• Journal of the Korean Institute of Rural Architecture
• /
• v.7 no.2
• /
• pp.1-8
• /
• 2005

In a rural house, it is necessary to make a quality model that choose effective design value engineering. So more effective models, this study examine requests of a rural house project designers, constructors, and tenants. Checked items are classified into four groups that working area, dwelling area, constructive and environmental parts. Each groups are also divided into detailed items for basic decisional elements. When basis points sets 10, it can be divided that working area parts 3pts, dwelling area parts 2.5pts, constructive parts 3pts and environmental parts 1.5pts. In this results, we can make a proposal of evaluation on additive value quality model for a rural house in design value engineering.

• Journal of the Korean Institute of Rural Architecture
• /
• v.8 no.1
• /
• pp.9-16
• /
• 2006

In campus project, "Barrier Free Design" is the essential part for the handicapped. It is necessary to make a quality model that choose effective design value engineering objects. So more effective models, this study examine requests of the university students. Checked items are classified into four groups that movement, guidable, safety and territoria parts, and each groups are also divided into detailed items for basic decisional elements. When basis points sets 10, it can be divided that movement parts 2.93pts, guidable parts 2.31pts, safety parts 2.41pts and territoria parts 2.35pts. in this results, we can make additive value and quality model for barrier free design in campus design value engineering.

• The KIPS Transactions:PartB
• /
• v.15B no.5
• /
• pp.457-464
• /
• 2008

In the real-world operational environment, most of text classification systems have the problems of insufficient training documents and no prior knowledge of feature space. In this regard, $Na{\ddot{i}ve$ Bayes is known to be an appropriate algorithm of operational text classification since the classification model can be evolved easily by incrementally updating its pre-learned classification model and feature space. This paper proposes the improving technique of $Na{\ddot{i}ve$ Bayes classifier through feature weighting strategy. The basic idea is that parameter estimation of $Na{\ddot{i}ve$ Bayes considers the degree of feature importance as well as feature distribution. We can develop a more accurate classification model by incorporating feature weights into Naive Bayes learning algorithm, not performing a learning process with a reduced feature set. In addition, we have extended a conventional feature update algorithm for incremental feature weighting in a dynamic operational environment. To evaluate the proposed method, we perform the experiments using the various document collections, and show that the traditional $Na{\ddot{i}ve$ Bayes classifier can be significantly improved by the proposed technique.

• The Journal of Bigdata
• /
• v.5 no.1
• /
• pp.197-208
• /
• 2020

Lexical ambiguity means that a word can be interpreted as two or more meanings, such as homonym and polysemy, and there are many cases of word sense ambiguation in words expressing emotions. In terms of projecting human psychology, these words convey specific and rich contexts, resulting in lexical ambiguity. In this study, we propose an emotional classification model that disambiguate word sense using bidirectional LSTM. It is based on the assumption that if the information of the surrounding context is fully reflected, the problem of lexical ambiguity can be solved and the emotions that the sentence wants to express can be expressed as one. Bidirectional LSTM is an algorithm that is frequently used in the field of natural language processing research requiring contextual information and is also intended to be used in this study to learn context. GloVe embedding is used as the embedding layer of this research model, and the performance of this model was verified compared to the model applied with LSTM and RNN algorithms. Such a framework could contribute to various fields, including marketing, which could connect the emotions of SNS users to their desire for consumption.

• Proceedings of the Materials Research Society of Korea Conference
• /
• 2007.11a
• /
• pp.153.2-153.2
• /
• 2007

• Journal of KIISE:Computing Practices and Letters
• /
• v.15 no.5
• /
• pp.345-349
• /
• 2009

Social bookmarking systems are a typical web 2.0 service based on folksonomy, providing the platform for storing and sharing bookmarking information. Spammers in social bookmarking systems denote the users who abuse the system for their own interests in an improper way. They can make the entire resources in social bookmarking systems useless by posting lots of wrong information. Hence, it is important to detect spammers as early as possible and protect social bookmarking systems from their attack. In this paper, we applied a diverse set of machine learning approaches, i.e., decision tables, decision trees (ID3), $na{\ddot{i}}ve$ Bayes classifiers, TAN (tree-augment $na{\ddot{i}}ve$ Bayes) classifiers, and artificial neural networks to this task. In our experiments, $na{\ddot{i}}ve$ Bayes classifiers performed significantly better than other methods with respect to the AUC (area under the ROC curve) score as veil as the model building time. Plausible explanations for this result are as follows. First, $na{\ddot{i}}ve$> Bayes classifiers art known to usually perform better than decision trees in terms of the AUC score. Second, the spammer detection problem in our experiments is likely to be linearly separable.

