Gui-Jung Kim

M2M-based smart health service for human UI/UX using motion recognition

Home networks currently dominated by human–object or human–human information production, exchange, processing, and paradigms are transitioning to machine to machine (M2M) due to the sudden introduction of embedded devices. Recently, due to the spread of IT equipment, more M2M-related devices are being used, and M2M-based projects are underway in various fields such as M2M-based u-city, u-port, u-work, u-traffic, etc. M2M has been applied in various fields, and u-healthcare is attracting attention in the M2M medical field. U-healthcare refers to technology in which ordinary patients can receive prescription services from experts by continuously monitoring changes in their health status during daily life at home based on wired and wireless communications infrastructures. In this paper, we propose an M2M-based smart health service for human UI/UX using motion recognition. Non-IP protocol, not TCP/IP protocol, has been used in sensor networks applied to M2M-based u-healthcare. However, sensors should be connected to the Internet in order to expand the use of services and facilitate management of the M2M-based sensor network. Therefore, we designed an M2M-based smart health service considering network mobility since data measured by the sensors should be transferred over the Internet. Unlike existing healthcare platforms, M2M-based smart health services have been developed for motion recognition as well as bio-information. Smart health services for motion recognition can sense four kinds of emotions, anger, sadness, neutrality, and joy, as well as stress using sensors. Further, they can measure the state of the individual by recognizing a user’s respiratory and heart rates using an ECG sensor. In the existing medical environment, most medical information systems managing patient data use a centralized server structure. Using a fixed network, it is easy to collect and process limited data, but there are limits to processing a large amount of data collected from M2M devices in real-time. Generally, M2M communication used in u-healthcare consists of many networked devices and gateways. An M2M network may use standardized wireless technology based on the requirements of a particular device. Network mobility occurs when the connecting point changes according to the movement of any network, and the terminal can be connected without changing its address. If the terminal within the network communicates with any corresponding node, communication between the terminal and corresponding node should be continuously serviced without discontinuation. The method proposed in this paper can easily respond to dynamic changes in the wireless environment and conduct systematic management based on user’s motion recognition using technology to support mobility among sensor nodes in M2M.

Policy on literature content based on software as service

Applications based on the software as a service (SaaS) platform that integrate the content of libraries across the country for personal and enterprise users are needed in the future. SaaS–based literature content integration has not yet been done in Korea. Therefore, this paper presents a SaaS application that integrates and manages the content of public organizations or libraries. This paper outlines a SaaS method and content framework for a literature content service that utilizes the metadata encoding and transmission specification and the metadata object description schema as the standard formats.

A method of intelligent recommendation using task ontology

This study proposed a method of developing an intelligent recommendation system for automotive parts assembly. The proposed system will display the detailed information and the list components which make up the relevant part that an user wants through the database using the ontology when selecting an automotive part that an user intends to learn or to be guided of. This study is to design task ontology based on Hierarchical Taxonomy so as to achieve productivity enhancement, cost reduction and outcome improvement through recommendations based on intelligence and personalization depending on the worker’s present situation or context of task in charge when assembly of automotive parts is conducted. For this, composing elements of an engine and upper/lower relationships were expressed using hierarchical structure Taxonomy. The intelligent recommendation system for parts is offered to users through determining the automatic recommendation order between parts using the weights. This study has experimented the principles of the recommendation system and the method of setting the weights by setting two scenarios.

Analysis of Correlation between Contacts to Cyber Obscene Materials and Sex Crime of High School Students

In this study, actual contact status to cyber obscene materials of high school students was grasped and the influence of contacts to cyber obscene materials on sex crime was analyzed. For this purpose, actual contacts status to cyber obscene materials, influence of cyber obscene materials, attitude of sex, and consciousness of sex crime were surveyed and accordingly correlation between each survey result were totally analyzed. This study result revealed that there is a close correlation between contacts to cyber obscene materials and sex crime. Obscene materials wrongly lead students who have big sexual curiosity to create distorted sexual behaviors and also commit sex crimes.

