• Journal of Intelligence and Information Systems
• /
• v.24 no.2
• /
• pp.85-109
• /
• 2018

Recommender system recommends the items expected to be purchased by a customer in the future according to his or her previous purchase behaviors. It has been served as a tool for realizing one-to-one personalization for an e-commerce service company. Traditional recommender systems, especially the recommender systems based on collaborative filtering (CF), which is the most popular recommendation algorithm in both academy and industry, are designed to generate the items list for recommendation by using 'overall rating' - a single criterion. However, it has critical limitations in understanding the customers' preferences in detail. Recently, to mitigate these limitations, some leading e-commerce companies have begun to get feedback from their customers in a form of 'multicritera ratings'. Multicriteria ratings enable the companies to understand their customers' preferences from the multidimensional viewpoints. Moreover, it is easy to handle and analyze the multidimensional ratings because they are quantitative. But, the recommendation using multicritera ratings also has limitation that it may omit detail information on a user's preference because it only considers three-to-five predetermined criteria in most cases. Under this background, this study proposes a novel hybrid recommendation system, which selectively uses the results from 'traditional CF' and 'CF using multicriteria ratings'. Our proposed system is based on the premise that some people have holistic preference scheme, whereas others have composite preference scheme. Thus, our system is designed to use traditional CF using overall rating for the users with holistic preference, and to use CF using multicriteria ratings for the users with composite preference. To validate the usefulness of the proposed system, we applied it to a real-world dataset regarding the recommendation for POI (point-of-interests). Providing personalized POI recommendation is getting more attentions as the popularity of the location-based services such as Yelp and Foursquare increases. The dataset was collected from university students via a Web-based online survey system. Using the survey system, we collected the overall ratings as well as the ratings for each criterion for 48 POIs that are located near K university in Seoul, South Korea. The criteria include 'food or taste', 'price' and 'service or mood'. As a result, we obtain 2,878 valid ratings from 112 users. Among 48 items, 38 items (80%) are used as training dataset, and the remaining 10 items (20%) are used as validation dataset. To examine the effectiveness of the proposed system (i.e. hybrid selective model), we compared its performance to the performances of two comparison models - the traditional CF and the CF with multicriteria ratings. The performances of recommender systems were evaluated by using two metrics - average MAE(mean absolute error) and precision-in-top-N. Precision-in-top-N represents the percentage of truly high overall ratings among those that the model predicted would be the N most relevant items for each user. The experimental system was developed using Microsoft Visual Basic for Applications (VBA). The experimental results showed that our proposed system (avg. MAE = 0.584) outperformed traditional CF (avg. MAE = 0.591) as well as multicriteria CF (avg. AVE = 0.608). We also found that multicriteria CF showed worse performance compared to traditional CF in our data set, which is contradictory to the results in the most previous studies. This result supports the premise of our study that people have two different types of preference schemes - holistic and composite. Besides MAE, the proposed system outperformed all the comparison models in precision-in-top-3, precision-in-top-5, and precision-in-top-7. The results from the paired samples t-test presented that our proposed system outperformed traditional CF with 10% statistical significance level, and multicriteria CF with 1% statistical significance level from the perspective of average MAE. The proposed system sheds light on how to understand and utilize user's preference schemes in recommender systems domain.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.38 no.4
• /
• pp.25-34
• /
• 2010

This study was conducted to develop evaluation indicators to improve athletic performance and operational management of golf courses and the results of the research are as follows. Through theoretical research and a preliminary professional survey, 15 on-going evaluations of golf course composition and operational management and 55 sub-evaluation indices were rejected while 10 on-going evaluations and 52 sub-evaluation indicators were reconfigured as final for environmental-friendliness, level of member services, level of human service of game personnel, difficulties of course, management level of the course, fairness of operational management, accessibility and location characteristic, traditions and ambiance of the golf club, quality of course, and course layout. When analyzing the important decision factors in golf course user preference evaluation indicators, the following contributed in the order of higher to lower contributions: the management level of the course, excellence of the course, level of human services for personnel, course layout and environmental-friendliness. When identifying the path coefficient of golf course evaluation indicators, the curvature of a hole and the length of the course had a causal effect on the 'course layout' section. Tournament facilities and various shot values had a causal relationship with 'excellence of the course', in the order of higher to lower, and convenience of waiting and fair allocation of reservations for 'fairness of operational management'. The history of the golf course and its environmental characteristics, history and culture of the region have relatively higher causal effects on 'traditions of the golf club' and geographical conditions on 'accessibility and location characteristics', pesticide and fertilizer usage and water pollution on 'environmental-friendliness', and member benefit and kindness of employees on 'level of member services'. The kindness and expertise of the game personnel had a relatively higher causal effect on the 'level of human services of game personnel', the location of tenning area, and location of OB and hazards on 'difficulties of course', and rough conditions and obstacles management on 'management level of the course'. There is a need to complete a systematic evaluation index system for golf course user preferences through future studies for a more detailed assessment, as well as a process to verify these evaluation indicators by application to domestic and international golf courses.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.39 no.4
• /
• pp.513-521
• /
• 2019

