• School Mathematics
• /
• v.11 no.1
• /
• pp.1-15
• /
• 2009

Some researchers (Jones et al., 1997; Tarr & Jones, 1997; Tarr & Lannin, 2005) have worked on students' probabilistic thinking framework. These studies contributed to an understanding of students' thinking in probability by depicting levels. However, understanding middle school students' probabilistic thinking is limited to the concepts in conditional probability and independence. In this study, the framework to understand middle school students' thinking in probability is integrated on the works of Jones et al. (1997), Polaki (2005) and Tarr and Jones (1997). As in their works, depicting levels of probabilistic thinking is focused on the concepts and skills for students in middle school. The concepts and skills considered as being necessary for middle school students were integrated from NCTM documents and NAEP frameworks.

• The Korean Journal of Applied Statistics
• /
• v.32 no.5
• /
• pp.763-782
• /
• 2019

This paper proposes a probabilistic document ranking model incorporating term dependencies. Document ranking is a fundamental information retrieval task. The task is to sort documents in a collection according to the relevance to the user query (Qin et al., Information Retrieval Journal, 13, 346-374, 2010). A probabilistic model is a model for computing the conditional probability of the relevance of each document given query. Most of the widely used models assume the term independence because it is challenging to compute the joint probabilities of multiple terms. Words in natural language texts are obviously highly correlated. In this paper, we assume a multinomial distribution model to calculate the relevance probability of a document by considering the dependency structure of words, and propose an information retrieval model to rank a document by estimating the probability with the maximum entropy method. The results of the ranking simulation experiment in various multinomial situations show better retrieval results than a model that assumes the independence of words. The results of document ranking experiments using real-world datasets LETOR OHSUMED also show better retrieval results.

• School Mathematics
• /
• v.17 no.2
• /
• pp.241-255
• /
• 2015

The purpose of this study is to analyze teachers' understanding on generating random number in Monte Carlo simulation and to provide educational implications in school practice. The results showed that the 70% of the teachers selected wrong ideas from three types for random-number as strategies for problem solving a probability problem and also they make some errors to justify their opinion. The first kind of the errors was that the probability of a point or boundary was equal to the value of the probability density function in the continuous probability distribution. The second kind of the errors was that the teachers failed to recognize that the sample space has been changed by conditional probability. The third kind of the errors was that when two random variables X, Y are independence of each other, then only, joint probability distribution is satisfied $P(X=x,\;Y=y)=p(X=x){\times}P(Y=y{\mid}X=x)$.

• Proceedings of the IEEK Conference
• /
• 1999.06a
• /
• pp.1009-1012
• /
• 1999

An analytical performance model that can predict the performance of a superscalar processor employing multiple branch prediction is introduced. The model is based on the conditional independence probability and the basic block size of instructions, with the degree of multiple branch prediction, the fetch rate, and the window size of a superscalar architecture. Trace driven simulation is performed for the subset of SPEC integer benchmarks, and the measured IPCs are compared with the results derived from the model. As the result, our analytic model could predict the performance of the superscalar processor using multiple branch prediction within 6.6 percent on the average.

• Journal of the Korean Statistical Society
• /
• v.27 no.1
• /
• pp.25-55
• /
• 1998

Maximum likelihood estimates (MLEs) for recursive models of categorical variables are discussed under an EM framework. Since MLEs by EM often depend on the choice of the initial values for MLEs, we explore reasonable rules for selecting the initial values for EM. Simulation results strongly support the proposed rules.

