• Journal of the Korea Society of Computer and Information
• /
• v.16 no.2
• /
• pp.83-91
• /
• 2011

Much research has been conducted on the e-learning systems for recommending a learning content to a student based on the difficulty of it. The difficulty is one of the most important factors for selecting a learning content. In the existing learning content recommendation systems, the difficulty of a learning content is determined by the creator. Therefore, it is not easy to apply a standard rule to the difficulty as it is determined by a subjective method. In this paper, we propose an ontology-based method for determining the difficulty of a learning content in order to provide an objective measurement. Previously, ontologies and knowledge maps have been used to recommend a learning content. However, their methods have the same problem because the difficulty is also determined by the creator. In this research, we use an ontology representing the IS-A relationships between words. The difficulty of a learning content is the sum of the weighted path lengths of the words in the learning content. By using this kind of difficulty, we can provide an objective measurement and recommend the proper learning content most suitable for the student's current level.

• The Journal of Information Systems
• /
• v.27 no.2
• /
• pp.157-173
• /
• 2018

Purpose Product characteristics and price value in website have strongly effects on customer satisfaction. Especially, in the online shopping site, the scarcity limits the customer's opportunity to purchase the product. Thus scarcity has been proposed as a important factor that makes the customer highly aware of the merchantability of the product. The scarcity in the web store is used as an important variable to make purchasing decisions of users easier by psychological pressure. In the case of scarce products with price discounts in online commerce, advertising formats that highlight scarcity value in the web commerce market are very effective in enhancing purchase intentions of consumers. Unlike offline stores, the importance of scarcity becomes more important when reflecting the characteristics of online commerce. Therefore, this study intends to confirm the influence of the degree of price discounts and scarcity information presented by Web sites on consumer purchase behavior in Web purchase behavior. Design/methodology/approach This study conducted a web-based experimental study on price sensitivity and price discount. Therefore, we created experimental web-sites that offer two stimuli according to the discount rate. The 200 respondents were randomly assigned. The stimuli were fictitious based on tourism products. The first stimulus presented the price discount(15% discount) with basic explanation about the package of the tourist package. The stimuli assigned to the second group were used for groups with high price discount intensity(65% discount). In this way, the two stimuli clearly distinguished the level of price discount intensity. This paper conducted t-test analysis and structural equation to analyze the experiemental results after confirming the reliability and validity. Findings The results of this study are as follows. The difference in price discount intensity (15% vs 65%) with scarcity showed the mean difference among all the variables. Therefore, this study concluded that there is a significant difference between the price discount of 15% and 65% for the acquisition value and transaction value of users. In particular, consumers' purchase intention is greater and product recommendation intensity is stronger when the price discount is 65%. As a result, the high degree of the price discount intensity with scarcity exerts a greater influence on consumers' purchase intentions. Product scarcity also have a significant impact on perceived value of users. Therefore, purchase intention of customers increases when perceived value increases their profit and pleasure feeling.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.10 no.3
• /
• pp.511-516
• /
• 2006

In this raper, we propose the Sequentiel Couleur Avec Memoire or Sequential Color with Memory (SECAM) video encoder system using modified anti-cloche filters and the SECAM video decoder system using a band pass filter (BPF) and an error-free square root. The SECAM encoder requires an anti-cloche filter recommended by International Telecommunication Union-Recommendation (ITU-R) Broadcasting service Television (BT) 470. However, the design of the anti-cloche filter is difficult because the frequency response of the anti-cloche filter is very sharp around rejection-frequency area. So, we convert the filter into a hish pass filter (HPF) by shifting the rejection frequency of 4.286MHz to 0Hz frequency. The design of HPF becomes very easy, compared to that of the anti-cloche filter. The proposed decoder also uses an error-free square root, two differentiators and trigonometric functions to extract color-component information of Db and Dr accurately from frequency modulation (FM) signals in SECAM systems. Also, the BPF in decoder it used for removing color noise in chrominance and dividing CVBS into chrominance and luminance. The proposed systems are experimentally demonstrated with Altera FPGA APEX20KE EP20K1000EBC652-3 device and TV sets.

