• Journal of Intelligence and Information Systems
• /
• v.17 no.4
• /
• pp.109-130
• /
• 2011

One of the major problems in the area of data mining is the size of the data, as most data set has huge volume these days. Streams of data are normally accumulated into data storages or databases. Transactions in internet, mobile devices and ubiquitous environment produce streams of data continuously. Some data set are just buried un-used inside huge data storage due to its huge size. Some data set is quickly lost as soon as it is created as it is not saved due to many reasons. How to use this large size data and to use data on stream efficiently are challenging questions in the study of data mining. Stream data is a data set that is accumulated to the data storage from a data source continuously. The size of this data set, in many cases, becomes increasingly large over time. To mine information from this massive data, it takes too many resources such as storage, money and time. These unique characteristics of the stream data make it difficult and expensive to store all the stream data sets accumulated over time. Otherwise, if one uses only recent or partial of data to mine information or pattern, there can be losses of valuable information, which can be useful. To avoid these problems, this study suggests a method efficiently accumulates information or patterns in the form of rule set over time. A rule set is mined from a data set in stream and this rule set is accumulated into a master rule set storage, which is also a model for real-time decision making. One of the main advantages of this method is that it takes much smaller storage space compared to the traditional method, which saves the whole data set. Another advantage of using this method is that the accumulated rule set is used as a prediction model. Prompt response to the request from users is possible anytime as the rule set is ready anytime to be used to make decisions. This makes real-time decision making possible, which is the greatest advantage of this method. Based on theories of ensemble approaches, combination of many different models can produce better prediction model in performance. The consolidated rule set actually covers all the data set while the traditional sampling approach only covers part of the whole data set. This study uses a stock market data that has a heterogeneous data set as the characteristic of data varies over time. The indexes in stock market data can fluctuate in different situations whenever there is an event influencing the stock market index. Therefore the variance of the values in each variable is large compared to that of the homogeneous data set. Prediction with heterogeneous data set is naturally much more difficult, compared to that of homogeneous data set as it is more difficult to predict in unpredictable situation. This study tests two general mining approaches and compare prediction performances of these two suggested methods with the method we suggest in this study. The first approach is inducing a rule set from the recent data set to predict new data set. The seocnd one is inducing a rule set from all the data which have been accumulated from the beginning every time one has to predict new data set. We found neither of these two is as good as the method of accumulated rule set in its performance. Furthermore, the study shows experiments with different prediction models. The first approach is building a prediction model only with more important rule sets and the second approach is the method using all the rule sets by assigning weights on the rules based on their performance. The second approach shows better performance compared to the first one. The experiments also show that the suggested method in this study can be an efficient approach for mining information and pattern with stream data. This method has a limitation of bounding its application to stock market data. More dynamic real-time steam data set is desirable for the application of this method. There is also another problem in this study. When the number of rules is increasing over time, it has to manage special rules such as redundant rules or conflicting rules efficiently.

• Journal of the Korean Society of Manufacturing Process Engineers
• /
• v.13 no.5
• /
• pp.28-34
• /
• 2014

A planetary gear train is more compact and endures greater amounts of transmission power compared to other gear systems. Although planetary gear systems operate in small volumes, they are capable of very high efficiency due to the compact combination of their gears in the planetary gear system. They also have outstanding efficiency of only 3% for power transmission, tantamount to the power loss that occurs in each of the shift stages. Given these advantages, planetary gear systems are used in the driving systems of, which are widely used in automobile transmissions, machine tools, semiconductor equipment, and in other areas in industrial fields. Current structural equipment requires higher efficiency and greater torque levels. According to these needs, we have designed a complex planetary gear system which creates higher levels of torque. In this paper, an evaluation of strength designs for the proposed planetary gear system was conducted to ensure the stability of the gear. In addition, a durability analysis based on Miner's rule was performed using RS B 0095 device.

