• Knowledge Management Research
• /
• v.21 no.1
• /
• pp.117-136
• /
• 2020

In the studies for the recommender systems which solve the information overload problem of users, the use of transactional data has been continuously tried. Especially, because the firms can easily obtain transactional data along with the development of IoT technologies, transaction-based recommender systems are recently used in various areas. However, the use of transactional data has limitations that it is hard to reflect domain knowledge and they do not directly show user preferences for individual items. Therefore, in this study, we propose a method applying the word embedding in the transaction-based recommender system to reflect preference differences among users and domain knowledge. Our approach is based on SAR, which shows high performance in the recommender systems, and we improved its components by using FastText, one of the word embedding techniques. Experimental results show that the reflection of domain knowledge and preference difference has a significant effect on the performance of recommender systems. Therefore, we expect our study to contribute to the improvement of the transaction-based recommender systems and to suggest the expansion of data used in the recommender system.

• Journal of the Korean Society for information Management
• /
• v.23 no.2
• /
• pp.167-183
• /
• 2006

This paper describes a metadata extraction technique based on natural language processing (NLP) which extracts personalized information from email communications between financial analysts and their clients. Personalized means connecting users with content in a personally meaningful way to create, grow, and retain online relationships. Personalization often results in the creation of user profiles that store individuals' preferences regarding goods or services offered by various e-commerce merchants. We developed an automatic metadata extraction system designed to process textual data such as emails, discussion group postings, or chat group transcriptions. The focus of this paper is the recognition of emotional contents such as mood and urgency, which are embedded in the business communications, as metadata.

• Journal of The Korean Digital Architecture Interior Association
• /
• v.9 no.2
• /
• pp.35-43
• /
• 2009

The purpose of this study is to know the way of practical space use and user behavior pattern at the livingroom of city apartment that is the representative residential space of modern citizen. The way of study is to compare the livingroom of model house with actual condition of livingroom, through this way, we will know that livingroom is in use as the original concept of design or not. From the research which sees consequently (1) Most of the current model houses show common kind of livingroom by using typical style and arrangement of furniture (TV, couch, table, and decorations such as pictures or other artworks). (2) Actual condition of livingroom is different from model house in furniture arrangement and in using space which is set depending on the residents' preferences and characteristics.(computer, desk, exercising equipments, and instruments etc.) (3) The actual condition of livingroom shows the various behavior pattern of space use as the actual condition of livingroom is a mixture of typical kind of livingroom and the livingroom that reflects the characteristics of residents'.

• Journal of Information Technology Applications and Management
• /
• v.21 no.1
• /
• pp.85-105
• /
• 2014

Exhibition industry is important business domains to many countries. Not only lots of countries designated the exhibition industry as tools to stimulate national economics, but also many companies offer millions of service or products to customers. Recommender systems can help visitors navigate through large information spaces of various booths. However, no study before has proposed a methodology for identifying and acquiring prospective visitors although it is important to acquire them. Accordingly, we propose a methodology for identifying, acquiring prospective visitors, and recommending the adequate booth information to their preferences in the exhibition industry. We assume that a visitor will be interested in an exhibition within same class of exhibition taxonomy as exhibition which the visitor already saw. Moreover, we use user-based collaborative filtering in order to recommend personalized booths before exhibition. A prototype recommender system is implemented to evaluate the proposed methodology. Our experiments show that the proposed methodology is better than the item-based CF and have an effect on the choice of exhibition or exhibit booth through automation of word-of-mouth communication.

• The KIPS Transactions:PartB
• /
• v.12B no.3 s.99
• /
• pp.313-322
• /
• 2005

In order to access digital items in a server via ubiquitous devices, the digital items should be adapted according to the system environment, device characteristics and user preferences. In ubiquitous environment, those device-dependent adaptation requirements are not statically determined and not predictable. Therefore an application specific adaptation mechanism can not be applied to a general digital item adaptation engine. In this paper, we propose an application independent digital item adaptation architecture which has a set of minimal transcoders, transcoding path generator for a required adaptation requirement, and adaptation scheduler. And a CFG based method of finding a sequence of multiple unit transcoders called a transcoding path Is described in detail followed by experimental results.

