information, a transformer model that is currently performing well in various AI fields such as natural language processing, voice recognition, and image recognition. We propose a multimodal-based improved transformer model that is improved to increase the accuracy of recommendation by using dialogue (text) and fashion (image) information together for data preprocessing and data representation. We also propose a method to improve accuracy through data improvement by analyzing the data. The proposed system has a recommendation accuracy score of 0.6563 WKT (Weighted Kendall's tau), which significantly improved the existing system's 0.3372 WKT by 0.3191 WKT or more.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.8
• /
• pp.41-47
• /
• 2022

In this paper, we propose how to implement text summaries for colloquial data that are not clearly organized. For this study, SAMSum data, which is colloquial data, was used, and the BERTSumExtAbs model proposed in the previous study of the automatic summary model was applied. More than 70% of the SAMSum dataset consists of conversations between two people, and the remaining 30% consists of conversations between three or more people. As a result, by applying the automatic text summarization model to colloquial data, a result of 42.43 or higher was derived in the ROUGE Score R-1. In addition, a high score of 45.81 was derived by fine-tuning the BERTSum model, which was previously proposed as a text summarization model. Through this study, the performance of colloquial generation summary has been proven, and it is hoped that the computer will understand human natural language as it is and be used as basic data to solve various tasks.

• Journal of Digital Contents Society
• /
• v.18 no.8
• /
• pp.1621-1626
• /
• 2017

This study conducts data visualization of common climax patterns of Korean blockbuster films to analyze shots and evaluate scene (subplot unit) arrangement. For this purpose, a model of editing patterns is used to analyze how many climax patterns a film contains. Moreover, a system, which automatically collects shot images and classifies shot sizes of collected data, is designed to demonstrate that a single scene is composed based on a climax pattern. As a scene is a subplot and thus its arrangement cannot fully be analyzed only by climax patterns, dialogues of starring actors are also used to identify scenes, and the result is compared with data visualization results. It detects dialogues between particular actors and visualizes dialogue formation in a network form. Such network visualization enables the arrangement of main subplots to be analyzed, and the box office performance of a film can be explained by the density of subplots. The study of two types comparison analysis is expected to contribute to planning, plotting, and producing films.

• Smart Media Journal
• /
• v.12 no.11
• /
• pp.77-85
• /
• 2023

After COVID-19, communication through online platforms has increased, leading to an accumulation of massive amounts of conversational text data. With the growing importance of summarizing this text data to extract meaningful information, there has been active research on deep learning-based abstractive summarization. However, conversational data, compared to structured texts like news articles, often contains missing or transformed information, necessitating consideration from multiple perspectives due to its unique characteristics. In particular, vocabulary omissions and unrelated expressions in the conversation can hinder effective summarization. Therefore, in this study, we restructured by considering the characteristics of Korean conversational data, fine-tuning a pre-trained text summarization model based on KoBART, and improved conversation data summary perfomance through a refining operation to remove redundant elements from the summary. By restructuring the sentences based on the order of utterances and extracting a central speaker, we combined methods to restructure the conversation around them. As a result, there was about a 4 point improvement in the Rouge-1 score. This study has demonstrated the significance of our conversation restructuring approach, which considers the characteristics of dialogue, in enhancing Korean conversation summarization performance.

• Journal of KIISE:Software and Applications
• /
• v.35 no.5
• /
• pp.306-312
• /
• 2008

In a goal-oriented dialogue, speaker's intention can be approximated by a semantic structure that consists of a pair of a speech act and a concept sequence. Therefore, it is very important to correctly identify the semantic structure of an utterance for implementing an intelligent dialogue system. In this paper, we propose a model to efficiently analyze the semantic structures based on an active teaming method. To reduce the burdens of high-level linguistic analysis, the proposed model only uses morphological features and previous semantic structures as input features. To improve the precisions of semantic structure analysis, the proposed model adopts CRFs(Conditional Random Fields), which show high performances in natural language processing, as an underlying statistical model. In the experiments in a schedule arrangement domain, we found that the proposed model shows similar performances(92.4% in speech act analysis and 89.8% in concept sequence analysis) to the previous models although it uses about a third of training data.

