• Journal of Broadcast Engineering
• /
• v.24 no.1
• /
• pp.58-66
• /
• 2019

Recognition of named entity such as proper nouns in conversation sentences is the most fundamental and important field of study for efficient conversational information prediction. The most important part of a task-oriented dialogue system is to recognize what attributes an object in a conversation has. The named entity recognition model carries out recognition of the named entity through the preprocessing, word embedding, and prediction steps for the dialogue sentence. This study aims at using user - defined dictionary in preprocessing stage and finding optimal parameters at word embedding stage for efficient dialogue information prediction. In order to test the designed object name recognition model, we selected the field of daily chemical products and constructed the named entity recognition model that can be applied in the task-oriented dialogue system in the related domain.

• Journal of Digital Convergence
• /
• v.17 no.11
• /
• pp.233-240
• /
• 2019

In order to predict and respond to cyber attacks, a number of security companies quickly identify the methods, types and characteristics of attack techniques and are publishing Security Intelligence Reports(SIRs) on them. However, the SIRs distributed by each company are huge and unstructured. In this paper, we propose a framework that uses five analytic techniques to formulate a report and extract key information in order to reduce the time required to extract information on large unstructured SIRs efficiently. Since the SIRs data do not have the correct answer label, we propose four analysis techniques, Keyword Extraction, Topic Modeling, Summarization, and Document Similarity, through Unsupervised Learning. Finally, has built the data to extract threat information from SIRs, analysis applies to the Named Entity Recognition (NER) technology to recognize the words belonging to the IP, Domain/URL, Hash, Malware and determine if the word belongs to which type We propose a framework that applies a total of five analysis techniques, including technology.

• Journal of the Korea Convergence Society
• /
• v.13 no.2
• /
• pp.13-20
• /
• 2022

This study is to develop a named entity recognition model specialized in criminal investigation domains using deep learning techniques. Through this study, we propose a system that can contribute to analysis of crime for prevention and investigation using data analysis techniques in the future by automatically extracting and categorizing crime-related information from text-based data such as criminal judgments and investigation documents. For this study, the criminal investigation domain text was collected and the required entity name was newly defined from the perspective of criminal analysis. In addition, the proposed model applying KoELECTRA, a pre-trained language model that has recently shown high performance in natural language processing, shows performance of micro average(referred to as micro avg) F1-score 98% and macro average(referred to as macro avg) F1-score 95% in 9 main categories of crime domain NER experiment data, and micro avg F1-score 98% and macro avg F1-score 62% in 56 sub categories. The proposed model is analyzed from the perspective of future improvement and utilization.

• Journal of the Korean Society of Food Science and Nutrition
• /
• v.45 no.12
• /
• pp.1769-1775
• /
• 2016

Aerial parts (leaves and stems) of radish are usually discarded due to the distinct undesirable flavors associated with inappropriate preparations, despite their many health benefits. In this study, we examined the role of extrusion process in the removal of off-flavors and elevation of antioxidant activity in radish (Raphanus sativus L.) leaves and stems. To optimize the extrusion conditions, we changed the barrel temperature (110, 120, and $130^{\circ}C$), screw speed (150, 200, 250, and 300 rpm), and moisture content (20, 25, and 30%). The polyphenol and flavonoid contents significantly increased in extruded radish leaves and stems (ER) under optimum extrusion conditions ($130^{\circ}C$, 250 rpm, and 20%). Under extrusion conditions, we compared off-flavors (as amount of sulfur-containing compound) levels between ER and non-extruded radish leaves and stems (NER) by an electronic nose. A total of six peaks (sulfur-containing compound) were similarly detected in both ER and NER, whereas the ER showed reduced off-flavors. Levels of glucosinolate (${\mu}g/g$), which can be hydrolyzed into off-flavors during mastication or processing, were significantly decreased in the ER. From these results, extrusion processing can be an effective method to increase anti-oxidant activity and removal of off-flavors in radish leaves and stems.

