• Asian Journal of Atmospheric Environment
• /
• v.9 no.4
• /
• pp.259-271
• /
• 2015

This study conducted analyses on biogenic volatile organic compounds (BVOC) emission sources contributing to urban ozone ($O_3$) concentration in Osaka Prefecture, Japan in summer 2010 by using the Weather Research and Forecasting model (WRF) version 3.5.1 and the Community Multiscale Air Quality model (CMAQ) version 5.0.1. This prefecture is characterized by highly urbanized area with small forest area. The contributions of source regions surrounding Osaka were estimated by comparing the baseline case and zero-out cases for BVOC emissions from each source region. The zero-out emission runs showed that the BVOC emissions substantially contributed to urban $O_3$ concentration in Osaka (10.3 ppb: 15.9% of mean daily maximum 1-h $O_3$ concentration) with day-by-day variations of contributing source regions, which were qualitatively explained by backward trajectory analyses. Although $O_3$ concentrations were especially high on 23 July and 2 August 2010, the contribution of BVOC on 23 July (35.4 ppb: 25.6% of daily maximum $O_3$) was much larger than that on 2 August (20.9 ppb: 14.2% of daily maximum $O_3$). To investigate this difference, additional zero-out cases for anthropogenic VOC (AVOC) emissions from Osaka and for VOC emissions on the target days were performed. On 23 July, the urban $O_3$ concentration in Osaka was dominantly increased by the transport from the northwestern region outside Osaka with large contribution of $O_3$ that was produced through BVOC reactions by the day before and was retained over the nocturnal boundary layer. On 2 August, the concentration was dominantly increased by the local photochemical production inside Osaka under weak wind condition with the particularly large contribution of AVOC emitted from Osaka on the day.

• The Journal of the Acoustical Society of Korea
• /
• v.18 no.2
• /
• pp.11-17
• /
• 1999

This paper proposes an observation probability smoothing technique for the robustness of a discrete hidden Markov(DHMM) model based speech recognizer. Also, an appropriate noise robust processing in car environment is suggested from experimental results. The noisy speech is often mislabeled during the vector quantization process. To reduce the effects of such mislabelings, the proposed technique increases the observation probability of similar codewords. For the noise robust processing in car environment, the liftering on the distance measure of feature vectors, the high pass filtering, and the spectral subtraction methods are examined. Recognition experiments on the 14-isolated words consists of the Korean digits and command words were performed. The database was recorded in a stopping car and a running car environments. The recognition rates of the baseline recognizer were 97.4% in a stopping situation and 59.1% in a running situation. Using the proposed observation probability smoothing technique, the liftering, the high pass filtering, and the spectral subtraction the recognition rates were enhanced to 98.3% in a stopping situation and to 88.6% in a running situation.

• The Journal of the Acoustical Society of Korea
• /
• v.39 no.2
• /
• pp.137-142
• /
• 2020

Many studies based on I-vector have been conducted in a variety of environments, from text-dependent short-utterance to text-independent long-utterance. In this paper, we propose a speaker verification system employing a combination of I-vector with Probabilistic Linear Discriminant Analysis (PLDA) and speaker embedding of Long Short Term Memory (LSTM) with attention mechanism in far-field and noisy environments. The LSTM model's Equal Error Rate (EER) is 15.52 % and the Attention-LSTM model is 8.46 %, improving by 7.06 %. We show that the proposed method solves the problem of the existing extraction process which defines embedding as a heuristic. The EER of the I-vector/PLDA without combining is 6.18 % that shows the best performance. And combined with attention-LSTM based embedding is 2.57 % that is 3.61 % less than the baseline system, and which improves performance by 58.41 %.

• Journal of the Korean Society of Propulsion Engineers
• /
• v.19 no.5
• /
• pp.37-45
• /
• 2015

Prediction of combustion instability within a solid-propellant rocket motor has been conducted with the classical acoustic analysis. The effect of particle size distribution on the instability has been analyzed by comparing the log-normal distribution to the fixed mono-sized particle followed by a survey of motor length scale effect between the baseline model and small scale model. Particle damping effect was more efficient for the small scale motor which has a relatively high unstable mode frequencies. It was also revealed that the prediction results by considering the particle size distribution show an overall attenuation of fluctuating pressure amplitude with respect to the mono-sized case.

