• 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.

• Journal of Digital Convergence
• /
• v.10 no.4
• /
• pp.147-157
• /
• 2012

In this study, User intention also factors that Consulting application and Subjective norm, Task media fit factors also impact on User intention was examined. In this process, especially Subjective norm and Task media fit of SNS in each dimension to User intention about the impact of Consulting application was in each carried out to examine the mediating effect. The test result used in this study the Subjective norm, Task media fit and Consulting application in a significant positive(+) User intention also was effective. And verifying the results of the mediating effect of Consulting application, Subjective norm and Task media fit has proven to be effective in all parameters. The results of such individuals or companies with Consulting application of SNS. SNS Subjective norm and Task media fit are aware of the factors influencing the Strategy envisions an appropriate response, such as to provide Suggestions.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.4 no.6
• /
• pp.1237-1252
• /
• 2010

This paper proposes an intelligent technique for data sharing and event transition among personal devices including smart phones, laptops, and desktops. We implemented the PES (Personal Event Service) system that proactively executes appropriate tasks across multiple devices without explicit user requests by sharing the data used by the user and recognizing user intention based on the observed actions of the user for specific devices. The client module of PES installed on each device monitors the user actions and recognizes the intention of the user. The server provides data sharing and maintenance for clients. The connection between client and server is established by Java RMI (Remote Method Invocation). A series of experiments were performed to evaluate user satisfaction and system accuracy, and the results showed that the PES system can proactively provide appropriate, personalized services with a high degree of satisfaction to the user in an effective and efficient manner.

• Journal of the Ergonomics Society of Korea
• /
• v.18 no.3
• /
• pp.55-72
• /
• 1999

Evaluation of user interfaces has to be performed in an intuitive and subjective manner by experts especially when the problem comes to the complexity and consistency of sophisticated interface procedures. The manual analysis and evaluation of logical interfaces also tends to be slow and laborious. To make the evaluation more formal and objective, the criteria and the procedure with which the evaluation can proceed must be explicitly prescribed. Furthermore, to make the formal procedure as practical as subjective expert evaluation, the criteria must reflect the user knowledge of the target tasks since the task knowledge plays the role of a basis for users to understand the interface organizations and procedures. This paper describes ESTIM, a support system for task-based evaluation of user interface, that includes the interface evaluation criteria and implies an evaluation procedure. The support system can be used either in an interactive manner by the analyst during the evaluation or in an automatic evaluation mode. It was verified that the result of automatic evaluation by ESTIM matched the results of expert evaluation fairly well.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.6 no.2
• /
• pp.1-9
• /
• 2010

The increasing popularity of mobile devices, such as cellular phones, smart phones, and PDAs, has fostered the need to recommend more effective information in ubiquitous environments. We propose the recommendation method for mobile contents service using contexts and prefetching in ubiquitous environment. The proposed method enables to find some relevant information to specific user's contexts and computing system contexts. The prefetching has been applied to recommend to user more effectively. Our proposed method makes more effective information recommendation. The proposed method is conceptually comprised of three main tasks. The first task is to build a prefetching zone based on user's current contexts. The second task is to extract candidate information for each user's contexts. The final task is prefetch the information considering mobile device's resource. We describe a new recommendation.

• Journal of Information Technology Applications and Management
• /
• v.12 no.2
• /
• pp.145-161
• /
• 2005

In this paper, we present an experiment conducted to examine how the web site's architecture, content quantity and the user task complexity affect its usability. The experiment was performed in two phases on college students to visit existing web sites which are readily accessible via the Internet. The results of experiment show that the web site's architecture significantly affects the efficiency and the effectiveness interactively with its content quantity and the user task's complexity. On the other hand, none of the above three factors show significant effects on the satisfaction and the learnability. In particular, the web site's content quantity does not have any statistically significant effects on the satisfaction and the learnability. It implies that the factors affecting the satisfaction and the learnability are different from those affecting the efficiency and the effectiveness. The analysis reassures that it is essential to consider the context of use in designing a web site.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.11 no.9
• /
• pp.1742-1748
• /
• 2007

In order to provide the convenient access to the mobile services and the efficient utilization of the mobile devices, it is required to devise an intelligent user interface which guarantees the efficient task selection and transition in the limited input/output environments of the mobile devices. In this paper, we propose user task prediction models which are essential for the intelligent user interface, and an accuracy estimation model is also proposed for evaluating the prediction models.

• Journal of Korea Multimedia Society
• /
• v.24 no.2
• /
• pp.320-327
• /
• 2021

We aim to investigate the effects of task type and user gender on the expected gender of chatbots. We conducted an online study of 381 participants who selected the gender (female, male, or neutral) for chabots performing six different tasks. Our results indicate that users expect human- gendered chatbots for all tasks and that the expected gender of a chatbot is significantly different depending on the task type. Users expected chatting, counseling, healthcare and clerical work to be done by female chatbots; professional and customer service work were expected to be done by male chatbots. A tendency for participants to prefer chatbots of the same-gendered as themselves is revealed in several tasks for both male and female users. However, this homophily tendency is stronger for female users. We conclude by suggesting practical guidelines for designing chatbot services that reflect user expectations.

• Asia pacific journal of information systems
• /
• v.5 no.2
• /
• pp.259-288
• /
• 1995

Using end-user attitude as intervening variable, this paper proposes a causal model of how environmental factors of EUC influence end-user computing (EUC) success, measured by the degree of IS use. This study identifies unique aspects of the EUC environment and generates key dimensions of end-user attitude appropriate for this context : perceived information quality, perceived design quality, and perceived usefulness. The research model includes six environmental factors - top management support, end-user education, task variability, task analyzability, end-user's computing ability, and end-user involvement. A field study was undertaken to test the hypothesized relationships among the research variables. The results generally support the assumption that end-user attitude intervene the impact of environmental factors on IS use. The implications of the findings are discussed.

• Journal of Information Technology Applications and Management
• /
• v.12 no.1
• /
• pp.213-229
• /
• 2005

Cost overrun or schedule delay of the software development project happens frequently despite that software developers continue to make every effort for the effective management of the projects. Previous researches have ascertained that these problems are ascribed to the uncertainty of projects and the improper management of the projects. The purposes of this research are to investigate the impacts of user participation and task interdependence on the performance of the projects and also to find out the appropriate project management method to improve the project performance. Even though the model fitness of the path model is proved to be very high, the verification of the hypotheses showed a variety of results including the four verifications and the one refutation of the hypotheses as well as the suggestion of one alternative hypothesis. The contribution of this research is that the integration model is proposed and verified, comprising the relationship among the user participation, the task interdependence, and the performance of software development projects. A project manager can utilize the implication of this research for an effective management of software development project.