• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.9
• /
• pp.381-390
• /
• 2022

Influenced by a culture that prefers non-face-to-face activity during the COVID-19 pandemic, chatbot usage is accelerating. Chatbots have been used for various purposes, not only for customer service in businesses and social conversations for fun but also for mental health. Chatbots are a platform where users can easily talk about their depressed moods because anonymity is guaranteed. However, most relevant research has been on social media data, especially Twitter data, and few studies have analyzed the commercially used chatbots data. In this study, we identified the characteristics of depressive discourse in user-chatbot interaction data by analyzing the chats, including the word 'depress,' using the topic modeling algorithm and the text-mining technique. Moreover, we compared its characteristics with those of the depressive moods in the Twitter data. Finally, we draw several design guidelines and suggest avenues for future research based on the study findings.

• The Journal of Society for e-Business Studies
• /
• v.26 no.4
• /
• pp.15-36
• /
• 2021

The airline industry faces with significant competition due to the rise of technology innovation and diversified customer needs. Therefore, continuous quality management is essential to gain competitive advantages. For this reason, there have been various studies to measure and manage service quality using customer reviews. However, previous studies have focused on measuring customer satisfaction only, neglecting systematic management between customer expectations and perception based on customer reviews. In response, this study suggests a framework to identify relevant criteria for service quality management, measure the importance, and assess the customer perception based on customer reviews. Machine learning techniques, topic models, and sentiment analysis are used for this study. This study can be used as an important strategic tool for evaluating service quality by identifying important factors for airline customer satisfaction while presenting a framework for identifying each airline's current service level.

• Journal of Korea Society of Industrial Information Systems
• /
• v.27 no.4
• /
• pp.65-79
• /
• 2022

This study identified smart manufacturing technologies using patent and topic modeling, and compared the technology development trends in countries such as the United States, Japan, Germany, China, and South Korea. To this purpose, this study collected patents in the United States and Europe between 1991 and 2020, processed patent abstracts, and identified topics by applying latent Dirichlet allocation model to the data. As a result, technologies related to smart manufacturing are divided into seven categories. At a global level, it was found that the proportion of patents in 'data processing system' and 'thermal/fluid management' technologies is increasing. Considering the fact that South Korea has relative competitiveness in thermal/fluid management technologies related to smart manufacturing, it would be a successful strategy for South Korea to promote smart manufacturing in heavy and chemical industry. This study is significant in that it overcomes the limitations of quantitative technology level evaluation proposed a new methodology that applies text mining.

• Journal of The Korean Association of Information Education
• /
• v.26 no.5
• /
• pp.307-315
• /
• 2022

The purpose of this study was to design a training program for pre-service elementary teachers, incorporating the concepts of extended reality technologies. This program contained the basic skills necessary for them to utilize in their future classrooms. To accomplish this, 12 undergraduate students of various majors enrolled in one of Korea's national universities of education were selected as research subjects. For a total of 6 times over 6 weeks, they participated in a training program learning the basic concepts of virtual, augmented, and mixed reality, as well as creating their own education software to use in simulated classes. To improve the quality of future research efforts, this study found it would be beneficial to: 1) expand the relevant support equipment, 2) provide students with preliminary, background knowledge of text-based programming, 3) introduce short-term, more intensive training, and 4) improve the survey methods for this research.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.05a
• /
• pp.2-5
• /
• 2022

With the advent of deep learning and the rapid development of ICT (Information and Communication Technology), research using artificial intelligence is being actively conducted in various fields of society such as politics, economy, and culture and so on. Deep learning-based artificial intelligence technology is subdivided into various domains such as natural language processing, image processing, speech processing, and recommendation system. In particular, as the industry is advanced, the need for a recommendation system that analyzes market trends and individual characteristics and recommends them to consumers is increasingly required. In line with these technological developments, this paper extracts and classifies attribute information from structured or unstructured text and image big data through deep learning-based technology development of 'language processing intelligence' and 'image processing intelligence', and We propose an artificial intelligence-based 'customized fashion advisor' service integration system that analyzes trends and new materials, discovers 'market-consumer' insights through consumer taste analysis, and can recommend style, virtual fitting, and design support.