• The KSFM Journal of Fluid Machinery
• /
• v.17 no.6
• /
• pp.52-58
• /
• 2014

In the present work, design optimization of film-cooling hole array on the pressure side of high pressure turbine nozzle was conducted. There are four rows of fan-shaped film cooling holes on the nozzle pressure side surface and each row has a straight array of holes in the spanwise direction for baseline model. For design optimization, hole distributions in streamwise and spanwise directions for three rows of holes except first row are parameterized as a 2nd-order shape function. Three-dimensional compressible RANS equations are used for flow and thermal analysis around the nozzle surface and optimization technique using Design of Experiment, Kriging surrogate model and Genetic Algorithm is used. The results shows that averaged adiabatic wall temperature at the whole nozzle surface decreases about 2.7% and averaged film cooling effectiveness at the pressure side of nozzle increased about 8.2%.

• Investigative Magnetic Resonance Imaging
• /
• v.20 no.1
• /
• pp.53-60
• /
• 2016

Purpose: To develop a technique for quantifying the $^{13}C$-metabolites by performing frequency-selective hyperpolarized $^{13}C$ magnetic resonance spectroscopy (MRS) in vitro which combines simple spectrally-selective excitation with spectrally interleaved acquisition. Methods: Numerical simulations were performed with varying noise level and $K_p$ values to compare the quantification accuracies of the proposed and the conventional methods. For in vitro experiments, a spectrally-selective excitation scheme was enabled by narrow-band radiofrequency (RF) excitation pulse implemented into a free-induction decay chemical shift imaging (FIDCSI) sequence. Experiments with LDH / NADH enzyme mixture were performed to validate the effectiveness of the proposed acquisition method. Also, a modified two-site exchange model was formulated for metabolism kinetics quantification with the proposed method. Results: From the simulation results, significant increase of the lactate peak signal to noise ratio (PSNR) was observed. Also, the quantified $K_p$ value from the dynamic curves were more accurate in the case of the proposed acquisition method compared to the conventional non-selective excitation scheme. In vitro experiment results were in good agreement with the simulation results, also displaying increased PSNR for lactate. Fitting results using the modified two-site exchange model also showed expected results in agreement with the simulations. Conclusion: A method for accurate quantification of hyperpolarized pyruvate and the downstream product focused on in vitro experiment was described. By using a narrow-band RF excitation pulse with alternating acquisition, different resonances were selectively excited with a different flip angle for increased PSNR while the hyperpolarized magnetization of the substrate can be minimally perturbed with a low flip angle. Baseline signals from neighboring resonances can be effectively suppressed to accurately quantify the metabolism kinetics.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.5
• /
• pp.2122-2140
• /
• 2020

Proper and precise embedding of commercial ads within Webpages requires Ad-hoc analysis and understanding of their content. By the successful implementation of this step, both publishers and advertisers gain mutual benefits through increasing their revenues on the one hand, and improving user experience on the other. In this research work, we propose a novel multi-level context-based ads serving approach through which ads will be served at generic publisher websites based on their contextual relevance. In the proposed approach, knowledge encoded in domain-specific and generic semantic repositories is exploited in order to analyze and segment Webpages into sets of contextually-relevant segments. Semantically-enhanced indexes are also constructed to index ads based on their textual descriptions provided by advertisers. A modified cosine similarity matching algorithm is employed to embed each ad from the Ads repository into one or more contextually-relevant segments. In order to validate our proposal, we have implemented a prototype of an ad serving system with two datasets that consist of (11429 ads and 93 documents) and (11000 documents and 15 ads), respectively. To demonstrate the effectiveness of the proposed techniques, we experimentally tested the proposed method and compared the produced results against five baseline metrics that can be used in the context of ad serving systems. In addition, we compared the results produced by our system with other state-of-the-art models. Findings demonstrate that the accuracy of conventional ad matching techniques has improved by exploiting the proposed semantically-enhanced context-based ad serving model.

