• Proceedings of the Korean Information Science Society Conference
• /
• 2008.06c
• /
• pp.482-486
• /
• 2008

본 논문에서는 하드 디스크 드라이브(Hard Disk Drive, HDD) 생산 공정 과정에서 발생할 수 있는 불량 HDD의 결함 분포에 대해서 패턴을 자동으로 분류해주는 기법을 제시한다. 이를 위해서 표준 패턴 클래스로 분류되어 있는 불량 HDD의 각 클래스의 확률 모델을 GMM(Gaussian Mixture Model)로 가정한다. 실험은 전문가에 의해 분류된 실제 HDD 결함 분포로부터 5가지의 특징 값들을 추출한 후, 결함 분포의 클래스를 표현할 수 있는 GMM의 파라미터(Parameter)를 학습한다. 각 모델의 파라미터를 추정하기 위해 EM(Expectation Maximization) 알고리즘을 사용한다. 학습된 GMM의 분류 테스트는 학습에 사용되지 않은 HDD 결함 분포에서 5가지의 특징 값을 입력 값으로 추정된 모델들의 파라미터 값에 의해 사후 확률을 구한다. 계산된 확률 값 중 가장 큰 값을 갖는 모델의 클래스를 표준 패턴 클래스로 분류한다. 그 결과 제시된 GMM을 이용한 HDD의 패턴 분류의 결과 96.1%의 정답률을 보여준다.

• Journal of the Korean Institute of Educational Facilities
• /
• v.24 no.6
• /
• pp.11-18
• /
• 2017

Education facilities have much affect to make a good condition for the learning environment. Therefore, various approaches have been conducted to improve the physical, social and educational achievement. Especially, the physical aspect is very important to get rid of the building defect and improve the student their learning environment. For these, it needs to explain the performance and function of components and materials, which is linked with the deterioration degree. The deterioration degree is a imperative factor to make a decision whether the component would be repaired or not and to provide the repair scope of its component. In this paper, it aimed at making the deterioration degree model of roof proof under the hypothesis of which deterioration degree would be equal the repair cost at this time. Results of the study are shown that first, the $3^{rd}$ function is most proper to explain the deterioration degree model among 11 functions in view of resulted statistics. Second, the inflection of deterioration is shown at 15yr of the elementary school and 13yr of the middle and high school. This study has a limit of disclassification of the component or materials and it is, therefore, favorable to include the classification of waterproof material and work. These results would make a change from the breakdown maintenance to preventive maintenance and give a decent the learning environment for student.

• Journal of the Korean Society of Manufacturing Process Engineers
• /
• v.19 no.7
• /
• pp.7-15
• /
• 2020

Recently, the fourth industrial revolution has been presented as a new paradigm and additive manufacturing (AM) has become one of the most important topics. For this reason, process monitoring for each cross-sectional layer of additive metal manufacturing is important. Particularly, deep learning can train a machine to analyze, optimize, and repair defects. In this paper, image classification is proposed by learning images of defects in the metal cross sections using the convolution neural network (CNN) image labeling algorithm. Defects were classified into three categories: crack, porosity, and hole. To overcome a lack-of-data problem, the amount of learning data was augmented using a data augmentation algorithm. This augmentation algorithm can transform an image to 180 images, increasing the learning accuracy. The number of training and validation images was 25,920 (80 %) and 6,480 (20 %), respectively. An optimized case with a combination of fully connected layers, an optimizer, and a loss function, showed that the model accuracy was 99.7 % and had a success rate of 97.8 % for 180 test images. In conclusion, image labeling was successfully performed and it is expected to be applied to automated AM process inspection and repair systems in the future.

• MALSORI
• /
• no.58
• /
• pp.67-81
• /
• 2006

For a large corpus of time-aligned data, HMM based approaches are most widely used for automatic segmentation, providing a consistent and accurate phone labeling scheme. There are two methods for training in HMM. Flat starting method has a property that human interference is minimized but it has low accuracy. Bootstrap method has a high accuracy, but it has a defect that manual segmentation is required In this paper, a new algorithm is proposed to minimize manual work and to improve the performance of automatic segmentation. At first phase, voiced, unvoiced and silence classification is performed for each speech data frame. At second phase, the phoneme sequence is aligned dynamically to the voiced/unvoiced/silence sequence according to the acoustic phonetic rules. Finally, using these segmented speech data as a bootstrap, phoneme model parameters based on HMM are trained. For the performance test, hand labeled ETRI speech DB was used. The experiment results showed that our algorithm achieved 10% improvement of segmentation accuracy within 20 ms tolerable error range. Especially for the unvoiced consonants, it showed 30% improvement.

• Soft Computing and Machine Intelligence
• /
• v.1 no.1
• /
• pp.1-10
• /
• 2021

Analysis and Tracking of bug reports is a challenging field in software repositories mining. It is one of the fundamental ways to explores a large amount of data acquired from defect tracking systems to discover patterns and valuable knowledge about the process of bug triaging. Furthermore, bug data is publically accessible and available of the following systems, such as Bugzilla and JIRA. Moreover, with robust machine learning (ML) techniques, it is quite possible to process and analyze a massive amount of data for extracting underlying patterns, knowledge, and insights. Therefore, it is an interesting area to propose innovative and robust solutions to analyze and track bug reports originating from different open source projects, including Mozilla and Eclipse. This research study presents an ML-based classification model to analyze and track bug defects for enhancing software engineering management (SEM) processes. In this work, Artificial Neural Network (ANN) and Naive Bayesian (NB) classifiers are implemented using open-source bug datasets, such as Mozilla and Eclipse. Furthermore, different evaluation measures are employed to analyze and evaluate the experimental results. Moreover, a comparative analysis is given to compare the experimental results of ANN with NB. The experimental results indicate that the ANN achieved high accuracy compared to the NB. The proposed research study will enhance SEM processes and contribute to the body of knowledge of the data mining field.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2003.09a
• /
• pp.539-542
• /
• 2003

