• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.5
• /
• pp.1413-1432
• /
• 2023

Recommender systems aim to recommend items to users by taking into account their probable interests. This study focuses on creating a model that utilizes multiple sources of information about users and items by employing a multimodality approach. The study addresses the task of how to gather information from different sources (modalities) and transform them into a uniform format, resulting in a multi-modal feature description for users and items. This work also aims to transform and represent the features extracted from different modalities so that the information is in a compatible format for integration and contains important, useful information for the prediction model. To achieve this goal, we propose a novel multi-modal recommendation model, which involves extracting latent features of users and items from a utility matrix using matrix factorization techniques. Various transformation techniques are utilized to extract features from other sources of information such as user reviews, item descriptions, and item categories. We also proposed the use of Principal Component Analysis (PCA) and Feature Selection techniques to reduce the data dimension and extract important features as well as remove noisy features to increase the accuracy of the model. We conducted several different experimental models based on different subsets of modalities on the MovieLens and Amazon sub-category datasets. According to the experimental results, the proposed model significantly enhances the accuracy of recommendations when compared to SVD, which is acknowledged as one of the most effective models for recommender systems. Specifically, the proposed model reduces the RMSE by a range of 4.8% to 21.43% and increases the Precision by a range of 2.07% to 26.49% for the Amazon datasets. Similarly, for the MovieLens dataset, the proposed model reduces the RMSE by 45.61% and increases the Precision by 14.06%. Additionally, the experimental results on both datasets demonstrate that combining information from multiple modalities in the proposed model leads to superior outcomes compared to relying on a single type of information.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.7
• /
• pp.1706-1725
• /
• 2024

The unique U-shaped structure of U-Net network makes it achieve good performance in image segmentation. This network is a lightweight network with a small number of parameters for small image segmentation datasets. However, when the medical image to be segmented contains a lot of detailed information, the segmentation results cannot fully meet the actual requirements. In order to achieve higher accuracy of medical image segmentation, a novel improved U-Net network architecture called multi-scale encoder-decoder U-Net+ (MEDU-Net+) is proposed in this paper. We design the GoogLeNet for achieving more information at the encoder of the proposed MEDU-Net+, and present the multi-scale feature extraction for fusing semantic information of different scales in the encoder and decoder. Meanwhile, we also introduce the layer-by-layer skip connection to connect the information of each layer, so that there is no need to encode the last layer and return the information. The proposed MEDU-Net+ divides the unknown depth network into each part of deconvolution layer to replace the direct connection of the encoder and decoder in U-Net. In addition, a new combined loss function is proposed to extract more edge information by combining the advantages of the generalized dice and the focal loss functions. Finally, we validate our proposed MEDU-Net+ MEDU-Net+ and other classic medical image segmentation networks on three medical image datasets. The experimental results show that our proposed MEDU-Net+ has prominent superior performance compared with other medical image segmentation networks.

• ETRI Journal
• /
• v.36 no.3
• /
• pp.385-395
• /
• 2014

In this paper, we consider a resource allocation scheme for millimeter wave-based wireless personal area networks using directional antennas. This scheme involves scheduling the reservation period of medium access control for IEEE 802.15.3c. Objective functions are considered to minimize the average delay and maximize throughput; and two scheduling algorithms-namely, MInMax concurrent transmission and MAxMin concurrent transmission-are proposed to provide a suboptimal solution to each objective function. These are based on an exclusive region and two decision rules that determine the length of reservation times and the transmission order of groups. Each group consists of flows that are concurrently transmittable via spatial reuse. The algorithms appropriately apply two decision rules according to their objectives. A real video trace is used for the numerical results, which show that the proposed algorithms satisfy their objectives. They outperform other schemes on a range of measures, showing the effect of using a directional antenna. The proposed scheme efficiently supports variable bit rate traffic during the reservation period, reducing resource waste.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2005.10a
• /
• pp.175-181
• /
• 2005

This paper describes the development of on-line computer vision laboratories to teach the detailed image processing and pattern recognition techniques. The computer vision laboratories include distant image acquisition method, basic image processing and pattern recognition methods, lens and light, and communication. This study introduces a case study that teaches computer vision in distance learning environment. It shows a schematic of a distant loaming workstation and contents of laboratories with image processing examples. The study focus more on the contents of the vision Labs rather than internet application method. The study proposes the ways to improve the on-line computer vision laboratories and includes the further research perspectives

• International journal of advanced smart convergence
• /
• v.9 no.1
• /
• pp.90-97
• /
• 2020

Oncolytic viruses are characterized by their ability to selectively kill cancer cells, and thus they have potential for application as novel anticancer agents. Despite an increase in the number of studies on methodologies involving oncolytic viruses, bioinformatic studies generating useful data are lacking. We constructed a database for oncolytic virus research (the oncolytic virus database, OVDB) by integrating scattered genetic information on oncolytic viruses and proposed a systematic means of using the biological data in the database. Our database provides data on 14 oncolytic viral strains and other types of viruses for comparative analysis. We constructed the OVDB using the basic local alignment search tool, and therefore can provides genetic information on highly homologous oncolytic viruses. This study contributes to facilitate systematic bioinformatics research, providing valuable data for development of oncolytic virus-based anticancer therapies.