• Proceedings of the Korean Association of Geographic Inforamtion Studies Conference
• /
• 2003.04a
• /
• pp.65-70
• /
• 2003

The weight-of-evidence model one of the Bayesian probability model was applied to the task of evaluating landslide susceptibility using GIS. Using the location of the landslides and spatial database such as topography, soil, forest, geology, land use and lineament, the weight-of-evidence model was applied to calculate each factor's rating at Boun area in Korea where suffered substantial landslide damage fellowing heavy rain in 1998, The factors are slope, aspect and curvature from the topographic database, soil texture, soil material, soil drainage, soil effective thickness, and topographic type from the soil database, forest type, timber diameter, timber age and forest density from the forest map, lithology from the geological database, land use from Landsat TM satellite image and lineament from IRS satellite image. Tests of conditional independence were performed for the selection of the factors, allowing the 43 combinations of factors to be analyzed. For the analysis, the contrast value, W$\^$＋/and W$\^$-/, as each factor's rating, were overlaid to map laudslide susceptibility. The results of the analysis were validated using the observed landslide locations, and among the combinations, the combination of slope, curvature, topographic, timber diameter, geology and lineament show the best results. The results can be used for hazard prevention and planning land use and construction

• Journal of KIISE:Software and Applications
• /
• v.27 no.8
• /
• pp.855-863
• /
• 2000

Although Behavior-Knowledge Space (BKS) method, one of well known decision combination methods, does not need any assumptions in combining the multiple experts, it should theoretically build exponential storage spaces for storing and managing jointly observed K decisions from K experts. That is, combining K experts needs a (K+1)st-order probability distribution. However, it is well known that the distribution becomes unmanageable in storing and estimating, even for a small K. In order to overcome such weakness, it has been studied to decompose a probability distribution into a number of component distributions and to approximate the distribution with a product of the component distributions. One of such previous works is to apply a conditional independence assumption to the distribution. Another work is to approximate the distribution with a product of only first-order tree dependencies or second-order distributions as shown in [1]. In this paper, higher order dependency than the first-order is considered in approximating the distribution and a dependency-based framework is proposed to optimally approximate the (K+1)st-order probability distribution with a product set of dth-order dependencies where ($1{\le}d{\le}K$), and to combine multiple experts based on the product set using the Bayesian formalism. This framework was experimented and evaluated with a standardized CENPARMI data base.

• Journal of Digital Convergence
• /
• v.11 no.11
• /
• pp.409-414
• /
• 2013

Naive Bayesian classifiers are a powerful and well-known type of classifiers that can be easily induced from a dataset of sample cases. However, the strong conditional independence assumptions can sometimes lead to weak classification performance. Normally, naive Bayesian classifiers use Gaussian distributions to handle continuous attributes and to represent the likelihood of the features conditioned on the classes. The probability density of attributes, however, is not always well fitted by a Gaussian distribution. Another eminent type of classifier is the neuro-fuzzy classifier, which can learn fuzzy rules and fuzzy sets using supervised learning. Since there are specific structural similarities between a neuro-fuzzy classifier and a naive Bayesian classifier, the purpose of this study is to apply learning distribution graphs constructed by a neuro-fuzzy network to naive Bayesian classifiers. We compare the Gaussian distribution graphs with the fuzzy distribution graphs for the naive Bayesian classifier. We applied these two types of distribution graphs to classify leukemia and colon DNA microarray data sets. The results demonstrate that a naive Bayesian classifier with fuzzy distribution graphs is more reliable than that with Gaussian distribution graphs.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.1
• /
• pp.329-338
• /
• 2019

In this study, we analyze consumer preferences based on the agricultural box scheme attributes, and make a suggestion for business revival. We estimate the marginal willingness to pay (MWTP) for box scheme attributes using a choice experiment. Attributes include the bundle method, the delivery method, and price. To select an efficient model for statistical analysis, we evaluate the conditional logit model, heteroscedastic extreme value model(HEV model), multinomial probit model, and mixed logit model under different assumptions. The results of these four models show that the bundle method, the delivery method, and price are statistically significant in explaining the probability of participation in a box scheme. The results of likelihood ratio tests show that the heteroscedastic extreme value model is the most appropriate for our survey data. The results also indicate that MWTP for a change from fixed type to selection type is KRW 7,096.6. MWTP for a change from parcel service to direct delivery and cold-chain delivery are KRW 3,497.5 and KRW 7,532.7, respectively. The results of this study may contribute to the government's local food policies.