• Journal of Advanced Information Technology and Convergence
• /
• v.10 no.1
• /
• pp.45-56
• /
• 2020

Despite extensive research, performance enhancement of keyphrase (KP) extraction remains a challenging problem in modern informatics. Recently, deep learning-based supervised approaches have exhibited state-of-the-art accuracies with respect to this problem, and several of the previously proposed methods utilize Bidirectional Encoder Representations from Transformers (BERT)-based language models. However, few studies have investigated the effective application of BERT-based fine-tuning techniques to the problem of KP extraction. In this paper, we consider the aforementioned problem in the context of scientific articles by investigating the fine-tuning characteristics of two distinct BERT models - BERT (i.e., base BERT model by Google) and SciBERT (i.e., a BERT model trained on scientific text). Three different datasets (WWW, KDD, and Inspec) comprising data obtained from the computer science domain are used to compare the results obtained by fine-tuning BERT and SciBERT in terms of KP extraction.

• Natural Product Sciences
• /
• v.28 no.2
• /
• pp.89-92
• /
• 2022

In Asia, the roots of Paeonia lactiflora and Astragalus membranaceus have been used as therapeutic agents for thousands of years. Once the medicinal plants are harvested, they are dried and their ingredients are extracted by heat-mediated reflux extraction. However, the condensed structure of organic products (especially roots) limits the extraction of bioactive components. In this study, we assessed the effect of the puffing method (using high temperature and pressure) before the extraction process in relation to the profile and antioxidant capacity of active ingredients. We demonstrated that the additional puffing process before extraction methods improves the yield of polyphenol concentrations and antioxidant activities from the roots of P. lactiflora and A. membranaceus.

• Journal of Pharmacopuncture
• /
• v.21 no.2
• /
• pp.48-60
• /
• 2018

Objective: The aim of this study is to optimize the extraction of beneficial substance from Cudrania tricuspidata leaves grown at Jirisan Mountain in South Korea by three different solvents depending on extraction time and at different temperature. Methods: The total phenolic contents were determined by the method reported by $S{\acute{a}}nchez$-Moreno et al. The total flavonoid contents were analyzed by Slinkard and Singleton. The DPPH radical scavenging activity was determined according to the method reported by Blois Results: The extraction yield for each solvent is 9.05-14.1%, 2.17-5.67%, and 2.3-3.9% for D.W., ethanol, and hexane, respectively. The overall results were maximized for the extract obtained with D.W. for 5 min at $100^{\circ}C$. The average phenol contents were 77.11, 45.64, and 0.343 mg/g at $100^{\circ}C$ in water, $78^{\circ}C$ in ethanol, and $68^{\circ}C$ in hexane, respectively. The flavonoid contents were the highest in the materials extracted with D.W., and were increased with increasing temperature, regardless of the extraction solvents, whether water (green), polar organic ethanol, or nonpolar organic hexane. In the ethanol extract, the flavonoid contents are increased gradually from 5.66 mg/g to 7.73 mg/g. The total flavonoid contents were proportional to the concentrations of the water extracts, ranging from 4.14 mg/g to 48.89 mg/g. The antioxidative activities of the water-extracted compounds are generally increased with increasing temperature from 42.5% to 85.5%. Those of the hexane extracts are increased slowly from 3.79% to 8.8%, while those of ethanol extracts are increased from 29.8% to 47.4%. Conclusion: The extraction yields were dependent upon solvents for extraction as well as extraction time and the temperature. The optimal extraction time was 5 min and the extraction yields were increased with increasing temperature excepted hexane. Of the three tested extraction solvents, the greenest solvent of water shows excellent results, suggesting that water is among the most effective solvents for natural sample extractions for general medicinal, pharmaceutical, and food applications.

