Yanggon Kim

Pupil and Iris Localization for Iris Recognition in Mobile Phones

Until now, iris recognition has been used in many fields. Recently, there have been attempts to adopt iris recognition technology for the security of mobile phones. For example, in case of bank transaction service by using a mobile phone, using a mobile phone can use high level of security based on iris recognition. In this paper, we propose a new pupil & iris segmentation method apt for the mobile environment. We find the pupil & iris at the same time, using both information of the pupil and iris. And we also use characteristic of the eye image. Experimental result shows that our algorithm has good performance in various images, which include motion or optical blurring, ghost, specular reflection and etc. from various environments for iris recognition system

Automated Twitter data collecting tool and case study with rule-based analysis

Applying data mining techniques to social media can yield interesting perspectives about individual human behavior, detecting hot issues and topics, or discovering a group and community. However, it is difficult to build your own data set to apply data mining techniques without an automated data gathering and filtering system because of main characteristics of social media: the data is large, noisy and dynamic. To overcome these challenges, we developed a java-based data gathering tool that continually collects social data from Twitter and filters noisy data. This allows us, as well as other researchers, to build our own Twitter database. In this paper, we introduce the design specifications and explain the implementation details of the Twitter Data Collecting Tool we developed. In addition, we provide an analysis of Twitter messages about various Super Bowl ads by applying data-mining techniques to a case study.

Automated Twitter data collecting tool for data mining in social network

Applying data mining techniques to social media can yield interesting perspectives about individual human behavior, detecting hot issues and topics, or discovering a group and community. However, it is difficult to build your own data set to apply data mining techniques without an automated data gathering system. To overcome this challenge, we developed a java-based data gathering tool that continually collects social data from Twitter. This allows us, as well as other researchers, to build our own Twitter database. In this paper, we introduce the design specifications and explain the implementation details of the Twitter Data Collecting Tool we developed. In addition, we provide an in-depth analysis of Twitter messages about various Super Bowl ads by applying data-mining techniques to a case study. The study aims to address the question of how people use Twitter and to assess the power of Twitter in terms of creating consumer interest in brands and commercials.

A new aggregation policy for RSS services

RSS is the XML-based format for syndication of Web contents, and users aggregate RSS feeds with RSS feed aggregators. There are RSS aggregation policies that help aggregate RSS feeds effectively. In this paper, we first propose an aggregation policy to minimize the number of missing postings within an aggregation. Second, we analyze and compare our aggregation policy with existing aggregation policies. Our analysis reveals that our aggregation policy can reduce approximately 23% of the missing posts in comparison with an existing policy while it increases only 6% of the aggregation delay.

The Prototype of the Massive Events Streams Service Architecture and its Application

This paper describes the prototype of the Massive Events Streams Service Architecture (MESSA), which is used as a framework of various RFID-enabled application systems. The MESSA captures events from a number of RFID readers, refines the events, processes the continuous queries, and delivers the event query results to the applications systems. In addition, the MESSA conforms to the ALE 1.0 interfaces from the EPCglobal in order to make the MESSA widely applicable to general events handling applications. The underlying data manipulation schemes of the MESSA support the continuous queries to manage the events streams from external data sources during specified periods of time. The main contribution of this research lies in its seamless integration of the event handling processes within a unified framework, MESSA, and realization of air cargo handling business logics into the RFID-enabled air cargo management system based on the MESSA.

The XCREAM Framework and Collaboration Validity Tests

The purpose of the XCREAM (XLogic Collaborative RFID/USN-Enabled Adaptive Middleware) is to enable collaboration among many RFID/USN-enabled appplications by providing them with flexible interface to the XCREAM through a web-based service scheme, called the Enterprise Manager, and XML infrastructure language, the XLogic script language. Rapidly growing demand for collaboration among numerous and heterogeneous business applications draws attention to this kind of framework, especially in the ubiquitous computing environment. The XCREAM framework gathers massive events from a variety of event sources and distributes them to the appropriate service parties depending on the predefined business scenarios. The scenarios are written in the XLogic script language and registered to the XCREAM framework. The paper includes simulation result which shows excellence in performance and validity of collaboration suppport of the XCREAM. Reviewing the test results, XCREAM works well, especially in the collaborative environment. This approach makes it possible to integrate many heterogeneous services with various data sources and to present a collaborative service framework.

