• Title/Summary/Keyword: 사전등록

Search Result 208, Processing Time 0.027 seconds

Voice Recognition using a Phoneme based Similarity Algorithm in Home Networks (음소 기반의 유사율 알고리즘을 이용한 Home Network 환경에서의 음성 인식)

  • Lee, Chang-Sub;Yu, Jae-Bong;Park, Joon-Seok;Yang, Soo-Ho;Kim, Yu-Seop;Park, Chan-Young
    • Proceedings of the Korea Information Processing Society Conference
    • /
    • 2005.05a
    • /
    • pp.767-770
    • /
    • 2005
  • 네트워크상에서 전달되는 음성데이터는 전달되는 과정에서 잡음 등의 외부 요인으로 인하여 데이터에 손실이 생기는 문제가 발생한다. 이렇게 전달된 음성데이터가 음성 인식기를 통과하면 바로 음성 인식기를 통과했을 때 보다 인식률이 낮아진다. 본 연구에서는 홈 네트워크를 제어하는데 있어서 음성 인식률을 향상시키기 위해서 음성 데이터를 입력받아, 이를 음소단위 기반의 유사율 알고리즘을 적용시켜 이미 구축된 홈 네트워크 용어 관련 사전에 등록된 단어와의 유사성을 검토하여 추출된 결과로 홈 네트워크를 제어하는 방안을 제안한다. 음소단위 기반의 유사율 알고리즘과 다중발화를 이용했을 때 Threshold 값이 85% 일 경우 사전에 구축된 단어와 매칭된 인식률은 100%였으며, 사전에 없는 단어의 오인식률은 2%로 감소되었다.

  • PDF

Improved Pattern Recoginition Coding System of a Handwriting Character with 3D (3D Magnetic Ball을 이용한 필기체 인식 향상 Coding System)

  • Sim, Kyu Seung;Lee, Jae Hong;Lee, Byoung Yup
    • The Journal of the Korea Contents Association
    • /
    • v.13 no.9
    • /
    • pp.10-19
    • /
    • 2013
  • This Paper proposed the development of new magnetic sensor and recognition system to expendite pattern recognition of a handwriting character. Received character graphics should be performed the session and balancing and no extraction of end points, bend points and juntions separately. The Artifical intelligence algorithm is adapted to structure snalysis and recognition process by individual basic letter dictionary except for the handwriing character graphic dictionaryimproving error of recognition algorithm and enomous dictionary for generalization. In this Paper, recognition rate of the received character are compared with pre registered character at letter dictionary for performance test of magnetic ball sensor. As a result of unicode conversion and eomparison, the artificial intelligence study have recognition rate more than 95% at initial recognition rate of 70%.

A Study on the Fast Enrollment of Text-Independent Speaker Verification for Vehicle Security (차량 보안을 위한 어구독립 화자증명의 등록시간 단축에 관한 연구)

  • Lee, Tae-Seung;Choi, Ho-Jin
    • Journal of Advanced Navigation Technology
    • /
    • v.5 no.1
    • /
    • pp.1-10
    • /
    • 2001
  • Speech has a good characteristics of which car drivers busy to concern with miscellaneous operation can make use in convenient handling and manipulating of devices. By utilizing this, this works proposes a speaker verification method for protecting cars from being stolen and identifying a person trying to access critical on-line services. In this, continuant phonemes recognition which uses language information of speech and MLP(mult-layer perceptron) which has some advantages against previous stochastic methods are adopted. The recognition method, though, involves huge computation amount for learning, so it is somewhat difficult to adopt this in speaker verification application in which speakers should enroll themselves at real time. To relieve this problem, this works presents a solution that introduces speaker cohort models from speaker verification score normalization technique established before, dividing background speakers into small cohorts in advance. As a result, this enables computation burden to be reduced through classifying the enrolling speaker into one of those cohorts and going through enrollment for only that cohort.

  • PDF

A Design of File Leakage Response System through Event Detection (이벤트 감지를 통한 파일 유출 대응 시스템 설계)

  • Shin, Seung-Soo
    • Journal of Industrial Convergence
    • /
    • v.20 no.7
    • /
    • pp.65-71
    • /
    • 2022
  • With the development of ICT, as the era of the 4th industrial revolution arrives, the amount of data is enormous, and as big data technologies emerge, technologies for processing, storing, and processing data are becoming important. In this paper, we propose a system that detects events through monitoring and judges them using hash values because the damage to important files in case of leakage in industries and public places is serious nationally and property. As a research method, an optional event method is used to compare the hash value registered in advance after performing the encryption operation in the event of a file leakage, and then determine whether it is an important file. Monitoring of specific events minimizes system load, analyzes the signature, and determines it to improve accuracy. Confidentiality is improved by comparing and determining hash values pre-registered in the database. For future research, research on security solutions to prevent file leakage through networks and various paths is needed.