• KIISE Transactions on Computing Practices
• /
• v.21 no.2
• /
• pp.101-108
• /
• 2015

This paper presents the implementation of a system that recommends empirical best routes based on classification of large trajectory data. As many location-based services are used, we expect the amount of location and trajectory data to become big data. Then, we believe we can extract the best empirical routes from the large trajectory repositories. Large trajectory data is clustered into similar route groups using Hadoop MapReduce framework. Clustered route groups are stored and managed by a DBMS, and thus it supports rapid response to the end-users' request. We aim to find the best routes based on collected real data, not the ideal shortest path on maps. We have implemented 1) an Android application that collects trajectories from users, 2) Apache Hadoop MapReduce program that can cluster large trajectory data, 3) a service application to query start-destination from a web server and to display the recommended routes on mobile phones. We validated our approach using real data we collected for five days and have compared the results with commercial navigation systems. Experimental results show that the empirical best route is better than routes recommended by commercial navigation systems.

• The KIPS Transactions:PartD
• /
• v.16D no.3
• /
• pp.451-462
• /
• 2009

Collaborative Filtering (CF), one of the most successful technologies among recommender systems, is a system assisting users in easily finding the useful information. However, despite its success and popularity, CF encounters a serious limitation with quality evaluation, called cold start problems. To alleviate this limitation, in this paper, we propose a unique method of building models derived from explicit ratings and applying the models to CF recommender systems. The proposed method is divided into two phases, an offline phase and an online phase. First, the offline phase is a building pre-computed model phase in which most of tasks can be conducted. Second, the online phase is either a prediction or recommendation phase in which the models are used. In a model building phase, we first determine a priori predicted rating and subsequently identify prediction errors for each user. From this error information, an error-reflected model is constructed. The error-reflected model, which is reflected average prior prediction errors of user neighbors and item neighbors, can make accurate predictions in the situation where users or items have few opinions; this is known as the cold start problems. In addition, in order to reduce the re-building tasks, the error-reflected model is designed such that the model is updated effectively and users'new opinions are reflected incrementally, even when users present a new rating feedback.

• Journal of Intelligence and Information Systems
• /
• v.29 no.4
• /
• pp.129-164
• /
• 2023

In a situation where the use and introduction of Large Language Models (LLMs) is expanding due to recent developments in generative AI technology, it is difficult to find actual application cases or implementation methods for the use of internal company data in existing studies. Accordingly, this study presents a method of implementing generative AI services using the LLM application architecture using the most widely used LangChain framework. To this end, we reviewed various ways to overcome the problem of lack of information, focusing on the use of LLM, and presented specific solutions. To this end, we analyze methods of fine-tuning or direct use of document information and look in detail at the main steps of information storage and retrieval methods using the retrieval augmented generation (RAG) model to solve these problems. In particular, similar context recommendation and Question-Answering (QA) systems were utilized as a method to store and search information in a vector store using the RAG model. In addition, the specific operation method, major implementation steps and cases, including implementation source and user interface were presented to enhance understanding of generative AI technology. This has meaning and value in enabling LLM to be actively utilized in implementing services within companies.

• Information Systems Review
• /
• v.24 no.1
• /
• pp.1-19
• /
• 2022

Social media-based communication has become crucial part of our personal and official lives. Therefore, it is no surprise that social media sentiment analysis has emerged an important way of detecting potential customers' sentiment trends for all kinds of companies. However, social media sentiment analysis suffers from huge number of sentiment features obtained in the process of conducting the sentiment analysis. In this sense, this study proposes a novel method by using Bayesian Network. In this model MBFS (Markov Blanket-based Feature Selection) is used to reduce the number of sentiment features. To show the validity of our proposed model, we utilized online review data from Yelp, a famous social media about restaurant, bars, beauty salons evaluation and recommendation. We used a number of benchmarking feature selection methods like correlation-based feature selection, information gain, and gain ratio. A number of machine learning classifiers were also used for our validation tasks, like TAN, NBN, Sons & Spouses BN (Bayesian Network), Augmented Markov Blanket. Furthermore, we conducted Bayesian Network-based what-if analysis to see how the knowledge map between target node and related explanatory nodes could yield meaningful glimpse into what is going on in sentiments underlying the target dataset.

