• 한국음향학회지
• /
• 제22권5호
• /
• pp.388-402
• /
• 2003

In this paper, we carried out the word, 4 continuous digits. continuous, and task-independent word recognition experiments to verify the effectiveness of the re-defined phoneme-likely units (PLUs) for the phonetic decision tree based HM-Net (Hidden Markov Network) context-dependent (CD) acoustic modeling in Korean appropriately. In case of the 48 PLUs, the phonemes /ㅂ/, /ㄷ/, /ㄱ/ are separated by initial sound, medial vowel, final consonant, and the consonants /ㄹ/, /ㅈ/, /ㅎ/ are also separated by initial sound, final consonant according to the position of syllable, word, and sentence, respectively. In this paper. therefore, we re-define the 39 PLUs by unifying the one phoneme in the separated initial sound, medial vowel, and final consonant of the 48 PLUs to construct the CD acoustic models effectively. Through the experimental results using the re-defined 39 PLUs, in word recognition experiments with the context-independent (CI) acoustic models, the 48 PLUs has an average of 7.06%, higher recognition accuracy than the 39 PLUs used. But in the speaker-independent word recognition experiments with the CD acoustic models, the 39 PLUs has an average of 0.61% better recognition accuracy than the 48 PLUs used. In the 4 continuous digits recognition experiments with the liaison phenomena. the 39 PLUs has also an average of 6.55% higher recognition accuracy. And then, in continuous speech recognition experiments, the 39 PLUs has an average of 15.08% better recognition accuracy than the 48 PLUs used too. Finally, though the 48, 39 PLUs have the lower recognition accuracy, the 39 PLUs has an average of 1.17% higher recognition characteristic than the 48 PLUs used in the task-independent word recognition experiments according to the unknown contextual factor. Through the above experiments, we verified the effectiveness of the re-defined 39 PLUs compared to the 48PLUs to construct the CD acoustic models in this paper.

• 스마트미디어저널
• /
• 제7권4호
• /
• pp.79-89
• /
• 2018

In recent years, falls among elderly people have gained serious attention as a major cause of injuries. Falls often lead to fatal consequences due to lack of prompt response and rescue. Therefore, a more accurate fall detection system and an effective feature extraction technique are required to prevent and reduce the risk of such incidents. In this paper, we proposed an efficient feature extraction technique based on multiple sliding windows and validated it through a series of experiments using supervised learning algorithms. The experiments were conducted using the public datasets obtained from tri-axial accelerometers. The results depicted that extraction of the feature from adjacent sliding windows led to high accuracy in supervised machine learning-based fall detection. Also, the experiments conducted in this study suggested that the best accuracy can be achieved by keeping the window size as small as 2 seconds. With the kNN classifier and dataset from wearable sensors, the experiments achieved accuracy rates of 94%.

• 한국경영과학회지
• /
• 제32권1호
• /
• pp.61-75
• /
• 2007

The purpose of this paper is to develop a new group classification method based on genetic algorithm and to com-pare its prediction performance with those of existing methods in the area of bond rating. To serve this purpose, we conduct various experiments with pilot and general models. Specifically, we first conduct experiments employing two pilot models : the one searching for the cluster center of each group and the other one searching for both the cluster center and the attribute weights in order to maximize classification accuracy. The results from the pilot experiments show that the performance of the latter in terms of classification accuracy ratio is higher than that of the former which provides the rationale of searching for both the cluster center of each group and the attribute weights to improve classification accuracy. With this lesson in mind, we design two generalized models employing genetic algorithm : the one is to maximize the classification accuracy and the other one is to minimize the total misclassification cost. We compare the performance of these two models with those of existing statistical and artificial intelligent models such as MDA, ANN, and Decision Tree, and conclude that the genetic algorithm-based group classification method that we propose in this paper significantly outperforms the other methods in respect of classification accuracy ratio as well as misclassification cost.

• 한국강구조학회 논문집
• /
• 제11권1호통권38호
• /
• pp.23-31
• /
• 1999

결합부위의 해석을 위하여, 계산 효율과 정확성이 모두 뛰어난 단순모델의 사용이 필요한 경우가 많다. 이 단순모델은 결합부위의 특성을 잘 묘사하는 파라미터들로 구성된다. 이들 파라미터의 값은 실험을 통하여 얻어지게 된다. 따라서 단순모델은 실험에서 사용된 하중하에서는 결합부위의 거동을 잘 묘사하지만, 그외 다른 하중하에서는 결합부위의 거동을 어느 정도 구현할지 알 수 없다. 따라서 단순모델의 정확성을 객관적으로 얻을 수 있는 방법이 필요하게 된다. 본 연구에서는 역최적화 (antioptimization) 개념으로 최악의 하중조건을 정의하여, 이 하중하에서 단순모델의 정확성을 평가하는 방법을 제시하였다. 최악의 하중조건 하에서, 용접으로 체결된 3차원 결합부위의 단순모델과 2차원 구조물에서의 결합부위 단순모델의 정확성을 평가하였다.

• 한국정밀공학회지
• /
• 제24권6호
• /
• pp.45-50
• /
• 2007

The aspheric lens has become the most popular optical component used in various optical devices such as digital cameras, pick-up lenses, printers, copiers etc. Using aspheric lenses not only miniaturizes and reduces the weight of products, but also lower prices and higher field angles can be realized. Additionally, plastic lenses are being changed to glass lenses more recently because of low accuracy, low acid-resistance and low thermal-resistance in the plastic lenses. Currently, one fabrication method of glass lenses is using a glass-mold method with a high precision mold core for mass production. In this paper, DOE (Design Of Experiments) and compensation machining were adopted to improve the surface roughness and the form accuracy of the mold core. The DOE has been done in order to discover the optimal grinding conditions which minimize the surface roughness with factors such as work spindle revolution, turbine spindle revolution, federate and cutting depth. And the compensation machining is used to generate high form accuracy of the mold core. From various experiments and analyses, we could obtain the best surface roughness 5 nm in Ra, form accuracy $0.167\;{\mu}m$ in PV.

