- [Show abstract] [Hide abstract] ABSTRACT: As the popularity of Web 2.0 increases, web users used to establish communication relationships and share resources with others online. Recently, one important research for retrieving information from the web has become a major issue. Moreover, due to the social network popularity, there is much more opportunity to find out various human resources to help people get what they want. In our work, we propose to help the social network users to find out the most reachable experts who are authorized researchers on a common topic area with the users. For this purpose, we propose how to calculate centrality closeness from requesting user to experts, analyze the relevancy of an expert to the topic, and finally combine both of the result to rank the most reachable experts for the requesting user.
Conference Paper: Mining Information of Anonymous User on a Social Network Service[Show abstract] [Hide abstract] ABSTRACT: 2 vntlffl, 3 uk3080789, Abstract— The growing number of individuals is recently writing their own opinions or information freely at the network space on the web such as the blog or Online Cafe and these network spaces are developed toward a new service called social network. Consequently, a lot of researchers are studying this social network lively. The social network is not only a tool for providing real time news, but also having an effect on an opinion or a policy decision. It uses mainly texts to present information in real-time, and its users use a computer or a mobile device to upload their own opinions or ideas. A lot of texts from users contain important and various information. We will get this information we need real-time if semantic can be extracted from texts on a social network. In this case, Opinion mining should be useful to extract semantic from social network. This paper suggests a noble method to grasp information of anonymous users through relationship information available and their psychology that is reflected on texts and also understand the meanings of contents in depth. Keywords-Anonymous; Opinion mining; LIWC; Social network service; Relation number
Conference Paper: Using wordmap and score-based weight in opinion mining with mapreduce[Show abstract] [Hide abstract] ABSTRACT: Cloud computing is newly rising as a novel drift of data management and many researchers find that opinion mining can be faster using cloud computing. Using the current opinion mining is, however, unfit for the Internet because the Internet has huge information and is changing at short intervals. In addition, utilizing marks or scores such as the number of stars awarded and sentiment classification will be more commonly used for analyzing opinions. For these reasons, we propose a new approach to opinion mining. We use MapReduce function as an opinion analyzing and clustering tool with score-based weight and try to make opinion mining simpler because of fixing in MapReduce. Our new approach can analyze results of documents with the opinion mining faster than using current methods and make products that meet requirements of users who want to employ outcomes of opinion mining. Our study is a new idea for opinion mining and done in a distinctive way and we are looking forward to applying this noble method to all related fields including searching engines.
Conference Paper: Credibility Evaluation and Results with Leader-Weight in Opinion Mining[Show abstract] [Hide abstract] ABSTRACT: A plethora of information is recently flooding the Internet with a rapid surge in Internet usage. Information easily available on the Internet help researchers understand people they study. The existing techniques of the opinion mining usually consist of sentiment classification, feature-based opinion mining, summarization, comparative sentence and relation mining, opinion search, opinion spam, and the linguistic dictionary construction such as the WordNet. This paper, however, proposes differing methods of opinion mining from existing ones. The methods we present here enable credibility evaluation and result conversion using influence of each opinion holder on the Internet and their personal information, which are an analysis-result of LIWC, including their background information and tendency.
Conference Paper: RBAC-based access control for privacy preserving in semantic web[Show abstract] [Hide abstract] ABSTRACT: Nowadays, remarkable development of information technologies makes information sharing and distributing possible in smooth water than ever. It enables to obtain information anytime and anywhere and provides technologies for superior services based on situation information. Also, it enables machines to grasp the meaning of information under semantic web technologies and new information paradigm. These technologies have been varied by information approach and acquisition method than ever before. However, illegal attacker behind development that uses better tools gives rise to side effects. Current systems can be intruded by a number of different ways of them. Therefore, system safety and privacy protection is being threatened by the invaders. There is access control technology for database system among the security technologies against this menace. Recently, security technique is carrying out researches of situation intelligence, privacy, and XML access etc. in order to correspond with new computing environments. In this paper, we propose extended Role Based Access Control (RBAC) in semantic web. Extended model can dynamically authorizes user's permission under context component. Also, we propose concept-enforcement model based on semantic web. OWL defines the terms used to describe and represent an area of knowledge. Proposed model enable to access semantic execution by suggested model even though it doesn't correspond with security policy.
