• Microbiology and Biotechnology Letters
• /
• v.48 no.4
• /
• pp.574-581
• /
• 2020

Next generation sequencing is commonly used to characterize the microbiome structure. MiSeq is commonly used to analyze the microbiome due to its relatively long read length. However, recently, Illumina introduced the 250x2 chip for HiSeq 2500. The purpose of this study was to compare the performance of MiSeq and HiSeq in the context of oral microbiome samples. The MiSeq Reagent Kit V3 and the HiSeq Rapid SBS Kit V2 were used for MiSeq and HiSeq 2500 analyses, respectively. Total read count, read quality score, relative bacterial abundance, community diversity, and relative abundance correlation were analyzed. HiSeq produced significantly more read sequences and assigned taxa compared to MiSeq. Conversely, community diversity was similar in the context of MiSeq and HiSeq. However, depending on the relative abundance, the correlation between the two platforms differed. The correlation between HiSeq and MiSeq sequencing data for highly abundant taxa (> 2%), low abundant taxa (2-0.2%), and rare taxa (0.2% >) was 0.994, 0.860, and 0.416, respectively. Therefore, HiSeq 2500 may also be compatible for microbiome studies. Importantly, the HiSeq platform may allow a high-resolution massive parallel sequencing for the detection of rare taxa.

• KIPS Transactions on Software and Data Engineering
• /
• v.7 no.4
• /
• pp.121-128
• /
• 2018

A Korean morphological analyzer adopts sequence-to-sequence (seq2seq) model, which can generate an output sequence of different length from an input. In general, a seq2seq based Korean morphological analyzer takes a syllable-unit based sequence as an input, and output a syllable-unit based sequence. Syllable-based morphological analysis has the advantage that unknown words can be easily handled, but has the disadvantages that morpheme-based information is ignored. In this paper, we propose a reranking model as a post-processor of seq2seq model that can improve the accuracy of morphological analysis. The seq2seq based morphological analyzer can generate K results by using a beam-search method. The reranking model exploits morpheme-unit embedding information as well as n-gram of morphemes in order to reorder K results. The experimental results show that the reranking model can improve 1.17% F1 score comparing with the original seq2seq model.

• ETRI Journal
• /
• v.39 no.4
• /
• pp.455-466
• /
• 2017

This study focuses on a method for sequential data augmentation in order to alleviate data sparseness problems. Specifically, we present corpus expansion techniques for enhancing the coverage of a language model. Recent recurrent neural network studies show that a seq2seq model can be applied for addressing language generation issues; it has the ability to generate new sentences from given input sentences. We present a method of corpus expansion using a sentence-chain based seq2seq model. For training the seq2seq model, sentence chains are used as triples. The first two sentences in a triple are used for the encoder of the seq2seq model, while the last sentence becomes a target sequence for the decoder. Using only internal resources, evaluation results show an improvement of approximately 7.6% relative perplexity over a baseline language model of Korean text. Additionally, from a comparison with a previous study, the sentence chain approach reduces the size of the training data by 38.4% while generating 1.4-times the number of n-grams with superior performance for English text.

• 한국어정보학회:학술대회논문집
• /
• 2016.10a
• /
• pp.217-219
• /
• 2016

본 논문은 형태소 분석 및 품사 태깅을 위해 seq2seq 주의집중 모델을 이용하는 접근 방법에 대하여 기술한다. seq2seq 모델은 인코더와 디코더로 분할되어 있고, 일반적으로 RNN(recurrent neural network)를 기반으로 한다. 형태소 분석 및 품사 태깅을 위해 seq2seq 모델의 학습 단계에서 음절 시퀀스는 인코더의 입력으로, 각 음절에 해당하는 품사 태깅 시퀀스는 디코더의 출력으로 사용된다. 여기서 음절 시퀀스와 품사 태깅 시퀀스의 대응관계는 주의집중(attention) 모델을 통해 접근하게 된다. 본 연구는 사전 정보나 자질 정보와 같은 추가적 리소스를 배제한 end-to-end 접근 방법의 실험 결과를 제시한다. 또한, 디코딩 단계에서 빔(beam) 서치와 같은 추가적 프로세스를 배제하는 접근 방법을 취한다.

