• Journal of the Korea Society of Computer and Information
• /
• v.28 no.1
• /
• pp.27-38
• /
• 2023

In this paper, we propose a conversational AI agent based on continual learning that can continuously learn and grow with new data over time. A continual learning-based conversational AI agent consists of three main components: Task manager, User attribute extraction, and Auto-growing knowledge graph. When a task manager finds new data during a conversation with a user, it creates a new task with previously learned knowledge. The user attribute extraction model extracts the user's characteristics from the new task, and the auto-growing knowledge graph continuously learns the new external knowledge. Unlike the existing conversational AI agents that learned based on a limited dataset, our proposed method enables conversations based on continuous user attribute learning and knowledge learning. A conversational AI agent with continual learning technology can respond personally as conversations with users accumulate. And it can respond to new knowledge continuously. This paper validate the possibility of our proposed method through experiments on performance changes in dialogue generation models over time.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2023.07a
• /
• pp.67-68
• /
• 2023

본 논문에서 제안한 Multi Agent Multi Action은 기존의 대화형 시스템 방식인 Single Agent Single Action 구조에 비해 확장성을 갖춘 대화 시스템을 구현하는 방식이다. 시스템을 여러 에이전트로 분할하고, 각 에이전트가 특정 액션에 대한 처리를 담당함으로써 보다 유연하고 효율적인 대화형 시스템을 구현할 수 있으며, 다양한 작업에 특화된 에이전트를 그룹화함으로써 작업의 효율성을 극대화하고, 사용자 경험을 향상 시킬 수 있다.

• Journal of the Korea Convergence Society
• /
• v.13 no.3
• /
• pp.11-22
• /
• 2022

Recently, research and development using various Artificial Intelligence (AI) technologies are being conducted in the field of education. Among the AI in Education (AIEd), conversational agents are not limited by time and space, and can learn more effectively by combining them with various AI technologies such as voice recognition and translation. This paper conducted a trend analysis on platforms that have a large number of users and used conversational agents for English learning among commercialized application. Currently commercialized educational platforms using conversational agent through trend analysis has several limitations and problems. To analyze specific problems and limitations, a comparative experiment was conducted with the latest pre-trained large-capacity dialogue model. Sensibleness and Specificity Average (SSA) human evaluation was conducted to evaluate conversational human-likeness. Based on the experiment, this paper propose the need for trained with large-capacity parameters dialogue models, educational data, and information retrieval functions for effective English conversation learning.

• The Journal of the Convergence on Culture Technology
• /
• v.8 no.1
• /
• pp.551-563
• /
• 2022

This study aims to identify the effect on the user experiences when the embodied agent in a self-driving car interacts with emotional expressions by using 'interjection'. An experimental study was designed with two conditions: the inclusion of injections in the agent's conversation feedbacks (with interjections vs. without interjections) and the type of conversation (task-oriented conversation vs. social-oriented conversation). The online experiment was conducted with the four video clips of conversation scenario treatments and measured intimacy, likability, trust, social presence, perceived anthropomorphism, and future intention to use. The result showed that when the agent used interjection, the main effect on social presence was found in both conversation types. When the agent did not use interjection in the task-oriented conversation, trust and future intention to use were higher than when the agent talked with emotional expressions. In the context of the conversation with the AI agent in a self-driving car, we found only the effect of adding emotional expression by using interjection on the enhancing social presence, but no effect on the other user experience factors.

• The Journal of the Korea Contents Association
• /
• v.18 no.8
• /
• pp.92-101
• /
• 2018

As the interactive AI speaker becomes popular, voice recognition is regarded as an important vehicle-driver interaction method in case of autonomous driving situation. The purpose of this study is to confirm whether multimodal interaction in which feedback is transmitted by auditory and visual mode of AI characters on screen is more effective in user experience optimization than auditory mode only. We performed the interaction tasks for the music selection and adjustment through the AI speaker while driving to the experiment participant and measured the information and system quality, presence, the perceived usefulness and ease of use, and the continuance intention. As a result of analysis, the multimodal effect of visual characters was not shown in most user experience factors, and the effect was not shown in the intention of continuous use. Rather, it was found that auditory single mode was more effective than multimodal in information quality factor. In the semi-autonomous driving stage, which requires driver 's cognitive effort, multimodal interaction is not effective in optimizing user experience as compared to single mode interaction.

