• Journal of the Korea Institute of Building Construction
• /
• v.16 no.4
• /
• pp.331-339
• /
• 2016

According to high-rise, complexation, and enlargement of buildings, various construction methods are being developed, and the significance of construction method selection about main work types has emerged as a major interest. However, it has been pointed out that hand-on workers cannot consider project characteristics carefully, and they lack an objective standard or reference for main construction method selection. Hence, the selection is being made depending on hand-on workers' experience and intuition. To solve this problem, various studies have proceeded for construction method selection of main work types using Artificial Intelligence like Fuzzy, AHP and Case-based reasoning. It is difficult to apply many different kinds of construction method selection to every main work type with consideration for characteristics of work types and condition of a construction site when selecting construction method in the field. Accordingly, this study proposed the decision-making model which can apply to fields easily. Using matrix analysis and liner transformation, this study verified consistency of study models applied in the process of soil retaining selection with a case study.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.5
• /
• pp.591-599
• /
• 2020

Deep learning is widely utilized to classify or recognize objects of real-world. An abundance of data is trained on high-performance computers and a trained model is generated, and then the model is loaded in an inferencer. The inferencer is used in various environments, so that it may cause unrecognized objects or low-accuracy objects. To solve this problem, real-world objects are collected and they are trained periodically. However, not only is it difficult to immediately improve the recognition rate, but is not easy to learn an inferencer on embedded devices. We propose a real-time handwriting recognizer based on partial learning on embedded devices. The recognizer provides a training environment which partially learn on embedded devices at every user request, and its trained model is updated in real time. As this can improve intelligence of the recognizer automatically, recognition rate of unrecognized handwriting increases. We experimentally prove that learning and reasoning are possible for 22 numbers and letters on RK3399 devices.

• Journal of the Korean Chemical Society
• /
• v.67 no.5
• /
• pp.378-392
• /
• 2023

The purpose of this study is to analyze the pre-service chemistry teachers' cognition of the nature of model in process of designing and developing models using technology. For this purpose, 19 pre-service chemistry teachers' in the 3rd grade of a education college located in the central region observe experimental phenomena related to Boyle's law presented in the 7th grade science textbook and researchers required the design and development of a model related to the observed experimental results using technology. Based on previous studies, the nature of model were classified into two aspect: 'Representational aspect' and 'Explanatory aspect'. The 'Representational aspect' was classified into 'Representation', 'Abstraction', and 'Simplification', and the 'Explanatory aspect' was classified into 'Analysis', 'Interpretation', 'Reasoning', 'Explanation', and 'Quantification'. The pre-service chemistry teachers' cognition were analyzed by the classification. As a result of the study, the 'Representation' of the 'expressive aspect' was uniformized in the form of space that changes in volume, and the pressure was expressed as the Brightness inside the cylinder or frequency of color change of particles for 'Abstraction'. In the case of 'Simplification', the particle collision was expressed as a perfectly elastic collision, but there was a group that could not simply indicate the type of particle. In the 'Explanatory aspect', in the case of 'Analysis', volume was classified as a manipulated variable, and in the case of 'Interpretation', most groups analyzed the change in pressure through the collision of gas particles. However, the cognition involved in 'Reasoning' was not observed much. In the case of 'Explanation', there were groups that did not succeed in explanation because the area where the particles collided was not set or incorrectly set, and in the case of 'Quantification', there was a group that formulated the number of collisions per unit time, and on the contrary, there was a group that could not quantify the number of collisions because they could not be expressed in numbers.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.15 no.4
• /
• pp.217-231
• /
• 2015

A novel fuzzy learning framework that employs fuzzy inference to solve the problem of Multiple Instance Learning (MIL) is presented. The framework introduces a new class of fuzzy inference systems called Multiple Instance Mamdani Fuzzy Inference Systems (MI-Mamdani). In multiple instance problems, the training data is ambiguously labeled. Instances are grouped into bags, labels of bags are known but not those of individual instances. MIL deals with learning a classifier at the bag level. Over the years, many solutions to this problem have been proposed. However, no MIL formulation employing fuzzy inference exists in the literature. Fuzzy logic is powerful at modeling knowledge uncertainty and measurements imprecision. It is one of the best frameworks to model vagueness. However, in addition to uncertainty and imprecision, there is a third vagueness concept that fuzzy logic does not address quiet well, yet. This vagueness concept is due to the ambiguity that arises when the data have multiple forms of expression, this is the case for multiple instance problems. In this paper, we introduce multiple instance fuzzy logic that enables fuzzy reasoning with bags of instances. Accordingly, a MI-Mamdani that extends the standard Mamdani inference system to compute with multiple instances is introduced. The proposed framework is tested and validated using a synthetic dataset suitable for MIL problems. Additionally, we apply the proposed multiple instance inference to fuse the output of multiple discrimination algorithms for the purpose of landmine detection using Ground Penetrating Radar.