• Journal of Korean Society for Quality Management
• /
• v.44 no.4
• /
• pp.951-963
• /
• 2016

Purpose: This study was conducted to analyze the potential customer's satisfaction for the concepts of smart audio features and utilize the results when developing the customer-oriented products. Methods: 16 different features were derived via the market research and professionals' interviews. The most satisfactory features were selected through "Kano model", the relative importance of Customer Satisfaction Coefficient, and respondents' preferences from 339 valid survey answers. Results: 15 out of the 16 features were categorized as attractive attribute. "'User Recognizing' and 'Strengthen Linking' groups", such as Auto connection with Smart-phone music player, Synchronization of TV & Audio, and Volume control situational awareness, were shown to provide higher satisfactions to those potential customers. On the other hand, Group 'Integrating Function', such as Aromatherapy and Auto lighting reaction, was shown to be relatively least preferred features. Conclusion: This study enabled which features could lead to the customer satisfaction. Nevertheless, it still requires extensive analyses in different countries and diverse cultures to target the global market. The audio product planners and R&D professionals are expected to learn useful information from such studies.

• Journal of the Korean Society of Physical Medicine
• /
• v.10 no.1
• /
• pp.23-35
• /
• 2015

PURPOSE: People with physical disabilities such as cerebral palsy usually experience obstacles when interacting with computer through conventional keyboard because of their motor disabilities. The purpose of this study is empirically compare of text entry(alphabet and word) speed and accuracy using the three different keyboard type on four students(male 2 and female 2) with cerebral palsy. METHODS: This research design used a replicated single-case experimental approach to compare the individual performance. An alternating treatments design was used to examine the effectiveness of standard QWERTY keyboard and alternative keyboard(mini and big keyboard) on computer access for students with cerebral palsy. To avoid changes in posture that influence a keyboard character entry training and evaluation was carried out using his sitting in a wheelchair. Compass software program used in this study as an assessment tool to measure speed and accuracy when performance of text entry(alphabet and word). This was repeated until the stable status of reaction time. RESULTS: As a result, the alternative keyboard seems to be the most effective device for students with cerebral palsy to perform text entry. But various factors such as peculiarity of motor disabilities, experience and preferences of the user are heavily related. CONCLUSION: Thus, we must perform the objective and systematic assessment for computer access and if sustained training is accomplished, it could to improve speed and accuracy of text entry(alphabet and word).

• Journal of Korean Society for Geospatial Information Science
• /
• v.19 no.1
• /
• pp.13-20
• /
• 2011

LBS services is a user-centric location based information service, where its importance has been discussed as an essential engine in an Ubiquitous Age. We aimed to develop an ontology reasoning system that enables users to derive recommended results suitable through selection standard reasoning according to various users' preferences. In order to achieve this goal, we designed the Geo-ontology system which enabled the construction of personal characteristics of users, knowledge on personal preference and knowledge on spatial and geographical preference. We also integrated a function of reasoning relevant information through the construction of Cost Value ontology using multi-criteria decision making by giving weight according to users' preference.

• Korean Institute of Interior Design Journal
• /
• v.14 no.3 s.50
• /
• pp.216-223
• /
• 2005

In this study, we would like to suggest a virtual model house where the users participate in the planning of the apartment directly in order to supply the apartment which is suitable for the fast change of the resident's characterization, multiplication, and preference in this like high-speed information society. The users have various preferences and demands in the high level information society. It is necessary for the architect to communicate with the users in order to suggest the various styles of habitats that one can treat various life style and life cycle which reflects the demand. In addition, the object of this paper is the construction and using of the participating virtual model house that the resident can plan the preferred space design of the main house before moving Into the apartment and change space design of the main house after moving into the apartment following the variety of life style and the changing of life cycle. Regarding the effect of this study, firstly, it is possible to supply the plate type following the user's preference in order to enhance the quality of uniformed type apartment. Secondly, it is possible to maintain and control the high quality building using sharing the building information with planning, construction, maintain, interior, and related company by constructing the database. Thirdly, the unified database which is constructed by XML can supply the effective environment of apartment information exchange between other companies.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2017.05a
• /
• pp.87-89
• /
• 2017

In this paper, we propose a multi-band network selection (MNS) system, which determines and seamlessly connects to the desired available communication band based on user preferences and network environments. The MNS system employs MNS server in the middleware layer which communicates with clients in order to combine and manage heterogeneous networks efficiently. In addition, we define a system architecture and function modules of the MNS system. The MNS system monitors availability in each communication band and updates the network status information table which is applied to determining the best available network. It is expected that the MNS system can be applied to the next generation maritime networks and communication infrastructure of e-navigation.