The purpose of this research is to find out the preference of users on the characteristics of the subway sign blank. In detail, The purpose is to introduce the concept of designing various characteristics of the subway sign blank according to characteristics such as gender and age, etc. The methodology of this study is to investigate the preference of the characteristics of the characteristics of the subway sign blank space and analyze the preference of the whole group and the cluster group. A survey was conducted to investigate preferences. A cluster analysis was conducted to analyze the preferences. And a demographic and conjoint analysis was conducted for whole group and the cluster group. The attributes of the subway sign blank space for preference survey are as follow : top and bottom blank, side blank, border line blank, arrow thickness, 'station name' and 'line number' order. The results of the preference analysis are as follows. The importance of the attributes in the whole group is shown in the order of the border line blank, 'station name' and 'line number' blank, side blank, top and bottom blank, and arrow thickness. The cluster group is composed of 3 groups, 1 cluster is a woman who uses the subway almost every day, three to four times a week, and seems to prefer half the side blank. 2 crowd is the user who thinks that 60 or more subway signs are uncomfortable, and preferring the order of 'station name' + 'line number' order without border. The 3 clusters were men in their 20s and 30s, with a preference for 1/5 border line blank and thin arrow thickness. The conclusion is as follows. First, the characteristics of the subway sign blank must be designed consistently. However, it is necessary to consider various factors according to gender, age, and frequency of subway use for specific regions or routes. Secondly, It has been shown that, depending on the specific area or route, it is possible to design two or more types of design, not one type of standardized marking of the characteristics of the subway sign blank.

• Journal of Internet Computing and Services
• /
• v.6 no.6
• /
• pp.139-148
• /
• 2005

The desired level for multimedia services in the mobile web environment, the next generation web environment, is expected to be of PC level quality. And great efforts have been made in the development of hadware technology, communication technology, various kinds of services and standardization to support these services, In the mobile web environment, multimedia contents adaptation services should be available through supporting various kinds of devices, network abilities and users' preferences. It means that due to the variety of both desired devices' hardware specifications, called destinations, and desired QoSes, the QoSes in the destinations are not fixed or defined. If a new user wants to stream multimedia contents in a server through a new kind of terminal device, it should be considered whether the existing transcoders are able to adapt the multimedia contents. However, the existing libraries for multimedia adaptation have heavy transcoder figures which include all adaptive functions in one library, The challenge of universal access is too complex to be solved with these all in one solutions. Therefore, in this paper we propose an application independent multimedia adaptation framework which meets the QoS of new and varied mobile devices. This framework is composed of a group of unit transcoders having only one transcoding function respectively, Instead of heavy transcoders. Also, It includes the transcoder manager supporting the dynamic connections of the unit transcoders in order to satisfy end to end QoS.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.47 no.8
• /
• pp.74-84
• /
• 2010

In this paper, we propose a preference-based shortest path algorithm which is combined with Ant Colony Optimization (ACO) and A* heuristic algorithm. In recent years, with the development of ITS (Intelligent Transportation Systems), there has been a resurgence of interest in a shortest path search algorithm for use in car navigation systems. Most of the shortest path search algorithms such as Dijkstra and A* aim at finding the distance or time shortest paths. However, the shortest path is not always an optimum path for the drivers who prefer choosing a less short, but more reliable or flexible path. For this reason, we propose a preference-based shortest path search algorithm which uses the properties of the links of the map. The preferences of the links are specified by the user of the car navigation system. The proposed algorithm was implemented in C and experiments were performed upon the map that includes 64 nodes with 118 links. The experimental results show that the proposed algorithm is suitable to find preference-based shortest paths as well as distance shortest paths.

• Journal of Korea Society of Industrial Information Systems
• /
• v.8 no.3
• /
• pp.76-84
• /
• 2003

Software agents reduce human involvement to a certain extent by automating routine tasks. However, most of agents have assisted with only a few steps in the multi-steps process of electronic transactions. In order to help users with the important steps in the electronic transactions, software agents need to persuade other parties to act in particular ways. While negotiations have many shapes and forms, this paper focuses on a particular class of negotiation, that is competitive business environment based negotiation. For negotiation with other parties in this contort, it is necessary for autonomous agents to consider environmental variables-the number of competitors, the number of negotiation parties, the maximum time by which they must finish their jobs, and user's preferences. Previous negotiation decision functions for the automated negotiation have used only time or the static numbs of negotiating parties as negotiation criteria, although competitive business environment should include potential competitors who can snatch negotiation parties away. This paper attempts to evaluate the performance of a negotiation decision function that considers the potential competitors in competitive market environment as well as that of a negotiation decision function that does not. For this evaluation, this study adopts the electronic marketplace as an application domain because many buyers and sellers compete for limited resources in the marketplace.