Conference Paper: A method for opinion mining of product reviews using association rules[Show abstract] [Hide abstract] ABSTRACT: When most people buy the products, they inquire about other people's opinion and refer to recommended product. Today, the result of explosive development of the Web makes it easy to consult other people's opinion information. These variety of opinion data are not only useful to customers, but also manufacturers. As a result, opinion mining research to analyze opinion data on the web has become a popular topic recently. In this paper, we proposed opinion mining method for product reviews. In our approach, we first do POS tagging on each review sentence, and we extract feature and opinion words in form of transaction data. Then we discover association rules of needed type from the transaction data, and provide information that is summarized advantages and disadvantages using PMI-IR algorithm.
- [Show abstract] [Hide abstract] ABSTRACT: Advance technologies in decentralized systems are the new building block of today's Internet and provide interoperability among heterogeneous databases. In these environments, interoperation and information sharing are one of the most critical issues. Interoperability enables users to access database in different domain. Furthermore, government, financial and medical institutions more require secure collaboration to share their data with organizations. However secure collaboration has many challenging problem in multi-domain environments. In this paper, we propose secure collaboration to effectively share resource by reconcilement structure. Proposed scheme is based on Role Based Access Control (RBAC) to support flexible control. In decentralized RBAC-based system, the number of roles can be in the hundreds or thousands and they share information among dispersed domains. We present solution which regulates the interoperability.
- [Show abstract] [Hide abstract] ABSTRACT: Recently sequential pattern has become an important research with broad applications. The task discovering frequent subsequences in sequence database is very worth. However, a frequent long sequence pattern, contains a combinatorial number of frequent subsequences, mining will generate an exponential number of frequent subsequences for long patterns, which is excessively expensive in both time and space. A more practical and scalable alternative is required which discovery of subsequential pattern. If a pattern has repetitive subsequences in a sequence, each subsequence must distinguish due to different weight in the pattern. For this reason, we need to gap weight for mining of repetitive subsequence. As yet, no subsequential pattern mining though gap weights are very important in the real world. We can mine the weighted mining of repetitive subsequences with gap weights. In the paper, we propose an algorithm, EWM (Efficient Gap-Weighted Mining), for the problem of mining repetitive subsequences. The EWM can address situations where distinguish between same subsequences. Furthermore, we introduce the concept of gap weight for subsequences which have different gap between events. To this end, we define and use a new type of database to represent sequence data efficiently. The EWM for the discovery of all subsequence patterns may lose information but is both efficient and scalable when pruning infrequent subsequences and discovering ordered subsequential patterns.
- [Show abstract] [Hide abstract] ABSTRACT: As growing interest in data publishing and analysis, privacy preserving data publication has become more important today. When a table containing the sensitive information is published, privacy of each individual should be protected. On the other hand, a data holder also considers minimizing information loss for analysis as long as the privacy is preserved. A few years ago, k-anonymity and l-diversity models have been suggested in order to protect privacy. However, these solutions are limited to static data release. Recently, the m-invariance model has been proposed to apply publication of dynamic environments. However, m-invariance generalization technique causes high information loss. In this paper, we propose a simple and safe anonymization technique without generalization while assuring high data utility in dynamic environments.
- [Show abstract] [Hide abstract] ABSTRACT: Group key exchange protocols allow a group of parties communicating over a public network to come up with a common secret key called a session key. Due to their critical role in building secure multicast channels, a number of group key exchange protocols have been suggested over the years for a variety of settings. Among these is the so-called EKE-M protocol proposed by Byun and Lee for password-based group key exchange in the different password authentication model, where group members are assumed to hold an individual password rather than a common password. While the announcement of the EKE-M protocol was essential in the light of the practical significance of the different password authentication model, Tang and Chen showed that the EKE-M protocol itself suffers from an undetectable on-line dictionary attack. Given Tang and Chen’s attack, Byun et al. have recently suggested a modification to the EKE-M protocol and claimed that their modification makes EKE-M resistant to the attack. However, the claim turned out to be untrue. In the current paper, we demonstrate this by showing that Byun et al.’s modified EKE-M is still vulnerable to an undetectable on-line dictionary attack. Besides reporting our attack, we also figure out what has gone wrong with Byun et al.’s modification and how to fix it.
