• Genomics & Informatics
• /
• v.17 no.2
• /
• pp.15.1-15.7
• /
• 2019

Automatically detecting mentions of pharmaceutical drugs and chemical substances is key for the subsequent extraction of relations of chemicals with other biomedical entities such as genes, proteins, diseases, adverse reactions or symptoms. The identification of drug mentions is also a prior step for complex event types such as drug dosage recognition, duration of medical treatments or drug repurposing. Formally, this task is known as named entity recognition (NER), meaning automatically identifying mentions of predefined entities of interest in running text. In the domain of medical texts, for chemical entity recognition (CER), techniques based on hand-crafted rules and graph-based models can provide adequate performance. In the recent years, the field of natural language processing has mainly pivoted to deep learning and state-of-the-art results for most tasks involving natural language are usually obtained with artificial neural networks. Competitive resources for drug name recognition in English medical texts are already available and heavily used, while for other languages such as Spanish these tools, although clearly needed were missing. In this work, we adapt an existing neural NER system, NeuroNER, to the particular domain of Spanish clinical case texts, and extend the neural network to be able to take into account additional features apart from the plain text. NeuroNER can be considered a competitive baseline system for Spanish drug and CER promoted by the Spanish national plan for the advancement of language technologies (Plan TL).

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.4
• /
• pp.81-86
• /
• 2022

In this work, we propose a method for automatically determining the domain of columns of file data structured by csv format. New data can be generated through convergence between data and data, and the consistency of the joined columns must be maintained in order for these new data to become an important resource. One of the methods for measuring data quality is a domain-based quality diagnosis method. Domain is the broadest indicator that defines the nature of each column, so a method of automatically determining it is necessary. Although previous studies mainly studied domain automatic discrimination of relational databases, this study developed a model that can automate domains using the characteristics of file data. In order to specialize in the domain discrimination of file data, the data were simplified and patterned using a regular expression, and the contents of the data header corresponding to the column name were analyzed, and the suffix used was used as a derived variable. When derivatives of regular expressions and suffixes were added, the result of automatically determining the domain with an accuracy of 95% greater than the existing method of 87% was derived. This study is expected to reduce the quality measurement period and number of people by presenting an automation methodology to the quality diagnosis of public data.

• Journal of KIISE:Information Networking
• /
• v.30 no.4
• /
• pp.559-571
• /
• 2003

ENUM(tElephon NUmbering Mapping) is protocol that brings convergence between PSTN Networks and IP Networks using a unique worldwide E.164 telephone number as an identifier between different communication infrastructure. The mechanism provides a bridge between two completely different environments with E.164 number; IP based application services used in PSTN networks, and PSTN based application services used in IP networks. We propose a new way to organize and handle ENUM Tier 2 name servers to improve performance at the name resolution process in ENUM based application service. We build an ENUM based network model when NAPTR(Naming Authority PoinTeR) resource record is registered and managed by area code at the initial registration step. ENUM promises convenience and flexibility to both PSTN and IP users, yet there is no evidence how much patience is required when users decide to use ENUM instead of non-ENUM based applications. We have estimated ENUM response time, and proved how to improve performance up to 3 times when resources are managed by the proposed mechanism. The proposition of this thesis favorably influences users and helps to establish the policy for Tier 2 name server management.

• Journal of KIISE
• /
• v.43 no.6
• /
• pp.678-685
• /
• 2016

Named Entity Recognition and Classification (NERC) is a task for recognition and classification of named entities such as a person's name, location, and organization. There have been various studies carried out on Korean NERC, but they have some problems, for example lacking some features as compared with English NERC. In this paper, we propose a method that uses word embedding as features for Korean NERC. We generate a word vector using a Continuous-Bag-of-Word (CBOW) model from POS-tagged corpus, and a word cluster symbol using a K-means algorithm from a word vector. We use the word vector and word cluster symbol as word embedding features in Conditional Random Fields (CRFs). From the result of the experiment, performance improved 1.17%, 0.61% and 1.19% respectively for TV domain, Sports domain and IT domain over the baseline system. Showing better performance than other NERC systems, we demonstrate the effectiveness and efficiency of the proposed method.

