• ETRI Journal
• /
• v.15 no.2
• /
• pp.35-51
• /
• 1993

This paper presents function approximation based on nonparametric estimation. As an estimation model of function approximation, a three layered network composed of input, hidden and output layers is considered. The input and output layers have linear activation units while the hidden layer has nonlinear activation units or kernel functions which have the characteristics of bounds and locality. Using this type of network, a many-to-one function is synthesized over the domain of the input space by a number of kernel functions. In this network, we have to estimate the necessary number of kernel functions as well as the parameters associated with kernel functions. For this purpose, a new method of parameter estimation in which linear learning rule is applied between hidden and output layers while nonlinear (piecewise-linear) learning rule is applied between input and hidden layers, is considered. The linear learning rule updates the output weights between hidden and output layers based on the Linear Minimization of Mean Square Error (LMMSE) sense in the space of kernel functions while the nonlinear learning rule updates the parameters of kernel functions based on the gradient of the actual output of network with respect to the parameters (especially, the shape) of kernel functions. This approach of parameter adaptation provides near optimal values of the parameters associated with kernel functions in the sense of minimizing mean square error. As a result, the suggested nonparametric estimation provides an efficient way of function approximation from the view point of the number of kernel functions as well as learning speed.

• Journal of Institute of Control, Robotics and Systems
• /
• v.6 no.8
• /
• pp.710-718
• /
• 2000

In this paper we described an approach to automation of visual inspection of solder joint defects of SMC(Surface Mounted Components) on PCBs(Printed Circuit Board) by using neural network and fuzzy rule-based classification method. Inherently the surface of the solder joints is curved tiny and specular reflective it induces difficulty of taking good image of the solder joints. And the shape of the solder joints tends to greatly vary with the soldering condition and the shapes are not identical to each other even though the solder joints belong to a set of the same soldering quality. This problem makes it difficult to classify the solder joints according to their qualities. Neural network and fuzzy rule-based classification method is proposed to effi-ciently make human-like classification criteria of the solder joint shapes. The performance of the proposed approach is tested on numerous samples of commercial computer PCB boards and compared with the results of the human inspector performance and the conventional Kohonen network.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 1999.06a
• /
• pp.29-34
• /
• 1999

We present a practical method for removing biotches and restoring their mission data. To detect blotches, we employ a robust approach of local analysis of spatiotemporal anisotropic brightness continuity Our approach uses first-order spatiotemporal directional derivatives to select the smoothest direction for each examined pixel, and puts out the incorruption probability that he examined pixel may not be corrupted by blotches. As the restoration filter, were employ a spatiotemporal fuzzy filter whose response is adaptively controlled according to a fuzzy rule defined by the incorruption probability. The fuzzy filter is composed of the two different filter of the identity filter and the spatiotemporal directional-weighted-mean filter, and will put out an intermediate value between the original input brightness and the directional-weighted-mean brightness. We design the fuzzy rule in advance by a standard supervised learning fuzzy rule in advance by a standard supervised learning method. The computer simulations are presented.

• Proceedings of the KIEE Conference
• /
• 1987.11a
• /
• pp.118-121
• /
• 1987

This paper presents a rule-based approach for automatically selecting critical contingencies in electric power systems. The rules required to perform the task are derived from inspection about results of simulation and expertise of operators. And inherent information of system, for example, topology of system configuration, and flow direction in a line by compensation theorem. etc., which are independent of operating point of system, is stored in the database using the off-line calculation. The approach was investigated using the study of a sample test system. Since it is based on the knowledge engineering technique, efficiency of selection can be improved by updating and adding the rules.