• Annual Conference on Human and Language Technology
• /
• 2021.10a
• /
• pp.453-456
• /
• 2021

대화형 에이전트가 일관성 없는 답변, 재미 없는 답변을 하는 문제를 해결하기 위하여 최근 페르소나 기반의 대화 분야의 연구가 활발히 진행되고 있다. 그러나 한국어로 구축된 페르소나 대화 데이터는 아직 구축되지 않은 상황이다. 이에 본 연구에서는 영어 원본 데이터에서 한국어로 번역된 데이터를 활용하여 최초의 페르소나 기반 한국어 대화 모델을 제안한다. 전처리를 통하여 번역 품질을 향상시킨 데이터에 사전 학습 된 한국어 모델인 KoBERT와 KoELECTRA를 미세조정(fine-tuning) 시킴으로써 모델에게 주어진 페르소나와 대화 맥락을 고려하여 올바른 답변을 선택하는 모델을 학습한다. 실험 결과 KoELECTRA-base 모델이 가장 높은 성능을 보이는 것을 확인하였으며, 단순하게 사용자의 발화만을 주는 것 보다 이전 대화 이력이 추가적으로 주어졌을 때 더 좋은 성능을 보이는 것을 확인할 수 있었다.

• Journal of Intelligence and Information Systems
• /
• v.29 no.3
• /
• pp.267-286
• /
• 2023

The conversational agents such as AI speakers utilize voice conversation for human-computer interaction. Voice recognition errors often occur in conversational situations. Recognition errors in user utterance records can be categorized into two types. The first type is misrecognition errors, where the agent fails to recognize the user's speech entirely. The second type is misinterpretation errors, where the user's speech is recognized and services are provided, but the interpretation differs from the user's intention. Among these, misinterpretation errors require separate error detection as they are recorded as successful service interactions. In this study, various text separation methods were applied to detect misinterpretation. For each of these text separation methods, the similarity of consecutive speech pairs using word embedding and document embedding techniques, which convert words and documents into vectors. This approach goes beyond simple word-based similarity calculation to explore a new method for detecting misinterpretation errors. The research method involved utilizing real user utterance records to train and develop a detection model by applying patterns of misinterpretation error causes. The results revealed that the most significant analysis result was obtained through initial consonant extraction for detecting misinterpretation errors caused by the use of unregistered neologisms. Through comparison with other separation methods, different error types could be observed. This study has two main implications. First, for misinterpretation errors that are difficult to detect due to lack of recognition, the study proposed diverse text separation methods and found a novel method that improved performance remarkably. Second, if this is applied to conversational agents or voice recognition services requiring neologism detection, patterns of errors occurring from the voice recognition stage can be specified. The study proposed and verified that even if not categorized as errors, services can be provided according to user-desired results.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.4
• /
• pp.23-30
• /
• 2024

In this paper, we propose a deep learning-based personalized senior care service application. The proposed application uses Speech to Text technology to convert the user's speech into text and uses it as input to Autogen, an interactive multi-agent large-scale language model developed by Microsoft, for user convenience. Autogen uses data from previous conversations between the senior and ChatBot to understand the other user's intent and respond to the response, and then uses a back-end agent to create a wish list, a shared calendar, and a greeting message with the other user's voice through a deep learning model for voice cloning. Additionally, the application can perform home IoT services with SKT's AI speaker (NUGU). The proposed application is expected to contribute to future AI-based senior care technology.

• Journal of the HCI Society of Korea
• /
• v.14 no.1
• /
• pp.15-22
• /
• 2019

The purpose of this study is to investigate the users' expectation and expectation gap about the attributes of smart speaker as an intelligent agent, ie autonomy, sociality, responsiveness, activeness, time continuity, goal orientation. To this end, semi-structured interviews were conducted for smart speaker users and analyzed based on ground theory. Result has shown that people have huge expectation gap about the sociality and human-likeness of smart speakers, due to limitations in technology. The responsiveness of smart speakers was found to have positive expectation gap. For the memory of time-sequential information, there was an ambivalent expectation gap depending on the degree of information sensitivity and presentation method. We also found that there was a low expectation level for autonomous aspects of smart speakers. In addition, proactive aspects were preferred only when appropriate for the context. This study presents implications for designing a way to interact with smart speakers and managing expectations.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.3
• /
• pp.797-802
• /
• 2023

In an environment where English is a foreign language, English learners can use AI voice chatbots in English-speaking practice activities to enhance their speaking motivation, provide opportunities for communication practice, and improve their English speaking ability. In this study, we propose a teaching-style AI voice chatbot that can be easily utilized by lower elementary school students and enhance their learning. To apply the Teachable Agent system to language learning, which is an activity based on tense, context, and memory, we proposed a new method of TA by applying the Teachable Agent to reflect the learner's English pronunciation and level and generate the agent's answers according to the learner's errors and implemented a Teachable Agent AI chatbot prototype. We conducted usability evaluations with actual elementary English teachers and elementary school students to demonstrate learning effects. The results of this study can be applied to motivate students who are not interested in learning or elementary school students to voluntarily participate in learning through role-switching.