• Journal of Broadcast Engineering
• /
• v.17 no.2
• /
• pp.270-283
• /
• 2012

With the advent of multi-channel TV, IPTV and smart TV services, excessive amounts of TV program contents become available at users' sides, which makes it very difficult for TV viewers to easily find and consume their preferred TV programs. Therefore, the service of automatic TV recommendation is an important issue for TV users for future intelligent TV services, which allows to improve access to their preferred TV contents. In this paper, we present a recommendation model based on statistical machine learning using a collaborative filtering concept by taking in account both public and personal preferences on TV program contents. For this, users' preference on TV programs is modeled as a latent topic variable using LDA (Latent Dirichlet Allocation) which is recently applied in various application domains. To apply LDA for TV recommendation appropriately, TV viewers's interested topics is regarded as latent topics in LDA, and asymmetric Dirichlet distribution is applied on the LDA which can reveal the diversity of the TV viewers' interests on topics based on the analysis of the real TV usage history data. The experimental results show that the proposed LDA based TV recommendation method yields average 66.5% with top 5 ranked TV programs in weekly recommendation, average 77.9% precision in bimonthly recommendation with top 5 ranked TV programs for the TV usage history data of similar taste user groups.

• Journal of Internet Computing and Services
• /
• v.18 no.4
• /
• pp.51-59
• /
• 2017

With the growth of Cloud computing, it is important to consider resource management techniques to minimize the overall costs of management. In cloud environments, each host's utilization and virtual machine's request based on user preferences are dynamic in nature. To solve this problem, efficient allocation method of virtual machines to hosts where the classification of virtual machines and hosts is undetermined should be studied. In reducing the number of active hosts to reduce energy consumption, thresholds can be implemented to migrate VMs to other hosts. By using Fuzzy logic in classifying resource requests of virtual machines and resource utilization of hosts, we proposed an adaptive VM allocation and migration approach. The allocation strategy classifies the VMs according to their resource request, then assigns it to the host with the lowest resource utilization. In migrating VMs from overutilized hosts, the resource utilization of each host was used to create an upper threshold. In selecting candidate VMs for migration, virtual machines that contributed to the high resource utilization in the host were chosen to be migrated. We evaluated our work through simulations and results show that our approach was significantly better compared to other VM allocation and Migration strategies.

• Therapeutic Science for Rehabilitation
• /
• v.9 no.3
• /
• pp.23-34
• /
• 2020

Objective : Recently, the demand for wearable mobile devices for the monitoring and improvement of one's health and quality of life is increasing. Therefore, the purpose of this study is to analyze the need of potential users in order to develop mobile health based and lifestyle care application for the elderly. Methods : Participants who lived in their community and used a mobile phone were recruited. Finally, a total of 84 participants completed the survey. Data were analyzed using descriptive statistics and a t-test, which was carried out with SPSS version 25.0. Results : The application functions that users deemed important for a lifestyle care app were the number of daily steps, physical activity, blood pressure, sleep, nutrition and participation in activity. Interestingly, there was a significant difference in the importance given to the app function of participation in activities between age groups. Conclusion : This study investigated the need and preferences of potential users of health and lifestyle care application for the promotion of health among the elderly. The, findings obtained from this study could be a valuable resource for the development of lifestyle care application.

• Knowledge Management Research
• /
• v.21 no.3
• /
• pp.27-44
• /
• 2020

In order to solve the user's information overload problem, recommender systems infer users' preferences and suggest items that match them. The collaborative filtering (CF), the most successful recommendation algorithm, has been improving performance until recently and applied to various business domains. Visual information, such as book covers, could influence consumers' purchase decision making. However, CF-based recommender systems have rarely considered for visual information. In this study, we propose VizNCS, a CF-based deep learning model that uses visual information as additional information. VizNCS consists of two phases. In the first phase, we build convolutional neural networks (CNN) to extract visual features from image data. In the second phase, we supply the visual features to the NCF model that is known to easy to extend to other information among the deep learning-based recommendation systems. As the results of the performance comparison experiments, VizNCS showed higher performance than the vanilla NCF. We also conducted an additional experiment to see if the visual information affects differently depending on the product category. The result enables us to identify which categories were affected and which were not. We expect VizNCS to improve the recommender system performance and expand the recommender system's data source to visual information.