A motor is the workhorse of our industry. The issues of preventive and condition-based maintenance, online monitoring, system fault detection, diagnosis, and prognosis are of increasing importance. Different internal motor faults (e.g., inter-turn short circuits, broken bearings, broken rotor bars) along with external motor faults (e.g., phase failure, mechanical overload, blocked rotor) are expected to happen sooner or later. This paper introduces the fault detection technique of induction motors based upon the stator current. The fault motors have rotor bar broken or rotor unbalance defect, respectively. The stator currents are measured by the current meters and stored by the time domain. The time domain is not suitable to represent the current signals, so the frequency domain is applied to display the signals. The Fourier Transformer is used for the conversion of the signal. After the conversion of the signals, the features of the signals have to be extracted by the signal processing methods like a wavelet analysis, a spectrum analysis, etc. The discovered features are entered to the pattern classification model such as a neural network model, a polynomial neural network, a fuzzy inference model, etc. This paper describes the fault detection results that use wavelet decomposition. The wavelet analysis is very useful method for the time and frequency domain each. Also it is powerful method to detect the features in the signals.

• Journal of Korean Society for Quality Management
• /
• v.48 no.2
• /
• pp.297-307
• /
• 2020

Purpose: The purpose of this study is to analyze the factors that affect ammunition performance by applying data mining techniques to the Ammunition Stockpile Reliability Program (ASRP) data of the 155mm propelling charge. Methods: The ASRP data from 1999 to 2017 have been utilized. Logistic regression and decision tree analysis were used to investigate the factors that affect performance of ammunition. The performance evaluation of each model was conducted through comparison with an artificial neural networks(ANN) model. Results: The results of this study are as follows; logistic regression and the decision tree analysis showed that major defect rate of visual inspection is the most significant factor. Also, muzzle velocity by base charge and muzzle velocity by increment charge are also among the significant factors affecting the performance of 155mm propelling charge. To validate the logistic regression and decision tree models, their classification accuracies have been compared with the results of an ANN model. The results indicate that the logistic regression and decision tree models show sufficient performance which conforms the validity of the models. Conclusion: The main contribution of this paper is that, to our best knowledge, it is the first attempt at identifying the significant factors of ASPR data by using data mining techniques. The approaches suggested in the paper could also be extended to other types ammunition data.

• Journal of Korean Society for Quality Management
• /
• v.49 no.3
• /
• pp.421-431
• /
• 2021

Purpose: For improving outgoing quality, this study presents a novel sampling framework based on predictive analytics. Methods: The proposed framework is composed of three steps. The first step is the variable selection. The knowledge-based and data-driven approaches are employed to select important variables. The second step is the model learning. In this step, we consider the supervised classification methods, the anomaly detection methods, and the rule-based methods. The applying model is the third step. This step includes the all processes to be enabled on real-time prediction. Each prediction model classifies a product as a target sample or random sample. Thereafter intensive quality inspections are executed on the specified target samples. Results: The inspection data of three Samsung products (mobile, TV, refrigerator) are used to check functional defects in the product by utilizing the proposed method. The results demonstrate that using target sampling is more effective and efficient than random sampling. Conclusion: The results of this paper show that the proposed method can efficiently detect products that have the possibilities of user's defect in the lot. Additionally our study can guide practitioners on how to easily detect defective products using stratified sampling

• Journal of Korea Foundry Society
• /
• v.42 no.6
• /
• pp.369-376
• /
• 2022

The die-casting process is an important process for various industries, but there are limitations in the profitability and productivity of related companies due to the high defect rate. In order to overcome this, this study has developed die-casting fault detection modules based on industrial AI technologies. The developed module is constructed from three-stage models depending on the characteristics of the dataset. The first-stage model conducts fault detection based on supervised learning from the dataset without labels. The second-stage model realizes one-class classification based on semi-supervised learning, where the dataset only has production success labels. The third-stage model corresponds to fault detection based on supervised learning, where the dataset includes a small amount of production failure cases. The developed fault detection module exhibited outstanding performance with roughly 96% accuracy for actual process data.

• Journal of Biosystems Engineering
• /
• v.33 no.1
• /
• pp.30-37
• /
• 2008

Dynamic excitation and response analysis is an acceptable method to determine some of physical properties of agricultural product for quality evaluation. There is a difference in the internal viscoelasticity between sound and defective fruits due to the difference of geometric structures, thereby showing different vibration characteristics. This study was carried out to develop a portable piezoelectric film-based glove sensor system that can separate internally damaged watermelons from sound ones using an acoustic impulse response technique. Two piezoelectric sensors based on polyvinylidene fluoride (PVDF) films to measure an impact force and vibration response were separately mounted on each glove. Various signal parameters including number of peaks, energy ratio, standard deviation of peak to peak distance, zero-crossing rate, and integral value of peaks were examined to develop a regression-estimated model. When using SMLR (Stepwise Multiple Linear Regression) analysis in SAS, three parameters, i.e., zeros value, number of peaks, and standard deviation of peaks were selected as usable factors with a coefficient of determination ($r^2$) of 0.92 and a standard error of calibration (SEC) of 0.15. In the validation tests using twenty watermelon samples (sound 9, defective 11), the developed model provided good capability showing a classification accuracy of 95%.