• Journal of KIISE
• /
• v.42 no.12
• /
• pp.1561-1567
• /
• 2015

Abstract Semi-supervised learning is an area in machine learning that employs both labeled and unlabeled data in order to train a model and has the potential to improve prediction performance compared to supervised learning. Graph-based semi-supervised learning has recently come into focus with two phases: graph construction, which converts the input data into a graph, and label inference, which predicts the appropriate labels for unlabeled data using the constructed graph. The inference is based on the smoothness assumption feature of semi-supervised learning. In this study, we propose an enhanced label inference algorithm by incorporating the importance of each vertex. In addition, we prove the convergence of the suggested algorithm and verify its excellence.

• JSTS:Journal of Semiconductor Technology and Science
• /
• v.7 no.3
• /
• pp.151-160
• /
• 2007

We report the label-free biomolecules detection based on nanomechanical micro cantilevers operated in dynamic mode for detection of two marker proteins (myoglobin and creatin kinase-MB (CK-MB)) of acute myocardical infarctions. When the specific binding between the antigen and its antibody occurred on the fuctionalized microcantilever surface, mechanical response (i.e. resonant frequency) of microcantilevers was changed in lower frequency range. We performed the label-free biomolecules detection of myoglobin and CK-MB antigen in the low concentration (clinical threshold concentration range) as much as 1 ng/ml from measuring the dynamic response change of micro cantilevers caused by the intermolecular force. Moreover, we estimate the surface stress on the dynamic microcantilevers generated by specific antibody-antigen binding. It is suggested that our dynamic microcantilevers may enable one to use the sensitive label-free biomolecules detection for application to the disease diagnosis system based on mechanical immuno-sensor.

• Journal of Sensor Science and Technology
• /
• v.19 no.6
• /
• pp.403-420
• /
• 2010

Nano and micro structure-based biosensors are promising tool for label-free detection of biomolecular interactions with great accuracy. This review gives a brief survey on nano and micro platforms to sense a variety of analytes such as DNA, proteins and viruses. Among incredible nano and micro structure for bio-analytical applications, the scope of this paper will be limited to micro and nano resonators and nanowire field-effect transistors. Nanomechanical motion of the resonators transducers biological information to readable signals. They are commonly combined with an optical, capacitive or piezo-resistive detection systems. Binding of target molecule to the modified surface of nanowire modulates the current of the nanowire through electrical field-effect. Both detection methods have advantages of label-free, real-time and high sensitive detection. These structures can be extended to fabricate array-type sensors for multiplexed detection and high-throughput analysis. The biosensors based on these structures will be applied to lab-on-a-chip platforms and point-of-care diagnostics. Basic concepts including detection mechanisms and trends in their fields will be covered in this review.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.30 no.6
• /
• pp.1291-1300
• /
• 2020

Recently, a deep learning-based non-profiling side-channel analysis was proposed. The deep learning-based non-profiling analysis is a technique that trains a neural network model for all guessed keys and then finds the correct secret key through the difference in the training metrics. As the performance of non-profiling analysis varies greatly depending on the neural network training model design, a correct model design criterion is required. This paper describes the two types of loss functions and eight labeling methods used in the training model design. It predicts the analysis performance of each labeling method in terms of non-profiling analysis and power consumption model. Considering the characteristics of non-profiling analysis and the HW (Hamming Weight) power consumption model is assumed, we predict that the learning model applying the HW label without One-hot encoding and the Correlation Optimization (CO) loss will have the best analysis performance. And we performed actual analysis on three data sets that are Subbytes operation part of AES-128 1 round. We verified our prediction by non-profiling analyzing two data sets with a total 16 of MLP-based model, which we describe.

• Journal of the Korean Home Economics Association
• /
• v.34 no.5
• /
• pp.235-248
• /
• 1996

The purpose of this study was to search for new food labeling system that would support effective consumer decision-making, by examining consumer behavior of food labels and its' related factors. The survey was conducted on 410 housewives under 60-year-old who live in Seoul or Metropolitan area, using questionnaires and the illustrated material of labels. Respondents generally considered food labels very important as devices. Among 12 items of label which are required to disclose, they evaluated circulation period, and nutritional information. Among 12 items of label, housewives read manufacture date, circulation period, and price more often than business admission number and self standard number. The respondents preferred the format of marking both manufacture date and expiration date on the same side, the format of printing both price and weight on the same side of package, the diagramatic type, and the format represented by ration graph per daily value.

