• Journal of Intelligence and Information Systems
• /
• v.21 no.2
• /
• pp.131-150
• /
• 2015

Omission of noun phrases for obligatory cases is a common phenomenon in sentences of Korean and Japanese, which is not observed in English. When an argument of a predicate can be filled with a noun phrase co-referential with the title, the argument is more easily omitted in Encyclopedia texts. The omitted noun phrase is called a zero anaphor or zero pronoun. Encyclopedias like Wikipedia are major source for information extraction by intelligent application systems such as information retrieval and question answering systems. However, omission of noun phrases makes the quality of information extraction poor. This paper deals with the problem of developing a system that can restore omitted noun phrases in encyclopedia documents. The problem that our system deals with is almost similar to zero anaphora resolution which is one of the important problems in natural language processing. A noun phrase existing in the text that can be used for restoration is called an antecedent. An antecedent must be co-referential with the zero anaphor. While the candidates for the antecedent are only noun phrases in the same text in case of zero anaphora resolution, the title is also a candidate in our problem. In our system, the first stage is in charge of detecting the zero anaphor. In the second stage, antecedent search is carried out by considering the candidates. If antecedent search fails, an attempt made, in the third stage, to use the title as the antecedent. The main characteristic of our system is to make use of a structural SVM for finding the antecedent. The noun phrases in the text that appear before the position of zero anaphor comprise the search space. The main technique used in the methods proposed in previous research works is to perform binary classification for all the noun phrases in the search space. The noun phrase classified to be an antecedent with highest confidence is selected as the antecedent. However, we propose in this paper that antecedent search is viewed as the problem of assigning the antecedent indicator labels to a sequence of noun phrases. In other words, sequence labeling is employed in antecedent search in the text. We are the first to suggest this idea. To perform sequence labeling, we suggest to use a structural SVM which receives a sequence of noun phrases as input and returns the sequence of labels as output. An output label takes one of two values: one indicating that the corresponding noun phrase is the antecedent and the other indicating that it is not. The structural SVM we used is based on the modified Pegasos algorithm which exploits a subgradient descent methodology used for optimization problems. To train and test our system we selected a set of Wikipedia texts and constructed the annotated corpus in which gold-standard answers are provided such as zero anaphors and their possible antecedents. Training examples are prepared using the annotated corpus and used to train the SVMs and test the system. For zero anaphor detection, sentences are parsed by a syntactic analyzer and subject or object cases omitted are identified. Thus performance of our system is dependent on that of the syntactic analyzer, which is a limitation of our system. When an antecedent is not found in the text, our system tries to use the title to restore the zero anaphor. This is based on binary classification using the regular SVM. The experiment showed that our system's performance is F1 = 68.58%. This means that state-of-the-art system can be developed with our technique. It is expected that future work that enables the system to utilize semantic information can lead to a significant performance improvement.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.18 no.9
• /
• pp.473-479
• /
• 2017

Due to environmental pollution, hazards of the human body, and global warning, changes in the power train of automobiles are intensifying, and the market forelectronic vehicles is rising. Also, in order to meet the stricter emission regulations forautomobiles with internal combustion engines based on fossil fuel, the proportion of after-treatments for vehicles and vessels is increasing gradually. The objective of this study is to investigate the effectsfrom additive ceric oxide ($CeO_2$) loading amounts to improve the methane ($CH_4$) and nitric oxide (NOx) abatement ability of the natural gas oxidation catalysts(NGOC) reducing toxic gases emitted from compressed natural gas (CNG) buses. Three kinds of NGOC were prepared under the following conditions: fresh and $700^{\circ}C$ for 12hr thermal aging, and the reduction performance of toxic gases was evaluated. Fresh $1Pt-3Pd-1Rh-3MgO-6CeO_2/(Al+Z)$ NGOC containing 6wt% $CeO_2$ had the highest dispersivity of palladium (Pd) with high selectivity to $CH_4$ and improved harmful gas reduction performance. The NGOC with 6wt% $CeO_2$ loaded the least decreased in the dispersivity of the noble metal, and showed the highest reduction of harmful gases due to the thermal durability of $CeO_2$.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.17 no.7
• /
• pp.740-748
• /
• 2016