Design and Implementation of Scenario-Based Collaborative Framework: XCREAM

The XCREAM (XLogic Collaborative RFID/USN-Enabled Adaptive Middleware), as a mediator between smart physical objects and collaborative cyber services, gathers massive events from a variety of event origins and distributes them to the appropriate service parties depending on the predefined business scenarios. Demand for collaboration among numerous business applications draws attention to this kind of middleware platform especially in the highly networked environment. The XCREAM platform, the scenario-based collaborative framework, integrates various kinds of RFID/USN enabled application services and provides more advanced services than the individual services. The goal of this research is to construct well-organized systematic RFID/USN prototyping framework, to which physical RFID readers or sensor devices are directly connected. Additionally, tag data and sensor signals from the devices are propagated to the proper services registered to the framework. The study encourages us to develop the XCREAM platform and build prototyping environment for presenting further comprehensive services. This approach makes it possible to integrate many heterogeneous services and to present collaborative service framework.

A Method for Extracting Lexicon for Sentiment Analysis Based on Morphological Sentence Patterns

In these days, people share their emotions, opinions, and experiences of products or services using online review services on their comments, and the people concern the reviews to make decision when buying products or services. Sentiment analysis is one of the solution to observe and summarize emotional opinions from the data. In spite of high demands for developing sentiment analysis, the development of the sentiment analysis faces some challenges to analyze the data, because the data is unstructured, unlabeled, and noisy. The aspect-based sentiment analysis approach helps for more in-depth analysis, however building aspect and emotional expression is one of the challenge for the aspect-based sentiment analysis approach. Accordingly, we propose an unsupervised system for building aspect-expressions to minimize human-coding efforts. The proposed method uses morphological sentence patterns through an aspect-expression pattern recognizer. It guarantees relatively higher accuracy. As well as, we found some characteristics for selecting patterns to extracting aspect-expressions accurately. The greatest advantage of our system is performing without any human coded train-set.

Semantic Reasoning with Contextual Ontologies on Sensor Cloud Environment

This research first focused on processing enormous number of sensor events from a variety of city-wide sensor networks. As a solution, the well-known big data handling scheme, Hadoop cluster framework, drew, unquestionably, our attention. The acquired sensor events are to be used to immediately detect a certain abnormal situation within the framework. Accordingly, we integrated our existing context-aware collaboration framework with Hadoop cluster framework by interfacing data collection and context-aware reasoning parts of the existing framework with the Hadoop cluster framework. This approach enabled us to effectively process massive sensor events and semantically analyze the big data within the cluster environment. The proposed smart city sensor cloud framework provides ontology-enabled semantic reasoning scheme with the XOntology in combination with the Context-Aware Inference (CAI) model. By applying the ontology technology, the proposed framework enhances the availability and interoperability of the contextual information across many cooperating parties according to semantic reasoning results. Further, this framework is flexible enough to integrate any heterogeneous platforms including many existing IT solutions as well as mobile platforms. In addition, this approach presents the direction of progressive migration of many existing sensor network solutions into big data handling sensor cloud framework.

Context-Aware Inference (CAI) Model on Smart Computing Environment

With the proliferation of smart computing equipment, the range of providing highly intelligent services is widely expanding in an attempt to include the present condition of each individual data source. Furthermore, broadly adopted RFID (Radio Frequency Identification) technology has led to the widespread application of automatic identification in numerous pervasive computing services including logistics, emergency control, and medical services. Another remarkable phenomenon is the explosively growing number of smartphone users, which allows them to stay connected to the Internet. The environment created by these factors, increases the need to make a decision out of the current context which can be inferred from various sensors and mobile devices. In response to the growing demand for the need to make a decision in the current context, the new XOnt Agent has been integrated into our collaboration middleware framework, the XCREAM (XLogic Collaborative RFID/USN-Enabled Adaptive Middleware). All the collected information should be examined too see if they meet any condition that triggers correspondent actions of a specific rule through reasoning process of a rule engine. We also suggest the context-aware inference (CAI) model to effectively support the reasoning and handling processes.