• International Journal of Internet, Broadcasting and Communication
• /
• 제13권2호
• /
• pp.173-178
• /
• 2021

In this paper, we find a suitable methodology for Korean Sentiment Analysis through a comparative experiment in which methods of embedding and natural network models are learned at the highest accuracy and fastest speed. The embedding method compares word embeddeding and Word2Vec. The model compares and experiments representative neural network models CNN, RNN, LSTM, GRU, Bi-LSTM and Bi-GRU with IKEA review data. Experiments show that Word2Vec and BiGRU had the highest accuracy and second fastest speed with 94.23% accuracy and 42.30 seconds speed. Word2Vec and GRU were found to have the third highest accuracy and fastest speed with 92.53% accuracy and 26.75 seconds speed.

• ETRI Journal
• /
• 제41권5호
• /
• pp.684-695
• /
• 2019

In a cloud environment, performance degradation, or even downtime, of virtual machines (VMs) usually appears gradually along with anomalous states of VMs. To better characterize the state of a VM, all possible performance metrics are collected. For such high-dimensional datasets, this article proposes a feature extraction algorithm based on unsupervised fuzzy linear discriminant analysis with kernel (UFKLDA). By introducing the kernel method, UFKLDA can not only effectively deal with non-Gaussian datasets but also implement nonlinear feature extraction. Two sets of experiments were undertaken. In discriminability experiments, this article introduces quantitative criteria to measure discriminability among all classes of samples. The results show that UFKLDA improves discriminability compared with other popular feature extraction algorithms. In detection accuracy experiments, this article computes accuracy measures of an anomaly detection algorithm (i.e., C-SVM) on the original performance metrics and extracted features. The results show that anomaly detection with features extracted by UFKLDA improves the accuracy of detection in terms of sensitivity and specificity.

• 산업경영시스템학회지
• /
• 제38권4호
• /
• pp.177-183
• /
• 2015

Incremental sheet metal forming is a manufacturing process to produce thin parts using sheet metals by a series of small incremental deformation. The process rarely needs dedicated dies and molds, thus, preparation time for the process is relatively short as to be compared to conventional metal forming. Spring back in sheet metal working is very common, which causes critical errors in dimensions. Incremental sheet metal forming is not fully investigated yet. Hence, incremental sheet metal forming frequently produces inaccurate parts. This paper proposes a method to minimize dimensional errors to improve shape accuracy of products manufactured by incremental forming. This study conducts experiments using an exclusive incremental forming machine and the material for these experiments are sheets of aluminum AL1015. This research defines a process parameter and selects a few factors for the experiments. The parameters employed in this paper are tool feed rate, tool diameter, step depth, material thickness, forming method, dies applied, and tool path method. In addition, their levels for each factor are determined. The plan of the experiments is designed using orthogonal array $L_8$ ($2^7$) which requires minimum number of experiments. Based on the measurements, dimensional errors are collected both on the tool contacted surfaces and on the non-contacted surfaces. The distances between the formed surfaces and the CAD models are scanned and recorded using a commercial software product. These collected data are statistically analyzed and ANOVAs (analysis of variances) are drawn up. From the ANOVAs, this paper concludes that the process parameters of tool diameter, forming depth, and forming method are the significant factors to reduce the errors on the tool contacted surface. On the other hand, the experimental factors of forming method and dies applied are the significant factors on the non-contacted surface. However, the negative forming method always produces better accuracy than the positive forming method.

• 한국항공우주학회지
• /
• 제42권4호
• /
• pp.291-297
• /
• 2014

회전익 항공기의 기체 공력특성 측정실험에 실험설계 방법론을 적용하여 풍동실험의 정밀도 향상에 관한 연구를 수행하였다. 블록화와 블록화 되지 않은 경우들에 대한 분산분석결과의 비교를 통해 블록화의 영향을 평가하였다. 2차 반응표면모델의 경우 블록화가 실험결과의 정밀도 향상에 실질적인 영향을 주고 있음을 알 수 있었다. 본 연구를 통해 랜덤화, 블록화 및 반복화와 같은 실험설계원리와 같은 풍동실험 절차의 재배치를 통해 이러한 정밀도 향상이 가능함을 보였다.

• 음성과학
• /
• 제12권4호
• /
• pp.31-42
• /
• 2005

Recently, Gaussian Mixture Model (GMM), a special form of CHMM, has been applied to speaker identification and it has proved that performance of GMM is better than CHMM. Therefore, in this paper the speaker models based on GMM and a new GMM using the modified EM algorithm are introduced and evaluated for text-independent speaker identification. Various experiments were performed to evaluate identification performance of two algorithms. As a result of the experiments, the GMM speaker model attained 94.6% identification accuracy using 40 seconds of training data and 32 mixtures and 97.8% accuracy using 80 seconds of training data and 64 mixtures. On the other hand, the new GMM speaker model achieved 95.0% identification accuracy using 40 seconds of training data and 32 mixtures and 98.2% accuracy using 80 seconds of training data and 64 mixtures. It shows that the new GMM speaker identification performance is better than the GMM speaker identification performance.