• Herbal Formula Science
• /
• v.22 no.2
• /
• pp.15-24
• /
• 2014

Objectives : Gwakhyangjeonggi-san (GJS) which consists of 13 herbal medicines has been used to treat gastrointestinal disorders caused by common cold. This study was performed to compare GJS decoctions produced using different pressure levels for various extraction times. Methods : Decoctions were prepared by the pressure levels of $0kgf/cm^2$ (non-pressurized) or $1kgf/cm^2$ (pressurized) for 30-180 min. The extraction yield, total soluble solid content (TSSC), and hydrogen ion concentration (pH) were measured, and the contents of the nine marker compounds were determined using high performance liquid chromatography. Results : The higher pressure and longer extraction time significantly increased TSSC value, while decreased the pH value. However, only extraction time affected the extraction yield of pressurized decoction. Variation of the amounts of chemical compounds was shown in pressurized and non-pressurized decoctions during extraction time. The result of regression analysis showed that pressure and extraction time can influence to extraction yield, TSSC, pH, and the content of chemical compounds. Conclusions : This study suggests that the pressure and extraction time can significantly affect the extraction efficiency of components from GJS decoctions.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2013.05a
• /
• pp.903-904
• /
• 2013

• ETRI Journal
• /
• v.35 no.1
• /
• pp.100-108
• /
• 2013

In this paper, a feature extraction (FE) method is proposed that is comparable to the traditional FE methods used in automatic speech recognition systems. Unlike the conventional spectral-based FE methods, the proposed method evaluates the similarities between an embedded speech signal and a set of predefined speech attractor models in the reconstructed phase space (RPS) domain. In the first step, a set of Gaussian mixture models is trained to represent the speech attractors in the RPS. Next, for a new input speech frame, a posterior-probability-based feature vector is evaluated, which represents the similarity between the embedded frame and the learned speech attractors. We conduct experiments for a speech recognition task utilizing a toolkit based on hidden Markov models, over FARSDAT, a well-known Persian speech corpus. Through the proposed FE method, we gain 3.11% absolute phoneme error rate improvement in comparison to the baseline system, which exploits the mel-frequency cepstral coefficient FE method.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.49 no.12
• /
• pp.651-658
• /
• 2000

This paper presents a new method to extract traffic information such as number of passing vehicles and average speed by a pair of stereo road images. The whole process consists of the extraction of vehicles and the tracking of the extracted vehicles. For the extraction of vehicles, the outline of each vehicle is obtained by using binary region growing technique applied to disparity map based on multi-resolution stereo matching. The Kalman filter tracking algorithm is applied to the extracted vehicle outlines to determine the flow of vehicles. Experimental results show that the proposed method significantly improved recognition rate of vehicles over the conventional methods-frame difference and background elimination methods.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.15 no.9
• /
• pp.5744-5749
• /
• 2014

An extraction technique for a common logic expression is an extremely important part of multiple-output logic synthesis. This paper presents a new Boolean extraction technique using an exclusive-OR logic synthesis principle. The logic circuits produced only have AND, OR and NOT gates. Heuristic methods can also be applied to reduce the execution time and the number of literals. The experimental results showed improvements in the literal counts over the previous methods.

• The Journal of Information Technology
• /
• v.12 no.4
• /
• pp.25-32
• /
• 2009

Iris recognition is a biometric technology which can identify a person using the iris pattern. It is important for the iris recognition system to extract the feature which is invariant to changes in iris patterns. Those changes can be occurred by the influence of lights, changes in the size of the pupil, and head tilting. This paper is appropriate for the embedded environment using local gradient histogram embedded system using iris feature extraction methods have implement. The proposed method enables high-speed feature extraction and feature comparison because it requires no additional processing to obtain the rotation invariance, and shows comparable performance to the well-known previous methods.

• Journal of the Korean Society of Food Science and Nutrition
• /
• v.27 no.4
• /
• pp.687-692
• /
• 1998

To use onion(Allium cepa L.) residue as raw materials of food product, yields of onion-juice according to various extracting methods and optimal extracting condition of quercetin and its related glycosides were carried out. Onion juices gained by the methods of pressing, rotary crushing, freeze pressing and enzyme treatment. The yield by the method of enzyme treatment was higher than others. The yields of juice from fresh onion and heat-treated onion(8$0^{\circ}C$/10min) by pressing were 66% and 83%, respectively. Ethanol extraction of onion was efficient at 75$^{\circ}C$ and acetic acid extraction was proper at 3% concentration for 3 hours at 11$0^{\circ}C$. The onion extract was fractionated in the order of hexane, chloroform, ethylacetate and butanol to test flavonoids. The highest abundant flavonoids were found in ethylacetate and butanol fraction.