• Journal of Distribution Science
• /
• v.18 no.4
• /
• pp.79-85
• /
• 2020

Purpose: This study explores how cognitive impairment caused by social exclusion experience can be explained through cognitive narrowing and how it influences consumer's judgment and reasoning and results pseudodiagnosticity bias towards distribution brands. This study examines the characteristics of cognitive narrowing, which is one of the strategies for overcoming the negative emotions resulting from social exclusion, and how cognitive errors called pseudodiagnosticity bias occur due to cognitive narrowing in the evaluation of distribution brands. Research design, data and methodology: Present study was performed with 77 college students in Seoul. Participants were randomly assigned to the group who experienced social exclusion and the group who did not experience social exclusion. The analysis has been made of how the degree of bias of pseudodiagnosticity differs according to the experience of social exclusion by t-test. Results: The group who experienced social exclusion had a higher level of pseudodiagnosticity bias towards distribution brands than the group who did not experience social exclusion. Conclusions: This study confirmed what characteristics of cognitive narrowing, which is one of the strategies for overcoming the negative emotions resulting from social exclusion, and how cognitive errors called pseudodiagnosticity bias occur due to cognitive narrowing. Implications and future research directions were discussed and suggested.

• Journal of Fisheries and Marine Sciences Education
• /
• v.29 no.3
• /
• pp.757-766
• /
• 2017

This study investigated not only the existence of differently functioned item due to gender but also domain. In this study, the randomly selected data of TIMSS 2007, which consist of 681 male and 646 women, were analyzed. To detect differently functioned items, this study employed Raju method. For Raju method, three-parameter logistic model was selected. Signed and unsigned area between two item characteristic curve were measured within the real ability range. An item which was detected commonly SA and UA area in Raju method was defined as a differently functioned item. As a result of this study, six items among twenty seven items of mathematics in the TIMSS 2007 were differently functioned item. Five items among those six items, were in favor of boys and one item was in favor of girls. Number, Geometric Shapes and Measures, and Applying were in favor of boys. but Data Display, Reasoning were in favor of girls. The conclusion of this study was summarized as existing differently functioned items in TIMSS 2007 and difference between favorable domain based gender. Finally, it is desirable to consider the differently functioned items by relating those item content for improving the test reliability of TIMSS 2007.

• Journal of Information Technology Services
• /
• v.15 no.3
• /
• pp.97-114
• /
• 2016

One of the key success factors for the intelligent IT service which is characterized by personalization and automation, is to obtain relevant data from either sensors or data storage for reasoning, analyzing and forecasting. The availability of the open data sources such as public portal sites remarkably increases the efficiency and quality of the intelligent IT service. However, with the condition that not all data in the existing public or private sites are opened or have various types of openness, it prohibits the value of utilization. For these reasons, it is highly required to evaluate the extent of openness of data storage. However, there are only a few studies which explore the factors which affect the degree of data openness with respect to intelligent IT services. Hence, this study aims to propose an evaluation model including the indices to evaluate a process of opening data for the intelligent IT service from a viewpoint of data utilization process. The indices are applied to evaluate the actual multinational websites, which provide public data for verification. We also discuss the implications of the evaluation according to the results.

• Asia pacific journal of information systems
• /
• v.7 no.1
• /
• pp.67-83
• /
• 1997

The utility of learning techniques in investment analysis has been demonstrated in many areas, ranging from forecasting individual stocks to entire market indexes. To date, however, the application of artificial intelligence to financial forecasting has focused largely on short predictive horizons. Usually the forecast window is a single period ahead; if the input data involve daily observations, the forecast is for one day ahead; if monthly observations, then a month ahead; and so on. Thus far little work has been conducted on the efficacy of long-term prediction involving multiperiod forecasting. This paper examines the impact of alternative procedures for extended prediction using knowledge discovery techniques. One dimension in the study involves temporal granularity: a single jump from the present period to the end of the forecast window versus a web of short-term forecasts involving a sequence of single-period predictions. Another parameter relates to the numerosity of input variables: a technical approach involving only lagged observations of the target variable versus a fundamental approach involving multiple variables. The dual possibilities along each of the granularity and numerosity dimensions entail a total of 4 models. These models are first evaluated using neural networks, then compared against a multi-input jump model using case based reasoning. The computational models are examined in the context of forecasting the S&P 500 index.

• Proceedings of the KIEE Conference
• /
• 2011.07a
• /
• pp.1113-1114
• /
• 2011

This paper is proposed design of high performance controller of SynRM drive using FNN and NNC. Also, This paper is proposed of designing fuzzy neural network controller(FNNC) which adopts the fuzzy logic to the artificial neural network(ANN). FNNC combines the capability of fuzzy reasoning in handling uncertain information and the capability of neural network in learning from processes. This controller is controlled speed using FNNC and model reference adaptive fuzzy control(MFC), and estimation of speed using ANN. The performance of proposed controller was demonstrated through response results. The results confirm that the proposed controller is high performance and robust under the variation of load torque and parameters.

• Journal of Institute of Control, Robotics and Systems
• /
• v.15 no.11
• /
• pp.1115-1121
• /
• 2009

For HRI (Human-Robot Interaction) in daily life, robots need to recognize non-rigid objects such as clothes and blankets. However, the recognition of non-rigid objects is challenging because of the variation of the shapes according to the places and laying manners. In this paper, the cognition of non-rigid object based on a cognitive system is presented. The characteristics of non-rigid objects are analysed in the view of HRI and referred to design a framework for the cognition of them. We adopt a linguistic cognitive system for describing all of the events happened to robots. When an event related to the non-rigid objects is occurred, the cognitive system describes the event into a sentential form and stores it at a sentential memory, and depicts the objects with a spatial model for being used as references. The cognitive system parses each sentence syntactically and semantically, in which the nouns meaning objects are connected to their models. For answering the questions of humans, sentences are retrieved by searching temporal information in the sentential memory and by spatial reasoning in a schematic imagery. Experiments show the feasibility of the cognitive system for cognizing non-rigid objects in HRI.