• Journal of Intelligence and Information Systems
• /
• v.24 no.1
• /
• pp.253-266
• /
• 2018

This paper proposes a methodology applying sequence tagging methodology to improve the performance of NER(Named Entity Recognition) used in QA system. In order to retrieve the correct answers stored in the database, it is necessary to switch the user's query into a language of the database such as SQL(Structured Query Language). Then, the computer can recognize the language of the user. This is the process of identifying the class or data name contained in the database. The method of retrieving the words contained in the query in the existing database and recognizing the object does not identify the homophone and the word phrases because it does not consider the context of the user's query. If there are multiple search results, all of them are returned as a result, so there can be many interpretations on the query and the time complexity for the calculation becomes large. To overcome these, this study aims to solve this problem by reflecting the contextual meaning of the query using Bidirectional LSTM-CRF. Also we tried to solve the disadvantages of the neural network model which can't identify the untrained words by using ontology knowledge based feature. Experiments were conducted on the ontology knowledge base of music domain and the performance was evaluated. In order to accurately evaluate the performance of the L-Bidirectional LSTM-CRF proposed in this study, we experimented with converting the words included in the learned query into untrained words in order to test whether the words were included in the database but correctly identified the untrained words. As a result, it was possible to recognize objects considering the context and can recognize the untrained words without re-training the L-Bidirectional LSTM-CRF mode, and it is confirmed that the performance of the object recognition as a whole is improved.

• Journal of the Korean Association of Oral and Maxillofacial Surgeons
• /
• v.34 no.5
• /
• pp.509-517
• /
• 2008

DNA damage accumulates in cells as a result of exposure to exogenous agents such as benzopyrene, cigarette smoke, ultraviolet light, X-ray, and endogenous chemicals including reactive oxygen species produced from normal metabolic byproducts. DNA damage can also occur during aberrant DNA processing reactions such as DNA replication, recombination, and repair. The major of DNA damage affects the primary structure of the double helix; that is, the bases are chemically modified. These modification can disrupt the molecules'regular helical structure by introducing non-native chemical bonds or bulky adducts that do not fit in the standard double helix. DNA repair genes and proteins scan the global genome to detect and remove DNA damage and damage to single nucleotides. Direct reversal of DNA damage, base excision repair, double strand break. DNA repair are known relevant DNA repair mechanisms. Four different mechanisms are distinguished within excision repair: direct reversal, base excision repair, nucleotide excision repair, and mismatch repair. Genetic variation in DNA repair genes can modulate DNA repair capacity and alter cancer risk. The instability of a cell to properly regulate its proliferation in the presence of DNA damage increase risk of gene mutation and carcinogenesis. This article aimed to review mechanism of excision repair and to understand the relationship between genetic variation of excision repair genes and head and neck cancer.

• Asian Pacific Journal of Cancer Prevention
• /
• v.15 no.21
• /
• pp.9347-9353
• /
• 2014

Background: Excision repair crossing-complementing group 2 (ERCC2), also called xeroderma pigmentosum complementary group D (XPD), plays a crucial role in the nucleotide excision repair (NER) pathway. Previous epidemiological studies have reported associations between ERCC2 polymorphisms and non-Hodgkin lymphoma (NHL) risk, but the results have remained controversial. Materials and Methods: We conducted this meta-analysis based on eligible case-control studies to investigate the role of two ERCC2 polymorphisms (Lys751Gln and Asp312Asn) in determining susceptibility to NHL. Ten case-control studies from several electronic databases were included in our study up to August 14, 2014. Pooled odds ratios (ORs) and 95% confidence intervals (CIs) were calculated using fixed- or random-effects models to estimate the association strength. Results: The combined results based on all studies did not show any association between Lys751Gln/Asp312Asn polymorphisms and NHL risk for all genetic models. Stratified analyses by histological subtype and ethnicity did not indicate any significant association between Lys751Gln polymorphism and NHL risk. However, a significant reduced risk of NHL was found among population-based studies (Lys/Gln versus Lys/Lys: OR=0.87, 95% CI=0.77-0.99, P=0.037) but not hospital-based studies. As for Asp312Asn polymorphism, there was no evidence for the association between this polymorphism and the risk of NHL in all subgroup analyses. Conclusions: This meta-analysis suggests that there may be no association between Lys751Gln/Asp312Asn polymorphism and the risk of NHL and its two subtypes, whereas ERCC2 Lys751Gln heterozygote genotype may provide protective effects against the risk of NHL in population-based studies. Therefore, large-scale and well-designed studies are needed to clarify the effects of haplotypes, gene-gene, and gene-environment interactions on these polymorphisms and the risk of NHL and its different histological subtypes in an ethnicity specific population.

