• Journal of Distribution Science
• /
• v.4 no.1
• /
• pp.103-120
• /
• 2006

This study is composed of analyses about whether intensifying the attribute level of alternatives presented to consumers makes the effect of intensifying attributes and there is any significant difference among effects of the intensified attributes. For this, the most favorite alternative is extracted from each group of computers and jeans through conjoint analysis, which is used as a standard alternative during the process of experimentation. In conclusion, this research suggests that in terms of business, by intensifying the latter attribute of products rather than the early attribute, an enterprise acquire distinct and competitive superiority over similar competing products (esp. functional products), and theoretically, by expanding and analyzing decision models it can predict consumers' behaviors more correctly.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2014.05a
• /
• pp.76-78
• /
• 2014

Naive Bayes (NB, for shortly) learning is more popular, faster and effective supervised learning method to handle the labeled datasets especially in which have some noises, NB learning also has well performance. However, the conditional independent assumption of NB learning imposes some restriction on the property of handling data of real world. Some researchers proposed lots of methods to relax NB assumption, those methods also include attribute weighting, kernel density estimating. In this paper, we propose a novel approach called NB Based on Attribute Weighting in Kernel Density Estimation (NBAWKDE) to improve the NB learning classification ability via combining kernel density estimation and attribute weighting.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.7
• /
• pp.3339-3352
• /
• 2016

In ciphertext-policy attribute-based encryption (CP-ABE) scheme, a user's secret key is associated with a set of attributes, and the ciphertext is associated with an access policy. The user can decrypt the ciphertext if and only if the attribute set of his secret key satisfies the access policy specified in the ciphertext. In the present schemes, access policy is sent to the decryptor along with the ciphertext, which means that the privacy of the encryptor is revealed. In order to solve such problem, we propose a CP-ABE scheme with hidden access policy, which is able to preserve the privacy of the encryptor and decryptor. And what's more in the present schemes, the users need to do excessive calculation for decryption to check whether their attributes match the access policy specified in the ciphertext or not, which makes the users do useless computation if the attributes don't match the hidden access policy. In order to solve efficiency issue, our scheme adds a testing phase to avoid the unnecessary operation above before decryption. The computation cost for the testing phase is much less than the decryption computation so that the efficiency in our scheme is improved. Meanwhile, our new scheme is proved to be selectively secure against chosen-plaintext attack under DDH assumption.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.10 no.6
• /
• pp.683-690
• /
• 2015

In Rough set, attribute missing values have several problems such as reduct and core estimation. Further, they do not give some discernable pattern for decision tree construction. Now, there are several methods such as substitutions of typical attribute values, assignment of every possible value, event covering, C4.5 and special LEMS algorithm. However, they are mainly substitutions into frequently appearing values or common attribute ones. Thus, decision rules with high information loss are derived in case that important attribute values are missing in pattern matching. In particular, there is difficult to implement cross validation of the decision rules. In this paper we suggest new method for substituting the missing attribute values into high information gain by using entropy variation among given attributes, and thereby completing the information table. The suggested method is validated by conducting the same rough set analysis on the incomplete information system using the software ROSE.

• Journal of the Korea Convergence Society
• /
• v.7 no.5
• /
• pp.181-188
• /
• 2016

This study investigated consumer convergence preferences for healthcare information service in social media using the multi-attribute utility theory. The multi-attribute utility theory is an approach for compositional modeling attributes wherein researchers compute the overall service utility by aggregating the evaluation results for attribute values. We found that healthcare information service consumers in social media evaluate the service preference through the multi attributes - Update, Cost, Usefulness, Usability, Reliability. And the study shows that individual preference by estimating weight of each attribute in the services influences the service use intention. So we believe that estimating preferences with multi-attribute utility theory will predict business success of the services in a new media, and for a successful business of the social media services marketers should assess the consumer's preference using the multi-attributes.