• Journal of Satellite, Information and Communications
• /
• v.9 no.2
• /
• pp.101-106
• /
• 2014

As ESL(Electronic Shelf Label) system is increased at the market in the world, the interference between ESL system and Wi-Fi(Wireless Fidelity) Cellular Phone at 2.4GHz becomes issue. The interference scenario and propagation of the Extended HATA Model were established to analyze the interference from Wi-Fi Cellular Phone into ESL system. Through simulation results based on SEAMCAT(Spectrum Engineering Advanced Monte Carlo Analysis Tool), separation distance was obtained to protect ESL system from Wi-Fi Cellular Phone interference.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• v.27 no.2
• /
• pp.135-145
• /
• 2023

Recently, as deep learning technology has developed, various deep learning technologies have been introduced in handwritten recognition, greatly contributing to performance improvement. The recognition accuracy of handwritten Hangeul recognition has also improved significantly, but prior research has focused on recognizing 520 Hangul characters or 2,350 Hangul characters using SERI95 data or PE92 data. In the past, most of the expressions were possible with 2,350 Hangul characters, but as globalization progresses and information and communication technology develops, there are many cases where various foreign words need to be expressed in Hangul. In this paper, we propose a model that recognizes and combines the consonants, medial vowels, and final consonants of a Korean syllable using a multi-label classification model, and achieves a high recognition accuracy of 98.38% as a result of learning with the public data of Korean handwritten characters, PE92. In addition, this model learned only 2,350 Hangul characters, but can recognize the characters which is not included in the 2,350 Hangul characters

• Annual Conference on Human and Language Technology
• /
• 2023.10a
• /
• pp.622-626
• /
• 2023

중첩 개체명 인식(Nested Named Entity Recognition)은 하나의 개체명 표현 안에 다른 개체명 표현이 들어 있는 중첩 구조의 개체명을 인식하는 작업으로, 중첩 개체명 인식을 위한 학습데이터 구축 작업은 일반 개체명 인식 학습데이터 구축보다 어렵다는 문제가 있다. 본 논문에서는 이러한 문제를 해결하기 위해 Few-shot Learning 환경에 강건한 중첩 개체명 인식 모델을 제안한다. 이를 위해, 기존의 Biaffine 중첩 개체명 인식 모델의 출력 레이어를 라벨 의미 정보를 활용하도록 변경하여 학습데이터가 적은 환경에서 중첩 개체명 인식의 성능을 향상시키도록 하였다. 실험 결과 GENIA 중첩 개체명 인식 데이터의 5-shot, 10-shot, 20-shot 환경에서 기존의 Biaffine 모델보다 평균 10%p이상의 높은 F1-measure 성능을 보였다.

• Journal of Digital Convergence
• /
• v.18 no.8
• /
• pp.371-376
• /
• 2020

This study is about convenience store lunchbox labels for efficient information transmission. Its purpose is to identify the importance of the information that buyers want and to establish labels that can help buyers when purchasing lunchbox. The literature research and survey identified the factors that buyers consider important when purchasing lunchbox, analyzed the information identified on the label and the order of importance of the information. The study found that label information that lunchbox buyers think is important was in order of price, expiration date, calories, lunchbox components, and date of manufacture. Also, buyers who do not check labels pointed the less readable labels. This study expects that the information that buyers think is important will be newly established on labels so that buyers can easily check and purchase information through labels.

• Proceedings of the IEEK Conference
• /
• 2000.11a
• /
• pp.37-40
• /
• 2000

Multi Protocol Label Switching (MPLS) is a high performance method for forwarding packets (frames) through a network. It enables routers at the edge of a network to apply simple labels to packets (frames). we use MPLS in the core network for internet. MPLS provide high speed switching and traffic engineering in MPLS domain but at the Label Edge Router(LER) there is frequently cell discarding via congestion and buffer management method. It is one of the most important reasons retransmission and congestion. In this paper we propose advanced LER scheme that provide less cell loss rate also efficient network infrastructure.