A Review Essay on Legal Mechanisms for Orbital Slot Allocation (정지궤도슬롯의 법적 배분기제에 관한 논고)

  • Jung, Joon-Sik;Hwang, Ho-Won
    • The Korean Journal of Air & Space Law and Policy
    • /
    • v.29 no.1
    • /
    • pp.199-236
    • /
    • 2014
  • This paper analyses from the perspective of distributive justice the legal mechanisms for international allocation of orbital slots, which are of co-owned nature and thereby limited natural resources in outer space. The allocative function is delegated to the International Telecommunication Union. The Radio Regulation, amongst such other legal instruments as the Constitution and Convention, by which the ITU and contracting States thereof abides, dictates how the orbital positions are distributed. Thus, the RR is thoroughly reviewed in the essay. The mechanisms are in a broad sense categorized into two systems: 'a posteriori system' where the 'first come, first served' principle prevails; and 'a priori system' designed to foster the utilisation of the slots by those who lack space resources and are, in especial, likely to be marginalised under the former system. The argument proceeds on the premise that a posteriori system places the under-resourced States in unfavourable positions in the securement of the slots. In contrast with this notion, seven factors were instantiated for an assertion that the degradation of the distributive justice derived from the 'first come, first served' rule, which lays the foundation for the system, could be either mitigated or counterbalanced by the alleged exceptions to the rule. However, the author of this essay argues for counterevidences against the factors and thereby demonstrating that the principle still remains as an overwhelming doctrine, posing a threat to the pursuit of fair allocation. The elements he set forth are as in the following: 1) that the 'first come, first served' principle only applies to assignments capable of causing harmful interferences; 2) the interoperability of the principle with the 'rule of conformity' with the all the ITU instruments; 3) the viability of alternative registrations, as an exception of the application of the principle, on the condition of provisional and informational purposes; 4) another reference that matters in deciding the priority: the types of services in the TFA; 5) the Rule of Procedure H40 proclaiming a ban on taking advantage of coming first to the Register; 6) the technical factors and equity-oriented norms under international and municipal laws along with; 7) the changes of 'basic characteristics' of registered assignments. The second half of this essay illustrates by examining the relevant Annexes to the Regulation that the planned allocation, i.e., a priori system, bear the structured flaws that hinder the fulfillment of the original purpose of the system. The Broadcasting and Fixed Satellite Systems are the reviewed Plans in which the 'first come, first served' principle re-emerges in the end as a determining factor to grant the 'right to international recognition' to administrations including those who has not the allotted portions in the Plan.

A Study on the Ways of Preparation of Disclosure Document and its Utilisation in Franchising: From a Franchisor Viewpoint (가맹사업에 있어 정보공개서의 작성 및 등록제도의 활용에 관한 연구 : 가맹본부입장에서)

  • Lee, Jae Yang;Kin, Pan Jin
    • The Korean Journal of Franchise Management
    • /
    • v.2 no.2
    • /
    • pp.1-23
    • /
    • 2011
  • The Freedom of Information System has been introduced into the society based on the Fair Trade Transactions Act, which was established by Fair Trade Commission (FTC) on May, 2002. However, the system itself has showed limitations in guaranteeing a reliability and transparency of the disclosure document. Thus, since February, 2008, FTC not only made franchisors to register disclosure documents but also adopted the Disclosure Document Registration System, which forced them to provide registered disclosure documents to franchise applicants and franchisee. Franchisors consider the newly adopted Disclosure Document Registration System a restrictive system. However, considering the recent trend of fast growing franchising industry and the importance of being competitive, franchisors need to utilize the disclosure documents to promote their business and to gain trusts from franchise applicants by providing truthful information. In that way, franchisors will be able to establish a foundation that franchising industry might be successful and reduce the agency fee by cutting out conflicts with franchisees. Thus, this study aims to study the ways of effective preparation of disclosure document and its utilization from a franchisor viewpoint.

Database metadata standardization processing model using web dictionary crawling (웹 사전 크롤링을 이용한 데이터베이스 메타데이터 표준화 처리 모델)

  • Jeong, Hana;Park, Koo-Rack;Chung, Young-suk
    • Journal of Digital Convergence
    • /
    • v.19 no.9
    • /
    • pp.209-215
    • /
    • 2021
  • Data quality management is an important issue these days. Improve data quality by providing consistent metadata. This study presents algorithms that facilitate standard word dictionary management for consistent metadata management. Algorithms are presented to automate synonyms management of database metadata through web dictionary crawling. It also improves the accuracy of the data by resolving homonym distinction issues that may arise during the web dictionary crawling process. The algorithm proposed in this study increases the reliability of metadata data quality compared to the existing passive management. It can also reduce the time spent on registering and managing synonym data. Further research on the new data standardization partial automation model will need to be continued, with a detailed understanding of some of the automatable tasks in future data standardization activities.