• Journal of KIISE:Databases
• /
• v.37 no.3
• /
• pp.137-145
• /
• 2010

Wireless sensors play important roles in various areas as ubiquitous computing is generalized. Depending on applications properties, each sensor can be equipped with limited computing power in addition to general function of gathering environment-related information. One of main issues in this environment is to improve energy-efficiency in sensor nodes. In this paper, we devise a new attribute-query processing algorithm. Each sensor has to maintain partial information locally about attributes values gathered at its all descendent nodes. As the volume is higher, however, the maintenance cost also increases. And the update cost also has to be considered in the proposed algorithm. Thus, some bits, AVB(Attribute-Value Bits), are delivered instead of the value itself, where each bit represents a bound of attribute. Thus, the partial information can decrease the number of exchanged messages with a little cost during query processing. Through simulation works, the proposed algorithm is analyzed from several points of view.

• Fashion & Textile Research Journal
• /
• v.8 no.6
• /
• pp.647-654
• /
• 2006

The purpose of this study is to identify the differences among goods attribute and brand awareness on fashion brand type. we were intended to suggest characteristics of each consumer group by identifying the differences of consumers' purchasing activities. 672 of consumers by brand who frequently purchase casual brand were chosen for the analysis according to common brand classification of national brand, private brand and no brand. For the purpose of data analysis, we performed factorial analysis of measuring tools and credibility test. Concerning the differences of goods attribute, brand awareness by brand type, MANOVA, ANOVA was employed, complimented with Sheffe-test as a post hoc test in case of occurrence of any differences by group. The findings from the analysis are described in the following. Regarding goods attribute by fashion brand type, there existed a significant difference between brand types in all the sub factors of goods attribute such as product attribute, shop attribute, and price attribute. Especially, the difference of product attribute is much more significant in the areas of material suitableness, product assortment, aesthetic expression, size & quality, clothing maintenance, and clothing comfortableness. In case of shop attribute, there was a significant difference between groups in all the factors such as shop environment, convenience of shopping, sales promotion, service quality of sales clerk, location, and shop reputation. Concerning price attribute, we found a significant difference between groups in the factors of price value, price reasonableness, price information, and economical efficiency of price. As for the difference of brand awareness by brand type, among other factors, brand value had a difference between groups; that is, private brand was found to obtain the highest brand value awareness.

• The KIPS Transactions:PartA
• /
• v.8A no.3
• /
• pp.209-216
• /
• 2001

The effective way to apply incremental attribute grammar to a complex language process is the use of optimized algorithm. In optimized algorithm for incremental attribute grammar, the new input attribute tree should be exactly compared with the previous input attribute tree, in order to determine which subtrees from the old should be used in constructing the new one. In this paper the new optimized algorithm was reconstructed by analyzing the algorithm suggested by Carle and Pollock, and a generation process of new attribute tree d’copy was added. Through the performance evaluation for the suggested matching algorithm, the run time is approximately improved by 19.5%, compared to the result of existing algorithm.

• Journal of the Korea Society of Computer and Information
• /
• v.12 no.4
• /
• pp.95-102
• /
• 2007

Recently, a lot of researchers focus on the sensor middleware that hide the complexity of application developments and provide the abstraction of functions to upper application layer. Although there we several factors to design sensor middleware, the attribute-based naming is considered to be an essential factor among them. However, most existing researches were not designed to reflect the characteristics of sensor networks and have the limitation of attribute-based query extension. This study adopts the concept of Virtual Counterpart to suggest the structure there attribute-based naming is supported by virtual sensor nodes of the middleware on the sink node. Unlike traditional data-centric middleware in which individual sensor nodes process attribute-based query, virtual sensor nodes mapped to physical sensor nodes are running on the middleware of the sink node and process attribute-based query as a proxy of the physical sensor. This approach enables attribute-based naming independent of physical infrastructure and easy extensibility.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.51 no.1
• /
• pp.57-70
• /
• 2014

Inner product encryption (IPE) scheme is a cryptographic primitive that provides fine grained relations between secret keys and ciphertexts. This paper proposes a new IPE scheme which achieves fully attribute hiding security. Our IPE scheme is based on bilinear groups of a composite order. We prove the fully attribute hiding security of our IPE by using dual encryption system framework. In performance analysis, we compare the computation cost and memory requirement of our proposed IPE to other existing IPE schemes.