The clustering algorithm of design pattern using object-oriented relationship

Clustering is representative method of components classification. But, previous clustering method that uses cohesion and coupling cannot be effective, because design pattern has consisted by relation between classes. In this paper, we classified design patterns with special quality of object-oriented relationship. Classification by clustering had expressed higher precision than classification by facet. Therefore, can do that it is effective that classify design patterns using clustering algorithms that is automatic classification method. When we are searching design patterns, using classification of design patterns we can compare and analyze similar patterns because similar patterns is saved to same category. Also we can manage repository efficiently because of using and storing link information of patterns.

Intelligent Recommendation System for Automotive Parts Assembly

This study proposed a method of developing an intelligent recommendation system for automotive parts assembly. The proposed system will display the detailed information and the list components which make up the relevant part that an user wants through the database using the ontology when selecting an automotive part that an user intends to learn or to be guided of. The intelligent recommendation system for parts is offered to users through determining the automatic recommendation order between parts using the weights. This study has experimented the principles of the recommendation system and the method of setting the weights by setting two scenarios.

Integration Technology of Literature Contents Based on SaaS

Applications based on SaaS platform have been appeared as services for common users and enterprise users and an integration management of literature contents of libraries in the country is needed in the future. SaaS based literature contents integration has not been done here yet. Therefore, this paper presents SaaS based technology which integrates literature contents of public organizations or libraries and manages. It designed a SaaS technology applying method and a literature contents framework for literature contents service and utilized METS and MODS for the format standard.

Customizing Intelligent Recommendation System based on Compound Knowledge

This research does focus on realization of customizing recommendation service that all of formal, or informal learning is accomplished at real time according to worker`s current situation or business context corresponding with the individual ability and the learning progress at industry or education field. For this, we designed the customizing intelligent recommendation system based on compound knowledge that workers can listen to coaching advices at real time and to retrieve and recommend multidimensional relation easily. Also, we constructed the repository based on compound knowledge and process engine for efficient management of compound knowledge. In specific industry, expert solution or coaching service will be created using the knowledge which is accumulated in long-term.

360° Projection Image Analysis Method for the Calibration

Image degradation will occur depending on hardware characteristics according to the lapse of time between beam projectors when multivision system is installed in the Theme park/Exhibition/Science Museum. In this paper, we have researched the 10-bit High-depth and high-resolution projection image analysis technique to solve the problems of quality and the maintenance of the theater. The goal is to minimize the economic losses and the development of special theater calibration system that can efficiently support a quality of an image. We proposed the method of image analysis technology, and explained the detailed functions and evaluation methods for image analysis technique. Evaluation method included the performance items, and proposed reasonable value to the experimental method and the goal value.

Performance analysis of non-PC/SC based mini-WiMAX connection manager

In order to provide WiMAX network services to specific purposed devices such as credit card systems and surveillance systems, we have proposed a new non-PC/SC based mini WiMAX Connection Manager (non-PC/SC m-WCM) that can be applied to small hardware system or architecture with Micro Controller Unit (MCU). The existing legacy WiMAX Connection Managers (WCMs) generally run on the operating systems such as Windows and Linux. This makes it hard to design and implement lightweight WCMs for the specific purposed devices that have very limited hardware capabilities. In order to solve the problem, we propose a new non-PC/SC m-WCM architecture which has almost same functionalities of the legacy WCM and implement non-PC/SC m-WCM without the performance degradations. The proposed non-PC/SC m-WCM mainly consists of three units; WiMAX module Control Unit (WCU), Data Surveillance Unit (DSU), and External device Program Unit (EPU). Besides these, it also has TCP/IP protocol stack and RTOS kernel, and can transmit date between external end user’s device and remote server system over the WiMAX networks. For the delay and throughput performance comparisons with the legacy WCMs, we have implemented the non-PC/SC m-WCM and tested it under a commercial mobile WiMAX network environment. From the experimental results, the non-PC/SC m-WCM is verified to show almost equal stability and delay performances with the legacy WCM and xDSL, but especially better than WCDMA. In the throughput tests, it has been shown that each of the modems shows almost similar throughput performance. Although the proposed WCM is designed with the very limited hardware specifications, it shows no performance degradations compared to the legacy WCMs. As a result, it has a strong possibility to be able to apply the non-PC/SC m-WCM to various types of specific purposed embedded systems, and to provide them with the same levels of WiMAX network services as those provided by the legacy WCMs.