Conference Paper: A Secure Delegation Model Based on Multi-agent in Pervasive Environments[Show abstract] [Hide abstract] ABSTRACT: In pervasive environments, the connected devices can be aware the status of users and provide the information to users automatically in anytime, anywhere. However these require more security technologies to protect private information. The users must temporarily delegate some or all of their rights to agents in order to perform actions on their behalf. This results in the exposure of user privacy information by agents. We propose a delegation model for providing safety and user privacy in pervasive computing environments. In order to provide safety and user privacy, XML-based encryption and a digital signature mechanism needs to be efficiently integrated. In this paper, we propose access control mechanism based on eXtensible Access Control Markup Language (XACML) in order to manage the services and policies. We also extend SAML, in order to declare delegation assertions transferred to service providers by delegation among agents.
- [Show abstract] [Hide abstract] ABSTRACT: With the continuous growth in XML data sources over the Internet, the discovery of useful information from a collection of XML documents is currently one of the main re- search areas occupying the data mining community. The most commonly adopted approach to this task is to extract frequently occurring subtree patterns from XML trees. But, the number of frequent subtrees usually grows exponentially with the size of trees, and therefore, mining all frequent subtrees becomes infeasible for large size trees. A more practical and scalable alternative is to use maximal frequent subtrees, the number of which is much smaller than that of frequent subtrees. Handling the maximal frequent subtrees is an inter- esting challenge, though, and represents the core of this paper. We present a novel, concep- tually simple, yet effective algorithm, called EXiT-B, that significantly simplifies the proc- ess of mining maximal frequent subtrees. This is achieved by two distinct features. First, EXiT-B represents all of string node labels of trees by some specified length of bits. Through fast bitwise operations, the process of deciding on which paths of trees contain a given node is accelerated. Second, EXiT-B avoids time-consuming tree join operations by using a spe- cially devised data structure called PairSet. To the best of our knowledge, EXiT-B is the first algorithm that discovers maximal frequent subtrees adopting bits representation. We also demonstrate the performance of our algorithm through extensive experiments using syn- thetic datasets which were generated artificially by a randomized tree-structure generator.
- [Show abstract] [Hide abstract] ABSTRACT: Key exchange protocols allow two or more parties communicating over a public network to establish a common secret key called a session key. Due to their significance in building a secure communication channel, a number of key exchange protocols have been suggested over the years for a variety of settings. Among these is the so-called S-3PAKE protocol proposed by Lu and Cao for password-authenticated key exchange in the three-party setting. In the current work, we are concerned with the password security of the S-3PAKE protocol. We first show that S-3PAKE is vulnerable to an off-line dictionary attack in which an attacker exhaustively enumerates all possible passwords in an off-line manner to determine the correct one. We then figure out how to eliminate the security vulnerability of S-3PAKE.
- [Show abstract] [Hide abstract] ABSTRACT: Pervasive environment is a post-desktop model of human-computer interaction in which information processing has been thoroughly integrated into everyday object and activities. In there environment access control is a critical issue, with many aspects relating to the establishment, authorization and enforcement of policies that protect the resources from adversaries. Recently, many researches have been worked methods that have access to resources in ubiquitous applications. However they are inadequate to meet the requirement for privacy safeguard and dynamic changes. In this paper we propose how to protect sensitive data and present extended Role Based Access Control (RBAC) that respond to all the requirements for privacy control. We also show how proposed model preserves safety properties in spite of dynamic changes to access control permission.
Conference Paper: Mining opinions from messenger[Show abstract] [Hide abstract] ABSTRACT: Increasing Internet users has created enormous important information of users in Internet. Opinion mining is technology that extracts meaningful opinions from that huge information. Becoming a hot research area, opinion mining has been studied in many different ways. These studies are mostly based on reviews, blogs. However, this paper focuses on messenger which generates many messages containing opinions of users. As messages may contain many opinions unrelated to our purpose, our aim is to extract only related opinions and features. Our approach initially collects messages from messengers and employs localized linguistic technique to extract candidate messages, opinions and features. Thereafter, we extract features from candidate features using association rule mining. Finally we summarize extracted opinions and features.