• International journal of advanced smart convergence
• /
• v.9 no.4
• /
• pp.167-172
• /
• 2020

In the 4th industrial revolution, there are many projects for diverse software applications of smart city environments. Most of the stakeholders focus on considering software quality for their developed software. Nobody doesn't guarantee requirement satisfaction after complete development. At this time, we can only work on user acceptance testing for requirement satisfaction on frequently changing requirements. Why keeps the requirement traceability? This traceability is to identify risks related to requirements, to assure correct software development based on customer requirements. To solve this, we are researching how to implement requirement traceability across each artifact's relationship to each activity of a whole development lifecycle.

• International journal of advanced smart convergence
• /
• v.10 no.2
• /
• pp.130-137
• /
• 2021

In the current software project, it is still very difficult to extract and define clear requirements in the requirement engineering. Informal requirements documents based on natural language can be interpreted in different meanings depending on the degree of understanding or maturity level of the requirements analyst. Also, Furthermore, as the project progresses, requirements continue to change from the customer. This change in requirements is a catastrophic failure from a management perspective in software projects. In the situation of frequent requirements changes, a current issue of requirements engineering area is how to make clear requirements with unclear and ambigousrequirements. To solve this problem, we propose to extract and redefine clear requirements by incorporating Design Thinking methodologies into requirements engineering. We expect to have higher possibilities to improve software quality by redefining requirements that are ambiously and unclearly defined.

• Journal of KIISE:Software and Applications
• /
• v.30 no.11
• /
• pp.1080-1091
• /
• 2003

Recently, the distance education has been rapidly proliferated with the rapid growth of the Internet and high speed networks. There has been relatively much research with regard to online lecture (teaching and studying) tools for the distance education, compared to the virtual laboratory tools (for self-study and experiments). In this paper, we design and implement a web-based simulation tool, named as SimDraw, for the virtual laboratory in the distance education. To apply the web-based simulation technology into the distance education, some requirements should be met; firstly, the user interface of the simulation should be very easy for students. Secondly, the simulation should be very portable to be run on various computer systems of remote students. Finally, the simulation program on remote computers should be very thin so that students can easily install the program onto their computers. To meet these requirements, SimDraw adopts the client/server architecture; the client program contains only model development and animation functions so that no installation of a client program onto student's system is required, and it can be implemented by a Java applet in Web browsers. The server program supports client programs by offering the functions such as remote compiling, model storing, library management, and user management. For the evaluation of SimDraw, we show the simulation process using the example experimentation of the RIP(Routing Information Protocol) Internet routing protocol.

• Journal of the Korea Society for Simulation
• /
• v.12 no.1
• /
• pp.1-10
• /
• 2003

Recently, various network protocols have been developed to keep step with the rapid growth of Internet users. For the education of students in the computer networks classes, there have been many attempts to allow students experiment routers setting, operations and management of networks by themselves, in addition to the classroom lectures based on textbooks. One of the attempts is to install expensive real router experiment sets in laboratories for students, and one another is using router and network simulators for virtual experiments. This paper presents the design and implementation of NetSim, a scalable, component-based simulator environment for the network experimentation. NetSim expands the existing component-based JavaSim simulation tool for the education purpose, and it allows the design and experiment of various networks and protocols. For the evaluation of NetSim, it is shown that the network design and RIP-based router setting process is possible in NetSim.

• The Journal of the Acoustical Society of Korea
• /
• v.28 no.3E
• /
• pp.99-108
• /
• 2009

In this paper, we suggest criteria for objective measurement of speech quality in mobile VoIP (Voice over Internet Protocol) services over wireless mobile internet such as mobile WiMAX networks. This is the case that voice communication service is available under other networks. When mobile VoIP service users in the mobile internet network based on packet call up PSTN and mobile network users, but there have not been relevant quality indexes and quality standards for evaluating speech quality of mobile VoIP. In addition, there are many factors influencing on the speech quality in packet network. Especially, if the degraded speech with packet loss transfers to the other network users through the handover, voice communication quality is significantly deteriorated by the transformation of speech codecs. In this paper, we eventually adopt the Gilbert-Elliot channel model to characterize packet network and assess the voice quality through the objective speech quality method of ITU-T P. 862. 1 MOS-LQO for the various call scenario from mobile VoIP service user to PSTN and mobile network users under various packet loss rates in the transmission channel environments. Our simulation results show that transformation of speech codecs results in the degraded speech quality for different transmission channel environments when mobile VoIP service users call up PSTN and mobile network users.