IT and communication technologies has contributed significantly to the convenience of passengers and the financial management of stations in accordance with the task automation in the field of the urban railway system. The foundation of the above development is based on the large amounts of data from various sensors installed in railways, trains, and stations. In particular, the sensor network that is installed in the station and train has played an important role in the railway information system. The performance of AP is affected by the number of APs and their locations installed in the station. In the installation of APs in stations, the intensity of the radio wave of the AP on its underlying position is considered to determine the number and position of APs. This paper proposes a method to estimate the number of APs and their position based on the structure of the underlying station and implemented a simulator to simulate the performance of the proposed method. The implemented simulator was applied to the decision of AP installation at Busan Seomyeon station to evaluate its performance.

• Journal of Intelligence and Information Systems
• /
• v.10 no.1
• /
• pp.109-123
• /
• 2004

Corporate financial distress and bankruptcy prediction is one of the major application areas of artificial neural networks (ANNs) in finance and management. ANNs have showed high prediction performance in this area, but sometimes are confronted with inconsistent and unpredictable performance for noisy data. In addition, it may not be possible to train ANN or the training task cannot be effectively carried out without data reduction when the amount of data is so large because training the large data set needs much processing time and additional costs of collecting data. Instance selection is one of popular methods for dimensionality reduction and is directly related to data reduction. Although some researchers have addressed the need for instance selection in instance-based learning algorithms, there is little research on instance selection for ANN. This study proposes a genetic algorithm (GA) approach to instance selection in ANN for bankruptcy prediction. In this study, we use ANN supported by the GA to optimize the connection weights between layers and select relevant instances. It is expected that the globally evolved weights mitigate the well-known limitations of gradient descent algorithm of backpropagation algorithm. In addition, genetically selected instances will shorten the learning time and enhance prediction performance. This study will compare the proposed model with other major data mining techniques. Experimental results show that the GA approach is a promising method for instance selection in ANN.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.49 no.3
• /
• pp.38-46
• /
• 2012

Atrial fibrillation is needed to be detected at paroxysmal stage and to be treated. But, paroxysmal atrial fibrillation ECG is hardly obtained with 12-lead electrocardiographs but Holter systems. Presently, the averaged beat subtraction(ABS) method is solely used to estimate atrial fibrillatory waves even with somewhat large residual error. As an alternative, in this study, we suggested an ESAF(event-synchronous adaptive filter) based algorithm, in which the AF ECG was treated as a primary input and event-synchronous impulse train(ESIT) as a reference. And, ESIT was generated so to be synchronized with the ventricular activity by detecting QRS complex. We tested proposed algorithm with simulated AF ECGs and real AF ECGs. As results, even with low computational cost, this ESAF based algorithm showed better performance than the ABS method and comparable performance to algorithm based on PCA(principal component analysis) or SVD(singular value decomposition). We also proposed an expanded version of ESAF for some AF ECGs with multi-morphologic ventricular activities and this also showed reasonable performance. Ultimately, with Holter systems including our proposed algorithm, atrial activity signal can be precisely estimated in real-time so that it will be possible to calculate atrial fibrillatory rate and to evaluate the effect of anti-arrhythmic drugs.

• Journal of the Institute of Convergence Signal Processing
• /
• v.11 no.2
• /
• pp.137-142
• /
• 2010

Although multi-model speech recognizer has been shown to be quite successful in noisy speech recognition, the results were based on general speech front-ends which do not take into account noise adaptation techniques. In this paper, for the accurate evaluation of the multi-model based speech recognizer, we adopted a quite noise-robust speech front-end, AFE, which was proposed by the ETSI for the noisy DSR environment. For the performance comparison, the MTR which is known to give good results in the DSR environment has been used. Also, we modified the structure of the multi-model based speech recognizer to improve the recognition performance. N reference HMMs which are most similar to the input noisy speech are used as the acoustic models for recognition to cope with the errors in the selection of the reference HMMs and the noise signal variability. In addition, multiple SNR levels are used to train each of the reference HMMs to improve the robustness of the acoustic models. From the experimental results on the Aurora 2 databases, we could see better recognition rates using the modified multi-model based speech recognizer compared with the previous method.