A Study on Potential Security Threats that Occurred Possibly by Permission Requests in Android Environment (안드로이드 환경에서 권한 요구를 통해 발생 가능한 잠재적 보안위협에 관한 연구)

  • Park, In-Kyu;Choi, Seul-Ki;Kwak, Jin
    • Proceedings of the Korea Information Processing Society Conference
    • /
    • 2017.04a
    • /
    • pp.258-261
    • /
    • 2017
  • 안드로이드 애플리케이션 개발 시 특정 기능을 수행하기 위해 필수적으로 AndroidManifest.xml 파일에 권한을 등록한다. 등록된 권한은 애플리케이션 설치 시 사용자에게 동의를 요구하며 애플리케이션을 설치하기 위해 해당 애플리케이션이 요구하는 사용 권한 모두를 동의해야 한다. 애플리케이션이 요구하는 권한에 동의하지 않을 경우 애플리케이션 설치가 취소되며, 권한 요구에 대한 동의 없이 애플리케이션을 설치할 수 없다. 이러한 이유로 대부분의 사용자는 애플리케이션이 본래의 기능과 맞지 않는 필요 범위 이상으로 요구하는 권한까지도 상세하게 살펴보지 않고 동의하여 본래의 기능과 관계없는 권한을 통해 개인정보 유출, 비정상적인 과금 유발 등 다양한 보안위협이 발생한다. 본 논문에서는 안드로이드 애플리케이션의 APK 파일을 통해 애플리케이션이 요구하는 권한과 실제 애플리케이션이 기능 제공을 위해 사용하는 API 목록을 확인하여 애플리케이션이 필요 범위 이상으로 요구하는 권한을 분석한다. 또한, 애플리케이션 업데이트 시 API 추가로 인해 발생할 수 있는 보안위협을 분석한다. 이를 통해 필요 범위 이상으로 요구하는 권한으로 발생할 수 있는 보안위협에 대해 사전에 인지하여 이를 방지할 수 있도록 한다.

Expanding the User Authentication Scheme in SIP (SIP에서의 강화된 사용자 인증 방식)

  • Go, Yun-Mi;Kwon, Kyung-Hee
    • The Journal of the Korea Contents Association
    • /
    • v.11 no.12
    • /
    • pp.88-93
    • /
    • 2011
  • Due to vulnerable authentication scheme of SIP, intruders can easily impersonate legitimate user. HTTP Digest authentication scheme or private key issued by trust third parties has been used to prevent impersonation attack. However, these methods have suffered security vulnerability or service delay due to computation overhead. In this paper, we propose new authentication method to generate automatically one-time password using the pre-shared password and time information of messages exchanged between SIP UA(User Agent) and SIP Registrar. This method protects against impersonation attack without significant modification of exiting SIP authentication procedure to build securer SIP environment.

Analysis of each appstores' certification way by code signing and security (앱스토어별 코드사이닝 기법을 이용한 인증 방식과 보안성 분석)

  • Kim, Ji-Hong;Kim, Young-Hun;Lim, Hun-Jung;Lee, Jun-Ho;Chung, Tai-Myoung
    • Proceedings of the Korea Information Processing Society Conference
    • /
    • 2010.11a
    • /
    • pp.1221-1224
    • /
    • 2010
  • 스마트폰 사용자의 폭발적 증가와 함께 모바일 애플리케이션 시장에 대한 관심도 크게 증가하고 있다. 더불어 애플리케이션의 보안 위협 역시 증가하고 있는데 각 모바일 애플리케이션 마켓에서는 이에 대한 대처방안 중 하나로 코드사이닝 기법을 활용하고 있다. 코드사이닝 기법을 이용하여 애플리케이션 및 개발자에 대해 인증을 받는 방식에는 각 모바일 마켓별로 다른 점이 존재한다. 그 종류로는 개발자가 생성한 공개키와 해당 애플리케이션을 애플에 전송하고 애플이 이에 대한 검증을 하여 마지막 서명을 부여하는 방식의 애플 앱스토어, 개발자가 키와 증명서를 이용해 스스로 애플리케이션에 대해 증명서를 발급하는 방식의 구글 안드로이드 마켓, RIM에 등록되어 있는 코드사이닝 키를 이용하여 애플리케이션을 제출하면 자동적으로 서명을 받을 수 있는 블랙베리 앱월드, 사전에 개발자가 자기 증명을 통해 아이디를 발급받고 애플리케이션을 테스트 하우스에 등록하여 검증 및 서명을 받는 노키아오비스토어 등이 있다. 이와 같이 개발자 및 애플리케이션을 인증 받는 방식이 각 마켓별로 차이가 있는데 이에 대한 자세한 분석과 그에 따른 보안 위협에 대한 안정성에 대해 분석해 보겠다.