• Proceedings of the Microbiological Society of Korea Conference
• /
• 2002.10a
• /
• pp.68-72
• /
• 2002

While the biosynthetic gene cluster encoding the pigmented antibiotic actinorhodin is present in the two closely related bacterial species, Streptomyces lividans and Streptomyces coelicolor, it normally is expressed only in S. coelicolor---generating the deep blue colonies responsible for the S. coelicolor name. However, multiple copies of the afsR2 gene, which activates actinorhodin synthesis, result in the ability of S. lividansto also synthesize large amounts of actinorhodin. Here we report that the phenotypic property that historicially distinguishes these two Streptomycesspecies is determined conditionally by the carbon source used for culture. Whereas growth on glucose repressed actinorhodin production in S. lividans, culture on solid media containing glycerol as the sole carbon source dramatically increased the expression of afsR2 mRNA---leading to extensive actinorhodin synthesis by S. lividansand obliterating its phenotypic distinction from S. coelicolor. afsR2 transcription under these conditions was developmentally regulated, rising sharply at the time of aerial mycelium formation and coinciding temporally with the onset of actinorhodin production. Our results, which identify media-dependent parallel pathways that regulate actinorhodin synthesis in S. lividans, demonstrate carbon source control of actinorhodin production through the regulation of afsR2 mRNA synthesis. The nucleotide sequences of afsR2 revealed two putative important domains; the domain containing direct repeats in the middle and the domain homologous to sigma factor sequence in the C-terminal end. In this work, we constructed various sized afsR2-derivatives and compared the actinorhodin stimulating effects in S. lividans TK21. The experimental data indicate that the domain homologous to sigma factor sequence in the C-terminal end of afsR2 plays a critical role as an antibiotic stimulating function. In addition, we also observed that the single copy integration of afsR2 regulatory gene into S. lividans TK21 chromosome significantly activates antibiotic overproduction.

• The KIPS Transactions:PartB
• /
• v.12B no.4 s.100
• /
• pp.479-486
• /
• 2005

The ontology was proposed to construct the logical basis of semantic web. Ontology represents domain knowledge in the formal form and it enables that machine understand domain knowledge and provide appropriate intelligent service for user request. However, the construction and the maintenance of ontology requires large amount of cost and human efforts. This paper proposes an automatic ontology construction method for defining relation between concepts in the documents. The Proposed method works as following steps. First we find concept pairs which compose association rule based on the concepts in domain specific documents. Next, we find pattern that describes the relation between concepts by clustering the context between two concepts composing association rule. Last, find generalized pattern name by clustering the clustered patterns. To verify the proposed method, we extract relation between concepts and evaluate the result using documents set provide by TREC(Text Retrieval Conference). The result shows that proposed method cant provide useful information that describes relation between concepts.

• Proceedings of the Korean Information Science Society Conference
• /
• 1998.10a
• /
• pp.323-325
• /
• 1998

인터넷에서 호스트의 인터넷 주소와 이름을 상호 매핑해주는 도메인 네임 시스템은 호스트수가 급증함에 따라 그 중요성 또한 증대되고 있다. 최근 인증 기능을 제공하는 도메인 네임 시스템이 제안됨에 따라 이를 이용하여 진보된 서비스를 제공하려는 연구가 활발히 진행중이다. 한편 도메인 네임 서버를 기존 네트워크 관리 체계 내에서 관리하기 위한 관리 정보개체는 이미 정의되어 있었다. 하지만 인증 기능을 제공하는 도메인 네임 서버 관리 정보 개체에 대한 연구는 진행되지 않고 있다. 본 논문에서는 인증 기능을 제공하는 도메인 네임 시스템을 관리하기 위한 관리 정보를 정의하고 이를 구현하기 위한 안전한 도메인 네임 서버관리모델을 제시한다.