• Convergence Security Journal
• /
• v.22 no.3
• /
• pp.33-39
• /
• 2022

Recently, cyberattacks on infrastructure have been continuously occurring with the starting of neutralizing the user authentication function of information systems. Accordingly, the vulnerabilities of system are increasing day by day, such as the increase in the vulnerabilities of the encryption system. In this paper, an alternative technique for the symmetric key algorithm has been developed in order to build the encryption algorithm that is not easy for beginners to understand and apply. Vigenere Cipher is a method of encrypting alphabetic text and it uses a simple form of polyalphabetic substitution. The encryption application system proposed in this study uses the simple form of polyalphabetic substitution method to present an application model that integrates the three steps of encryption table creation, encryption and decryption as a framework. The encryption of the original text is done using the Vigenère square or Vigenère table. When applying to the automatic generation of secret keys on the information system this model is expected that integrated authentication work, and analysis will be possible on target system. ubstitution alphabets[3].

• Journal of Software Assessment and Valuation
• /
• v.14 no.2
• /
• pp.1-9
• /
• 2018

The proportion of users and companies using open source continues to grow. The size of open source software market is growing rapidly not only in foreign countries but also in Korea. However, compared to the continuous development of open source software, there is little research on open source software subject classification, and the classification system of software is not specified either. At present, the user uses a method of directly inputting or tagging the subject, and there is a misclassification and hassle as a result. Research on open source software classification can also be used as a basis for open source software evaluation, recommendation, and filtering. Therefore, in this study, we propose a method to classify open source software by using machine learning model and propose performance comparison by machine learning model.

• Korean Journal of Social Welfare Studies
• /
• v.42 no.2
• /
• pp.235-274
• /
• 2011

The purpose of this study is to understand suffering of torture for victims with suspicion of espionage under the military government in Korea by knowing meaning and structure of empirical phenomena. Methods is to applied to Phenomenological and heuristic Human Becoming Methodology, and the subjects of this study are text for three tortured victims. Results is the structure that the victims accepted their act of espionage under the torture and horror, living with retribution from heaven, surviving pressured times, and fighting for human rights upon release from prison. The conceptual integration of relationship issues were: valuing, imaging, languaging with powering and transforming under the process of revealing-concealing and enabling-limiting. Finally, discussion and practical meaning was reviewed.

• Journal of Digital Convergence
• /
• v.22 no.1
• /
• pp.15-21
• /
• 2024

The purpose of this study is to analyze the use of generative AI and the perception of differences between user groups. This study explored the perceptions of different user groups regarding generative AI, aiming to derive implications for enhancing AI utilization capabilities for each group. Upon analysis, it was found that there were no significant differences in perceptions across age groups. However, notable differences were observed between professional backgrounds, particularly in the areas of generative AI application and ethical perspectives. Consequently, this study suggests the need for diversified AI solutions tailored to specific fields of expertise. It underscores the importance of customized education and training programs, as well as specialized education focused on ethical considerations. Additionally, this research contributes academically by proposing varied AI usage strategies for different age and professional groups. It also highlights the role of text mining techniques in developing and improving AI utilization skills.

• The Transactions of the Korea Information Processing Society
• /
• v.13 no.9
• /
• pp.429-436
• /
• 2024

As Large Language Models (LLMs) continue to advance, effectively harnessing their potential has become increasingly important. LLMs, trained on vast datasets, are capable of generating text across a wide range of topics, making them useful in applications such as content creation, machine translation, and chatbots. However, they often face challenges in generalization due to gaps in specific or specialized knowledge, and updating these models with the latest information post-training remains a significant hurdle. To address these issues, Retrieval-Augmented Generation (RAG) models have been introduced. These models enhance response generation by retrieving information from continuously updated external databases, thereby reducing the hallucination phenomenon often seen in LLMs while improving efficiency and accuracy. This paper presents the foundational architecture of RAG, reviews recent research trends aimed at enhancing the retrieval capabilities of LLMs through RAG, and discusses evaluation techniques. Additionally, it explores performance optimization and real-world applications of RAG in various industries. Through this analysis, the paper aims to propose future research directions for the continued development of RAG models.