• The Journal of the Acoustical Society of Korea
• /
• v.20 no.3
• /
• pp.56-62
• /
• 2001

In this paper, we adopt a missing data theory to speech recognition. It can be used in order to maintain high performance of speech recognizer when the missing data occurs. In general, hidden Markov model (HMM) is used as a stochastic classifier for speech recognition task. Acoustic events are represented by continuous probability density function in continuous density HMM(CDHMM). The missing data theory has an advantage that can be easily applicable to this CDHMM. A marginalization method is used for processing missing data because it has small complexity and is easy to apply to automatic speech recognition (ASR). Also, a spectral subtraction is used for detecting missing data. If the difference between the energy of speech and that of background noise is below given threshold value, we determine that missing has occurred. We propose a new method that examines the reliability of detected missing data using voicing probability. The voicing probability is used to find voiced frames. It is used to process the missing data in voiced region that has more redundant information than consonants. The experimental results showed that our method improves performance than baseline system that uses spectral subtraction method only. In 452 words isolated word recognition experiment, the proposed method using the voicing probability reduced the average word error rate by 12％ in a typical noise situation.

• Journal of computational fluids engineering
• /
• v.21 no.3
• /
• pp.98-109
• /
• 2016

The flow and heat transfer characteristics of a heat exchanger having rectangular pin-fin in the flow direction have been investigated numerically. On the bottom plate, the convective boundary conditions for the hot side was given, and the fins were arranged in a channel-type geometric model using the periodic boundary condition in the span-wise direction. Three-dimensional numerical calculations for the flow and conjugate heat transfer problem were conducted using SIMPLE algorithm and $k-{\varepsilon}$ turbulence model. For the slanted pin-fin models, it was found that the downward cooling flow is generated due to the downward pressure gradient component, which can enhance the heat transfer performance near the bottom surface and the fin stem region. Four different inclined angles were considered in the Reynolds number range of 13,500-55,000. The aero-thermal performance of the slanted pin-fin heat exchangers, such as the volume and area goodness factors, were summarized and compared with the baseline plate-fin type heat exchanger quantitatively.

• Asian Pacific Journal of Cancer Prevention
• /
• v.14 no.5
• /
• pp.3255-3261
• /
• 2013

Background: The efficacy of erlotinib is controversial in patients with unknown EGFR mutational status. The aim of this study was to identify the clinicopathological factors that are predictive of erlotinob treatment outcomes for NSCLC patients with unknown EGFR mutational status. Materials and Methods: A retrospective analysis of 109 patients with advanced NSCLC who had previously failed at least one line of chemotherapy and received subsequent treatment with erlotinib (150 mg/day orally) was performed. A Cox proportional hazard model for univariate and multivariate analyses was used to identify the baseline clinical parameters correlating with treatment outcome, expressed in terms of hazard ratios (HRs) and 95% confidence intervals. Results: The median treatment duration was 15 weeks (range, 4-184). The disease control rate was 55%, including disease stability for ${\geq}3$ months for 40% of the patients. Median progression-free survival and median overall survival (OS) were 4.2 and 8.5 months, respectively. The Cox model indicated that an Eastern Cooperative Oncology Group performance status (ECOG PS) ${\geq}2$ (HR 3.82; p<0.001), presence of intra-abdominal metastasis (HR 3.42; p=0.002), 2 or more prior chemotherapy regimens (HR 2.29; p=0.021), and weight loss >5% (HR 2.05; p=0.034) were independent adverse prognostic factors for OS in NSCLC patients treated with erlotinib. Conclusions: This study suggests that NSCLC patients should be enrolled in erlotinib treatment after a first round of unsuccessful chemotherapy to improve treatment success, during which they should be monitored for intra-abdominal metastasis and weight loss.