• Therapeutic Science for Rehabilitation
• /
• v.10 no.2
• /
• pp.95-108
• /
• 2021

Objective : This study aimed to develop prototypes of "smart stacking cones" by combining ICT to evaluate and train the upper extremity function and dual task performance in patients with central nervous system impairment, and to identify the complementary point to the completion of the device through a usability test. Methods : This prototype comprised of a hardware and software system that enabled the evaluation and training of patients and the management of data obtained from patients' performance. Specific measurement variables were established so that patient performance could be measured correctly. Based on the measurement variables, a the prototype included a 'single task evaluation modes', 'dual task evaluation mode', 'single task training mode', and 'dual task training modes'. Additionally, a usability test was conducted to assess clinical applicability and overall satisfaction for the prototype. Results : The results of the usability test were generally found to be appropriate. The 'content adequacy' in the usability test was the area with the highest level of adequacy and the lowest level of inadequacy. Additionally, overall 'satisfaction' in the usability test was the area with the highest appropriate and inappropriate levels. Hence, the overall satisfaction results were unstable. Conclusion : Future studies should be conducted to identify the clinical effectiveness of the device by applying an upgraded smart stacking cone to an actual patient group.

• The Journal of the Acoustical Society of Korea
• /
• v.40 no.2
• /
• pp.139-147
• /
• 2021

This paper introduces and implements a Sound Event Detection (SED) system based on weakly-supervised learning where only part of the data is labeled, and analyzes the effect of parameters. The SED system estimates the classes and onset/offset times of events in the acoustic signal. In order to train the model, all information on the event class and onset/offset times must be provided. Unfortunately, the onset/offset times are hard to be labeled exactly. Therefore, in the weakly-supervised task, the SED model is trained by "strongly labeled data" including the event class and activations, "weakly labeled data" including the event class, and "unlabeled data" without any label. Recently, the SED systems using the mean-teacher model are widely used for the task with several parameters. These parameters should be chosen carefully because they may affect the performance. In this paper, performance analysis was performed on parameters, such as the feature, moving average parameter, weight of the consistency cost function, ramp-up length, and maximum learning rate, using the data of DCASE 2020 Task 4. Effects and the optimal values of the parameters were discussed.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.20 no.6
• /
• pp.157-165
• /
• 2020

Aiming at the development of neural network and self-driving data set, it is also an idea to improve the performance of network model to detect moving objects by dividing the data set. In Darknet network framework, the YOLOv4 (You Only Look Once v4) network model was used to train and test Udacity data set. According to 7 proportions of the Udacity data set, it was divided into three subsets including training set, validation set and test set. K-means++ algorithm was used to conduct dimensional clustering of object boxes in 7 groups. By adjusting the super parameters of YOLOv4 network for training, Optimal model parameters for 7 groups were obtained respectively. These model parameters were used to detect and compare 7 test sets respectively. The experimental results showed that YOLOv4 can effectively detect the large, medium and small moving objects represented by Truck, Car and Pedestrian in the Udacity data set. When the ratio of training set, validation set and test set is 7:1.5:1.5, the optimal model parameters of the YOLOv4 have highest detection performance. The values show mAP50 reaching 80.89%, mAP75 reaching 47.08%, and the detection speed reaching 10.56 FPS.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.9
• /
• pp.1172-1179
• /
• 2020

With the establishment of the railway integrated radio network (LTE-R) environment, radio-based train control transmission and reception and various forms of service are provided. The smooth delivery of these services requires improved performance in a highly reliable and available wireless environment. This paper measured the LTE-R radio communication environment to improve radio communication performance of railway integrated wireless network reliability and availability, analyzed the results, and established the wireless environment model. Based on the built-up model, we also proposed an improved radio-access algorithm to control trains for improved reliability, suggesting a way to improve stability for handover that occur during open-air operation, and proposed an algorithm for frequency auto-heating to improve availability. For simulation, data were collected from the Korea Rail Network Authority (Daejeon), Manjong-Gangneung KTX route, which can measure the actual data of LTE-R wireless environment, and the results of the simulation show performance improvement through algorithm.