• Proceedings of the Korean Association of Geographic Inforamtion Studies Conference
• /
• 1999.12a
• /
• pp.21-25
• /
• 1999

Planning support systems(PSS) add more advanced spatial analysis functions than Geographic information systems(GIS) and intertemporal functions to the functions of spatial decision support systems(SDSS). This paper reports the continuing development of a PSS providing a framework that facilitates urban planners and civil engineers in conducting coherent deliberations about planning, design and operation & maintenance(O&M) of water-distribution networks for urban growth management. The PSS using dynamic optimization model, modeling-to-generate-alternatives, value engineering(VE) and life-cycle cost(LCC) can generate network alternatives in consideration of initial cost and O&H cost. Users can define alternatives by the direct manipulation of networks or by the manipulation of parameters in the models. The water-distribution network analysis model evaluates the performance of the user-defined alternatives. The PSS can be extended to include the functions of generating sewer network alternatives, combining water-distribution and sewer networks, eventually the function of planning, design and O&H of housing sites. Capacity expansion by the dynamic water-distribution network optimization model using MINLP includes three advantages over capacity expansion using optimal control theory(Kim and Hopkins 1996): 1) finds expansion alternatives including future capacity expansion times, sizes, locations, and pipe types of a water-distribution network provided, 2) has the capabilities to do the capacity expansion of each link spatially and intertemporally, and 3) requires less interaction between models. The modeling using MINLP is limited in addressing the relationship between cost, price, and demand, which the optimal control approach can consider. Strictly speaking, the construction and O&M costs of water-distribution networks influence the price charged for the served water, which in turn influence the. This limitation can be justified in rather small area because price per unit water in the area must be same as that of neighboring area, i.e., the price is determined administratively. Planners and engineers can put emphasis on capacity expansion without consideration of the relationship between cost, price, and demand.

• Journal of Korea Multimedia Society
• /
• v.17 no.2
• /
• pp.232-239
• /
• 2014

As online texts have been rapidly growing, their automatic classification gains more interest with machine learning methods. Nevertheless, comparatively few research could be found, aiming for Korean texts. Evaluating them with statistical methods are also rare. This study took a sample of tweets and used machine learning methods to classify emotions with features of morphemes and n-grams. As a result, about 76% of emotions contained in tweets was correctly classified. Of the two methods compared in this study, Support Vector Machines were found more accurate than Na$\ddot{i}$ve Bayes. The linear model of SVM was not inferior to the non-linear one. Morphological features did not contribute to accuracy more than did the n-grams.

• Proceedings of the KSR Conference
• /
• 2008.06a
• /
• pp.128-138
• /
• 2008

VE of the scripture season enterprises and it respected LCC analyzes from the research which it sees and to use AHP techniques and definite LCC techniques and probabilistic LCC techniques selects the optimum route the case study which it executed. It presented the quality rating model in about the resultant most route lascivious at the time of VE evaluation, in order to select the alternative of optimum AHP techniques which are one in decision-making technique and an evaluation item by weight and a grade it applied the mountaintop it did. Also the definite LCC analyzer law departments of existing together it applied the probabilistic LCC techniques which use Monte Carlo Simulation in about analytical prices and reliability height boil. The economical efficiency was excellent with VE/LCC analytical resultant route and facility size abridgment, the rivers most it will be able to minimize an environmental effect with short distance traverse, the selection this hit preparation LCC which separates from the land use side decreased, the value (V) above 22.0% with the fact that it improves. And, the reliability of the probabilistic LCC analytical resultant analytical results in compliance with Monte Carlo Simulation with 90.3% was very analyzed with the fact that it is a high level.

• Journal of Radiation Industry
• /
• v.6 no.3
• /
• pp.281-287
• /
• 2012

The model plant, Arabidopsis thaliana is the subject of an international genome research project. Massive doses of ionizing radiation have been shown to induce physiological changes in plants. The wild-type (Ler) Arabidopsis plants were irradiated with 100 Gy and 800 Gy of gamma-ray. Gibberellin (GA) affects developmental processes and responses according to the various environment conditions in diverse plant. The 13 GA isomers were analyzed at vegetative (VE) and reproductive (RE) stages by HPLC. Total GA contents were reduced with the increase in radiation doses at VE and RE stages. Specifically, levels of GA3, GA4, GA12, and GA34 were significantly reduced with the increase of radiation doses. Oligonucleotide microarrays analysis was performed with Arabidopsis plants at different developmental stages and doses of gamma-ray. Through the microarray data, we isolated 41 genes related to GA biosynthesis and signaling transduction. Expression of these genes was also decreased as the reduction of GA contents. Interestingly, in GA signaling related gene expression, gibberellin-responsive protein, putative (At2g18420) was down-regulated at VE and RE stages. Myb21 (At3g27810), Myb24 (At5g40350), and Myb57 (At3g01530) was down-regulated at RE stage. In GA biosynthesis related gene expression, YAP169 (At5g07200) and GA20ox2 (At5g51810) were down-regulated at 100 Gy treatment of VE stage and 800 Gy treatment of RE stage in cytoplasm, respectively. However, exceptively, GA3ox2 (At1g80340) was up-regulated at 100 Gy treatment of RE stage in cytoplasm. In this study, the wild type (Ler) Arabidopsis plants showed differences in response with development stage at the various doses of gamma-rays. GA contents change was reported in gamma irradiated plant.