• ETRI Journal
• /
• v.39 no.5
• /
• pp.652-661
• /
• 2017

A mention has a noun or noun phrase as its head and constructs a chunk that defines any meaning, including a modifier. Mention detection refers to the extraction of mentions from a document. In mentions, coreference resolution refers to determining any mentions that have the same meaning. Pointer networks, which are models based on a recurrent neural network encoder-decoder, outputs a list of elements corresponding to an input sequence. In this paper, we propose mention detection using pointer networks. This approach can solve the problem of overlapped mention detection, which cannot be solved by a sequence labeling approach. The experimental results show that the performance of the proposed mention detection approach is F1 of 80.75%, which is 8% higher than rule-based mention detection, and the performance of the coreference resolution has a CoNLL F1 of 56.67% (mention boundary), which is 7.68% higher than coreference resolution using rule-based mention detection.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.9
• /
• pp.2299-2308
• /
• 1997

While most machine learning research has been primarily concerned with the development of systems that implement one type of learning strategy, we use a multistrategy approach which integrates rule induction learning and instance-based learning, and show how this marriage allows for overall better performance. In the rule induction learning phase, we derive an entropy function, based on Hellinger divergence, which can measure the amount of information each inductive rule contains, and show how well the Hellinger divergence measures the importance of each rule. We also propose some heuristics to reduce the computational complexity by analyzing the characteristics of the Hellinger measure. In the instance-based learning phase, we improve the current instance-based learning method in a number of ways. The system has been implemented and tested on a number of well-known machine learning data sets. The performance of the system has been compared with that of other classification learning technique.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.1
• /
• pp.116-132
• /
• 2022

Automated medical claims processing and billing is a popular application domain of information technology. Managing medical related data is a tedious job for healthcare professionals, which distracts them from their main job of healthcare. The technology used in data management has a sound impact on the quality of healthcare data. Most of Information Technology (IT) organizations use conventional software development technology for the implementation of healthcare systems. The objective of this experimental study is to devise a mechanism for use of rule-based expert systems in medical related edits and compare it with the conventional software development technology. A sample of 100 medical edits is selected as a dataset to be tested for implementation using both technologies. Besides empirical analysis, paired t-test is also used to validate the statistical significance of the difference between the two techniques. The conventional software development technology took 254.5 working hours, while rule-based technology took 81 hours to process these edits. Rule-based technology outperformed the conventional systems by increasing the confidence value to 95% and reliability measure to 0.462 (which is < 0.5) which is three times more efficient than conventional software development technology.

• ETRI Journal
• /
• v.37 no.3
• /
• pp.541-550
• /
• 2015

In this paper, we propose a classification-based approach for hybridizing statistical machine translation and rulebased machine translation. Both the training dataset used in the learning of our proposed classifier and our feature extraction method affect the hybridization quality. To create one such training dataset, a previous approach used auto-evaluation metrics to determine from a set of component machine translation (MT) systems which gave the more accurate translation (by a comparative method). Once this had been determined, the most accurate translation was then labelled in such a way so as to indicate the MT system from which it came. In this previous approach, when the metric evaluation scores were low, there existed a high level of uncertainty as to which of the component MT systems was actually producing the better translation. To relax such uncertainty or error in classification, we propose an alternative approach to such labeling; that is, a cut-off method. In our experiments, using the aforementioned cut-off method in our proposed classifier, we managed to achieve a translation accuracy of 81.5% - a 5.0% improvement over existing methods.

• Proceedings of the KIEE Conference
• /
• 1998.11a
• /
• pp.301-303
• /
• 1998

This paper presents a new approach to diagnose and detect faults in oil-filled power transformers based on various dissolved gas analyses. A theoretic fuzzy information model is introduced, An inference scheme which yields the 'most' consistent conclusion proposed. A framework is established that allows various dissolved gas analyses to be combined in a systematic way such as the Dempster-Shafer rule. Good diagnosis accuracy is obtained with the proposed approach.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2003.06a
• /
• pp.1235-1238
• /
• 2003

A rule-based CAPP(computer aided process planning) system is proposed in this research to develop integrated manufacturing process of machining and inspection using OMM(On-Machine Measurement) device. Generally workpiece composed of many primitive form features. This features are determined optimum inspection sequence by analyzing the feature information such as features-relationship, probe approach direction and etc. Proposed paper is more efficient method of CAIP(computer aided inspection planning) considered machining process