• Journal of the Korea Convergence Society
• /
• v.9 no.12
• /
• pp.33-38
• /
• 2018

Recently, we are trying to predict the emotion from various information and to convey the emotion information that the supervisor wants to inform the audience. In addition, audiences intend to understand the flow of emotions through various information of non-dialogue parts, such as cinematography, scene background, background sound and so on. In this paper, we propose to extract emotions by mixing not only the context of scripts but also the cinematography information such as color, background sound, composition, arrangement and so on. In other words, we propose an emotional prediction system that learns and distinguishes various emotional expression techniques into dialogue and non-dialogue regions, contributes to the completeness of the movie, and quickly applies them to new changes. The precision of the proposed system is improved by about 5.1% and 0.4%, and the recall is improved by about 4.3% and 1.6%, respectively, when compared with the modified n-gram and morphological analysis.

• Annual Conference on Human and Language Technology
• /
• 2022.10a
• /
• pp.531-534
• /
• 2022

신경망 기반 심층학습 기술은 대화처리 분야에서 대폭적인 성능 개선을 가져왔다. 특히 GPT-2와 같은 대규모 사전학습 언어모델을 백본 네트워크로 하고 특정 도메인 타스크 대화 데이터에 대해서 미세조정 방식으로 생성되는 종단형 대화모델의 경우, 해당 도메인 타스크에 대해서 높은 성능을 내고 있다. 하지만 이런 연구들은 대부분 하나의 도메인에 대해서만 초점을 맞출 뿐 싱글 모델로 두 개 이상의 도메인을 고려하고 있지는 않다. 특히 순차적인 미세 조정은 이전에 학습된 도메인에 대해서는 catastrophic forgetting 문제를 발생시킴으로써 해당 도메인 타스크에 대한 성능 하락이 불가피하다. 본 논문에서는 이러한 문제를 해결하기 위하여 MultiWoz 목적지향 대화 데이터에 오픈 도메인 칫챗 대화턴을 유사도에 기반하여 추가하는 데이터 증강 방식을 통해 사용자 입력 및 문맥에 따라 MultiWoz 목적지향 대화와 오픈 도메인 칫챗 대화를 함께 생성할 수 있도록 하였다. 또한 목적지향 대화와 오픈 도메인 칫챗 대화가 혼합된 대화에서의 시스템 응답 생성 성능을 평가하기 위하여 오픈 도메인 칫챗 대화턴을 수작업으로 추가한 확장된 MultiWoz 평가셋을 구축하였다.

• KIPS Transactions on Software and Data Engineering
• /
• v.8 no.5
• /
• pp.213-222
• /
• 2019

Determining the meaning of a keyword in a speech dialogue system is an important technology for the future implementation of an intelligent speech dialogue interface. After extracting keywords to grasp intention from user's utterance, the intention of utterance is determined by using the semantic mark of keyword. One keyword can have several semantic marks, and we regard the task of attaching the correct semantic mark to the user's intentions on these keyword as a problem of word sense disambiguation. In this study, about 23% of all keywords in the corpus is manually tagged to build a semantic mark dictionary, a synonym dictionary, and a context vector dictionary, and then the remaining 77% of all keywords is automatically tagged. The semantic mark of a keyword is determined by calculating the context vector similarity from the context vector dictionary. For an unregistered keyword, the semantic mark of the most similar keyword is attached using a synonym dictionary. We compare the performance of the system with manually constructed training set and semi-automatically expanded training set by selecting 3 high-frequency keywords and 3 low-frequency keywords in the corpus. In experiments, we obtained accuracy of 54.4% with manually constructed training set and 50.0% with semi-automatically expanded training set.