Conference Paper: Mining association rules in tree structured XML data[Show abstract] [Hide abstract] ABSTRACT: XML is increasingly popular for knowledge representations. However, mining association rules from them is a challenging issue since XML data is usually poorly supported by the current database systems due to its tree structure. Several encouraging attempts at developing methods for mining rules in tree dataset have been proposed, but simplicity and efficiency still remain significant impediments for further development. What is needed is a clear and simple methodology for finding the rules that are hidden in the heterogeneous tree data. In this paper, we adjust and fine-tune the label projection method which has been recently published to compute association rules from trees. The suggested approach avoids the computationally intractable problem caused by the number of nodes contained in the tree dataset.
Conference Paper: A Data Sanitization Method for Privacy Preserving Data Re-publication[Show abstract] [Hide abstract] ABSTRACT: When a table containing personal information is published, sensitive information should not be revealed. Although k-anonymity and l-diversity models are popular approaches to protect privacy, they are limited to one time data publishing. After a dataset is updated with insertions and deletions, a data holder cannot safely release up-to-date information. Recently, m-invariance model has been proposed to support re-publication of dynamic datasets. However, m-invariance model has two drawbacks. First, the m-invariant generalization can cause high information loss. Second, if the adversary already obtained sensitive values of some individuals before accessing released information, m-invariance leads to severe privacy breaches. In this paper, we propose a new data sanitization technique for safely releasing dynamic datasets. The proposed technique prevents two drawbacks of m-invariance and provides a simple and effective method for handling inserted and deleted records.
Conference Paper: Distributed File Discovery Protocol in Mobile Peer-to-Peer Networks[Show abstract] [Hide abstract] ABSTRACT: In mobile peer-to-peer networks, it is important to discovery file to increase the file accessibility and to decrease the wireless network traffic. In this paper, we design and evaluate a discovery scheme to effectively search files by using only local information in a mobile peer-to-peer network. Our discovery protocol is based on the concept of peer-to-peer caching of file information advertisement and lower ID-based forwarding of file requests. Our protocol is that physical hop counts and the number of messages exchanged have been significantly reduced, since it does not require a central lookup server and does not rely on flooding. In the proposed scheme, physical hop counts and the number of messages exchanged have been significantly reduced, compared with the other protocol. We will show via a simulation how this system works efficiently.
- [Show abstract] [Hide abstract] ABSTRACT: Due to its highly flexible tree structure, XML data is used to capture most kinds of data and provides a substrate in which almost any other data structure may be presented. With the continuous growth of XML tree data in electronic environments, the discovery of useful knowledge from them has been a main research area in the information retrieval community. The mostly used approach to this task is to extract frequently occurring subtree patterns from a set of trees. However, because the number of frequent subtrees grows exponentially with the size of trees, a more practical and scalable alternative is required, which is the discovery of maximal frequent subtrees. The maximal frequent subtrees hold all the useful information, though, the number of them is much smaller than that of frequent subtrees. Handling the maximal frequent subtrees is an interesting challenge, and represents the core of this paper. As far as we know, this is one of the first studies to directly discover maximal frequent subtrees without any candidate sets generations as well as eliminating the process of useless subtree pruning. To this end, we define and use a new type of projected database to represent XML tree data efficiently. It significantly improves the entire process of mining maximal frequent subtree patterns. We study the performance and the scalability of the proposed approach through experiments based on synthetic datasets.
- [Show abstract] [Hide abstract] ABSTRACT: In this paper, we proposed a multi-agent based healthcare system (MAHS) which is the combination of a medical sensor module and wireless communication technology. This MAHS provides broad services for mobile telemedicine, patient monitoring, emergency management, doctor’s diagnosis and prescription, patients and doctors and information exchange between hospital workers over a wide area. Futher more, MAHS is connected to a Body Area Network (BAN) and a doctor and hospital support staff. In this paper, we demonstrate how we can collect diagnosis patterns, classify them into normal, and emergency and be ready for an emergency by using the real-time biosignal data obtained from a patient’s body. This proposed method deals with the enormous quantity of real-time sensing data and performs analysis and comparison between the data of patient’s history and the real-time sensory data. In this paper, we separate Association rule exploration into two data groups: one is the existing enormous quantity of medical history data. The other group is real-time sensory data which is collected from sensors measuring body temperature, blood pressure, pulse. We suggest methods to analyze and model patterns of a patient’s state for normal, and very emergency, and making decisions about a patient’s present status by utilizing these two data groups.
Sŏul, Seoul, South Korea
- Department of Computer Engineering