• Korean Management Science Review
• /
• v.11 no.3
• /
• pp.87-102
• /
• 1994

Conceptual database design is an important activity because the performance of a DBMS heavily depends on the proper design of the database, and view integration is one of the most important phases of conceptual database design. During this phase, the individual views designed by separate user groups are integrated into a global view for the entire system. After the view integration, a database schema must be validated before it becomes a stable product of the design. In this paper, a methodology for view integration an validation is suggested. The individual views are modeled by entity-relationship model. The entities are integrated according to their name and domain information. After that, the relationships are merged by their constraints. We also present a methodology for checking redundant relationships which has an algorithmic description.

• Proceedings of the Korean Information Science Society Conference
• /
• 2003.04c
• /
• pp.256-258
• /
• 2003

정보화 시대에는 사람들의 모든 활동이 인터넷을 통해서 대부분 이루어진다. 이중에서 전자 메일이 차지하는 비중은 매우 크다. 고객 유치를 위한 기업들의 광고와 배움을 위한 강의, 자신의 관심 분야에 대한 정보 등을 전자 매일로 받아보게 되는 것이 더 많아 질것이다. 이러한 상황에서 사람들은 자신이 필요로 하는 메일과 필요로 하지 않는 메일을 분류하는데 많은 시간을 낭비한다. 사람들은 이러한 시간 낭비를 줄이기 위해서 메일 분류 시스템을 사용한다. 현재 사용되고 있는 매일 분류 시스템은 스팸 매일을 기준으로 하고 있다. 그러나 오분류되는 메일들이 있어 사용자가 스팸 메일을 다시 보는 경우가 있어 한계를 보인다. 본 논문에서는 사람들이 자신이 원하는 메일과 그렇지 않은 메일을 분류하기 위해서 1차 분류로 긍정어와 부정어를 이용하여 전자 메일을 분류하고 2차 분류로 도메인 네임을 이용하여 분류한다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.10c
• /
• pp.256-258
• /
• 2001

이동 환경에서의 서비스가 대중화 되면서 문제가 되고 있는 것이 바로 이동 호스트의 전력 제어 문제이다. 이동 호스트는 충분한 전력을 얻기 힘들기 때문에 데이터 전송을 받지 않는 경우에는 최소한의 정보만을 교환하는 상태로의 전의가 필수적이다. 현재 IETF의 Seamoby Working Group에서 논의되고 있는 IP 페이징 프로토콜의 경우에는 요구사항과 논리적인 구조만이 정의되어 있다. 현재 정의되어 있는 구조의 문제점은 유휴 (formant) 모드에 있는 이동 호스트에 데이터를 전송하기 위해서는 Dormant Monitoring Agent(DMA)로 전송된 데이터에 대하여 버퍼링이 수행되어야만 한다. 버퍼링의 목적은 이동 호스트의 유휴 모드에서 활성 (active) 모드로 전의 되는 시간 동안의 데이터의 손실을 방지하기 위함이다. 따라서, 하나의 DMA해서 관리해야 하는 이동 호스트가 많은 경우에는 대회의 양이 문제가 된다. 즉, 확장성 문제점이 존재하게 된다. 본 논문은 이러한 문제점을 줄이기 위하여 Domain Name Server(DNS)를 이용한 IP 페이징 제공 방안을 제안한다. 본 메커니즘은 송신자가 DNS에 호스트의 IP주소를 얻고자 하는 경우 IP 페이징 프로토콜을 먼저 이동 호스트의 페이징 에이전트에 전송함으로써 미리 이동 호스트를 활성 모드로 변경하게 함으로써 이전 메커니즘에서 적용되어야 하는 버퍼의 크기를 줄일 수 있는 메커니즘이다.