• Korean Chemical Engineering Research
• /
• v.44 no.5
• /
• pp.498-504
• /
• 2006

The etch rates of photoresist (PR) and the etch selectivity of $SiO_2$ to PR in a high density $CHF_3$ plasma were investigated at different ion-incident angles and bias voltages. A Faraday cage was employed for the accurate control of ion-incident angles. The ion energy was controlled by changing bias voltages. The etch rate of $SiO_2$ continuously decreased with ion-incident angles but the etch rate of PR remained constant up to the middle angle region and decreased afterwards. The etch rates of $SiO_2$ normalized to those at $0^{\circ}$ incident angle changed with the ion-incident angle following a cosine(${\theta}$) curve. On the other hand, the normalized etch rates of the PR changed showing a drastic over-cosine shape in the middle angle region. The etch selectivity of $SiO_2$ to PR decreased with an increase in the ion-incident angle because the etch yields of PR were enhanced by physical sputtering in the middle angle region compared to the case of $SiO_2$ etching. The etch selectivity of $SiO_2$ to PR decreased with an increase in the bias voltage at nearly all ion-incident angles.

• Applied Biological Chemistry
• /
• v.38 no.2
• /
• pp.111-117
• /
• 1995

In Escherichia coli, CRP forms a complex with cAMP and acts as a transcriptional regulator of many genes, including sugar metabolism operons. The E. coli MK2001, which is introduced the altered crp, is functional in the expression of lac, ara and man, in the absence of cAMP. However, the expression of mal gene is fully activated by the addition of cAMP or cGMP. The object of the study is cloning of the sfs (sugar fermentation stimulation) genes, which was involved in regulation of mal gene expression with the altered crp gene, and structural analysis and characterization of the genes at the molecular level. We have cloned 5 different E. coli genes which stimulate the maltose metabolism in a crp, cya::km (MK2001) background. Newly identified genes were designated as sfs. One of the sfs genes (pPC1), located at the 53.2 min map position on the E. coli chromosome, was further analyzed. Expression of the genes, which is involved in maltose metabolism, malQ (amylomaltase), was increased to 5.8-fold in the presence of a plasmid, pAP5, containing the subcloned sfs4 gene. The nucleotide seguence of a common 2,126 bp segment of the pPCM1 was determined and two open reading frames (ORF1 and ORF2) were detected. The ORF1 encodes the sfs4 gene and ORF2 encodes a truncated protein. Potential CRP binding site is located in the upstream of the putative promoter in the regulatory region. Expression of the cloned sfs4 gene was positively regulated by the cAMP-CRP complex.

• Journal of the Korean Institute of Gas
• /
• v.2 no.3
• /
• pp.37-48
• /
• 1998

A cryogenic liquid stored in the closed cryogenic tank has been studied at various liquid levels. The change of pressure, temperature, and liquid-vapor ratio in the tank depended on the liquid levels. The various phenomena were shown at different liquid levels as follows: (1) liquid level was increased with condensation of vapor: (2) liquid was vaporized in spite of liquid level going up for a certain initial period and then condensation of vapor occurred at higher pressure; (3) liquid was vaporized without liquid level change; (4) liquid was vaporized with liquid level decreasing. If the tank is full with cryogenic liquid, it is extremely dangerous because of soaring the pressure. Therefore the tank must be filled with $90\%$ liquid according to the safety rules. If the tank was filled with $0\%$ ullage, the pressure increment as high as 80bar during first 5 days. With $90\%$ liquid level, however, the pressure was increased as low as 1.5bar in the same period. No matter what the liquid level is, it is very dangerous if the tank is locked-up with filled cryogenic liquid for a long time.