• IE interfaces
• /
• v.11 no.3
• /
• pp.23-40
• /
• 1998

In this paper, a method to determine the optimal configuration of operating policies in an integrated-automated manufacturing system using the Taguchi method and computer simulation experiments is presented. An integrated-automated manufacturing system called direct-input-output manufacturing system(DIOMS) is described. We only consider the operational aspect of the DIOMS. Four operating policies including input sequencing control, dispatching rule for the storage/retrieval(S/R) machine, machine center-based part type selection rule, and storage assignment policy are treated as design factors. The number of machine centers, the number of part types, demand rate, processing time and the rate of each part type, vertical and horizontal speed of the S/R machine, and the size of a local buffer in the machine centers are considered as noise factors in generating various manufacturing system environment. For the performance characteristics, mean flow time and throughput are adopted. A robust design experiment with inner and outer orthogonal arrays are conducted by computer simulation, and an optimal configuration of operating policies is presented which consists of a combination of the level of each design factor. The validity of the optimal configurations is investigated by comparing their signal-to-noise ratios with those obtained with full factorial designs.

• Proceedings of the IEEK Conference
• /
• 2008.06a
• /
• pp.20-21
• /
• 2008

In the IEEE 802.16 Wireless MAN standard, the best effort service class is ranked on the lowest position in priority so that only scarce resource may be available for it. Also, the best effort service is usually assisted by a MAC scheme based on reservation ALOHA without explicit acknowledgement. However, the standard only specifies the skeleton of the MAC scheme. In this paper, we propose some rules to specify the indefinite part of the MAC scheme as well as to improve the throughput and delay performance of the MAC scheme. As generic rules for demanding and granting resource, we first propose non-gated exhaustive, gated exhaustive, non-gated limited, and gated limited demand rules, and deficient and full grant rules. Secondly, we propose grant regulation rules, identified as replacement and lifetime rules, to avoid excessive grant incurred by not giving acknowledgement. Noting the residual resource is inflated by adopting a grant regulation rule, we finally propose a rule for investing the residual resource to a subscriber station. Simulation results confirm that a combination of proposed rules improves the performance of the MAC scheme.

• ETRI Journal
• /
• v.39 no.2
• /
• pp.191-201
• /
• 2017

This paper proposes an approach using taxonomic relatedness for answer-type recognition and type coercion in a question-answering system. We introduce a question analysis method for a lexical answer type (LAT) and semantic answer type (SAT) and describe the construction of a taxonomy linking them. We also analyze the effectiveness of type coercion based on the taxonomic relatedness of both ATs. Compared with the rule-based approach of IBM's Watson, our LAT detector, which combines rule-based and machine-learning approaches, achieves an 11.04% recall improvement without a sharp decline in precision. Our SAT classifier with a relatedness-based validation method achieves a precision of 73.55%. For type coercion using the taxonomic relatedness between both ATs and answer candidates, we construct an answer-type taxonomy that has a semantic relationship between the two ATs. In this paper, we introduce how to link heterogeneous lexical knowledge bases. We propose three strategies for type coercion based on the relatedness between the two ATs and answer candidates in this taxonomy. Finally, we demonstrate that this combination of individual type coercion creates a synergistic effect.

• International Journal of Korean Welding Society
• /
• v.1 no.2
• /
• pp.36-44
• /
• 2001

To get the appropriate welding process variables, mathematical modeling in conjunction with many experiments is necessary to predict the magnitude of weld bead shape. Even though the experimental results are reliable, it has a difficulty in accurately predicting welding process variables for the desired weld bead shape because of nonlinear and complex characteristics of welding processes. The welding condition determined for the desired weld bead shape may cause the weld defect if the welding current/voltage/speed combination is improperly selected. In this study, the $2^{n-1}$ fractional factorial design method and correlation parameter were used to investigate the effect of the welding process variables on the fillet joint shape, and the multiple non-linear regression analysis was used for modeling the gas metal arc welding(GMAW）parameters of the fillet joint. Finally, a fuzzy rule-based method and a neural network method were proposed so that the complexity and non-linearity of arc welding phenomena could be effectively overcome. The performance of the proposed neuro-fuzzy system was evaluated through various experiments. The experimental results showed that the proposed neuro-fuzzy system could effectively check the welding conditions as to whether or not weld defects would occur, and also adjust the welding conditions to avoid these weld defects.