• Information Systems Review
• /
• v.22 no.2
• /
• pp.25-41
• /
• 2020

Along with the expectation about 5G network commercialization, interests in realistic and immersive media industries such as virtual reality (VR) are increasing. However, most of studies on VR still focus on video technologies instead of factors for popularity and consumption. Thus, the main objective of this research is to identify meaningful factors, which affect the view counts of VR videos and to provide business implications of the content strategies for VR video creators and service providers. Using a regression analysis with 700 VR videos, this study tries to find major factors that affect the view counts of VR videos. As a result, user assessment factors such as number of likes and sicknesses have a strong influence on the view counts. In addition, the result shows that both general information factors (video length and age) and content characteristic factors (series, one source multi use (OSMU), and category) are all influential factors. The findings suggest that it is necessary to support recommendation and curation based on user assessments for increasing popularity and diffusion of VR video streaming.

• Information Systems Review
• /
• v.20 no.1
• /
• pp.1-16
• /
• 2018

The development and generalization of the Internet increased the popularity of food delivery service applications in Korea. The food delivery market based on online-to-offline service is growing rapidly. This study compares the usability of Korean food delivery service application between that of Chinese food delivery service application. This study suggests improvement points for Korean food delivery service applications. To conduct this study, we explore the status of various food delivery service applications and conduct interviews and surveys based on the honeycomb model developed by Peter Morville. This study obtained the following results. First, all restaurants participating in the Korean food delivery service must be able to accept order through the application. Second, the shopping cart function must be able to accept order of all restaurants simultaneously. Third, when users look for menu recommendation, their purchase history and shopping cart functions should appear at the first page of the website. Users should be able to perceive the improved usability of the website using those functions. Fourth, when the search window is fixed on the top of each page, users should be able to find the information they need. Fifth, the application must allow users to find the exact location of the delivery person and the estimated delivery time. Finally, the restaurants'address should be disclosed and fast delivery time should be confirmed to enhance users'trust on the application. This study contributes to academia and industry by suggesting useful insight into food delivery service applications and improving the point of food delivery service application in Korea.

• Journal of the korean academy of Pediatric Dentistry
• /
• v.31 no.4
• /
• pp.636-644
• /
• 2004

Recently, self-etching adhesive system have been developed and bonding procedures simplified into one or two steps, which are simultaneously applied to both enamel and dentin. These systems are easy to use and have the potential for good clinical success. The purpose of this study is to evaluate in vitro the microleakage on the cementum/dentin and enamel walls in composite resin restoration of Class V cavities, regarding the use of different adhesive systems. 30 human premolars were divided into 3 groups. A standardized Class V preparation was prepared on the buccal and lingual surface of each premolar. The preparation were made parallel to the cementoenamel junctions, with the gingival half of the preparation extending 1mm apical to the cementoenamel junction. After adhesive system was applied to teeth as manufacture's recommendation, hybrid resin composite was filled in bulk into the preparation and light polymerized according to manufacturer's recommendations. Specimen were stored in distilled water at $37^{\circ}C$ for 5 days and thermocycled 1000 times ($5^{\circ}C{\pm}2^{\circ}C\;and\;55^{\circ}C{\pm}2^{\circ}C)$, then immersed in a 2% methylene blue solution for 12 hours. After sectioning mesio distally through the restorations, the degree of dye penetration was scored under a stereomicroscope at ${\times}\;25$ magnification. The data were analyzed statistically using t-test and one-way ANOVA. The results were as follows: ${\cdot}$ There is no adhesive system which can prevent microleakage perfectly. ${\cdot}$ There is significant difference in microleakage between enamel margin and dentin margin (p<0.0001). ${\cdot}$ In enamel margin, self-etching primer systems did not show any significant difference comparing total-etching system. In denin margin, self-etching primer systems did not show any significant difference comparing one-bottle adhesive system used in combination with total-etching.