• Journal of Mechanical Science and Technology
• /
• v.19 no.spc1
• /
• pp.388-394
• /
• 2005

Recently, the authors have developed a method for real-time dynamics of multibody systems, which combines a semi-recursive formulation to derive the equations of motion in dependent relative coordinates, along with an augmented Lagrangian technique to impose the loop closure conditions. The following numerical integration procedures, which can be grouped into the so-called structural integrators, were tested : trapezoidal rule, Newmark dissipative schemes, HHT rule, and the Generalized-${\alpha}$ family. It was shown that, for large multi body systems, Newmark dissipative was the best election since, provided that the adequate parameters were chosen, excellent behavior was achieved in terms of efficiency and robustness with acceptable levels of accuracy. In the present paper, the performance of the described method in combination with another group of integrators, the Implicit Runge-Kutta family (IRK), is analyzed. The purpose is to clarify which kind of IRK algorithms can be more suitable for real-time applications, and to see whether they can be competitive with the already tested structural family of integrators. The final objective of the work is to provide some practical criteria for those interested in achieving real-time performance for large and complex multibody systems.

• Proceedings of the KIEE Conference
• /
• 1997.07c
• /
• pp.891-893
• /
• 1997

In this paper an intelligent power transformer protective relaying algorithm based on Fuzzy Decision-Making is presented. The introduced protection algorithm contains several internal fuzzy rule-bases including bpa(Basic Probability Assignment: m) which are subject to off-line pre-installation by the analysis of the transformer transient characteristics for detecting the internal fault. Dempster-Shafer's rule of combination is used for the inference method with rules to decide the situation of a transformer, The proposed algorithm immunes to the saturation of transformer, inrush conditions, over excitation, and external fault. The included results of testing show practically sufficient sensitivity and selectivity of the proposed algorithm.

• Journal of the Ergonomics Society of Korea
• /
• v.23 no.1
• /
• pp.1-12
• /
• 2004

The purpose of this study was to develop an automatic expert system for the evaluation of human sensibility, where human sensibility can be inferred from objective physiological signals. The study aim was also to develop an algorithm in which human arousal and pleasant level can be judged by using measured physiological signals. Fuzzy theory was applied for mathematical handling of the ambiguity related to evaluation of human sensibility. and the degree of belonging to a certain sensibility dimension was quantified by membership function through which the sensibility evaluation was able to be done. Determining membership function was achieved using results from a physiological signal database of arousal/relaxation and pleasant/unpleasant that was generated from imagination. To induce one final result (arousal and pleasant level) based on measuring the results of more than 2 physiological signals and the membership function of each physiological signal. Dempster-Shafer's rule of combination in evidence was applied, through which the final arousal and pleasant level was inferred.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.61 no.8
• /
• pp.1164-1171
• /
• 2012

Recently, machine olfactory systems as an artificial substitute of the human olfactory system are being studied actively because they can scent dangerous gases and identify the type of gases in contamination areas instead of the human. In this paper, we present an effective design method for the gas identification system. Even though dimensionality reduction is the very important part, in pattern analysis, We handled effectively the dimensionality reduction by grouping the sensors of which the measured patterns are similar each other, where genetic algorithms were used for combination optimization. To identify the gas type, we constructed the hierarchical rule base with two frames by using rough set theory. The first frame is to accept measurement characteristics of each sensor and the other one is to reflect the identification patterns of each group. Thus, the proposed methods was able to accomplish effectively dimensionality reduction as well as accurate gas identification. In simulation, we demonstrated the effectiveness of the proposed methods by identifying five types of gases.