• Annual Conference on Human and Language Technology
• /
• 2016.10a
• /
• pp.217-219
• /
• 2016

본 논문은 형태소 분석 및 품사 태깅을 위해 seq2seq 주의집중 모델을 이용하는 접근 방법에 대하여 기술한다. seq2seq 모델은 인코더와 디코더로 분할되어 있고, 일반적으로 RNN(recurrent neural network)를 기반으로 한다. 형태소 분석 및 품사 태깅을 위해 seq2seq 모델의 학습 단계에서 음절 시퀀스는 인코더의 입력으로, 각 음절에 해당하는 품사 태깅 시퀀스는 디코더의 출력으로 사용된다. 여기서 음절 시퀀스와 품사 태깅 시퀀스의 대응관계는 주의집중(attention) 모델을 통해 접근하게 된다. 본 연구는 사전 정보나 자질 정보와 같은 추가적 리소스를 배제한 end-to-end 접근 방법의 실험 결과를 제시한다. 또한, 디코딩 단계에서 빔(beam) 서치와 같은 추가적 프로세스를 배제하는 접근 방법을 취한다.

• Annual Conference on Human and Language Technology
• /
• 2022.10a
• /
• pp.257-261
• /
• 2022

본 연구에서는 한국어 AMR 자동 파싱을 하기 위해 seq2seq 방법론을 적용하였다. Seq2seq 방법론은 AMR 파싱 태스크를 자연어 문장을 바탕으로 선형화된(linearization) 그래프의 문자열을 번역해내는 과정을 거친다. 본고는 Transformer 모델을 파싱 모델로 적용하여 2020년 공개된 한국어 AMR와 자체적으로 구축된 한국어 <어린 왕자> AMR 데이터에서 실험을 진행하였다. 이 연구에서 seq2seq 방법론 기반 한국어 AMR 파싱의 성능은 Smatch F1-Score 0.30으로 나타났다.

• Korean Journal of Clinical Laboratory Science
• /
• v.55 no.4
• /
• pp.235-243
• /
• 2023

As next-generation sequencing has been developed and used widely, RNA-sequencing (RNA-seq) has rapidly emerged as the first choice of tools to validate global transcriptome profiling. With the significant advances in RNA-seq, various types of RNA-seq have evolved in conjunction with the progress in bioinformatic tools. On the other hand, it is difficult to interpret the complex data underlying the biological meaning without a general understanding of the types of RNA-seq and bioinformatic approaches. In this regard, this paper discusses the two main sections of RNA-seq. First, two major variants of RNA-seq are described and compared with the standard RNA-seq. This provides insights into which RNA-seq method is most appropriate for their research. Second, the most widely used RNA-seq data analyses are discussed: (1) exploratory data analysis and (2) pathway enrichment analysis. This paper introduces the most widely used exploratory data analysis for RNA-seq, such as principal component analysis, heatmap, and volcano plot, which can provide the overall trends in the dataset. The pathway enrichment analysis section introduces three generations of pathway enrichment analysis and how they generate enriched pathways with the RNA-seq dataset.

• Annual Conference on Human and Language Technology
• /
• 2018.10a
• /
• pp.151-154
• /
• 2018

본 논문에서는 CNN Seq2Seq 구조를 이용해 한국어 대화 시스템을 개발하였다. 기존 Seq2Seq는 RNN 혹은 그 변형 네트워크에 데이터를 입력하고, 입력이 완료된 후의 은닉 층의 embedding에 기반해 출력열을 생성한다. 우리는 CNN Seq2Seq로 입력된 발화에 대해 출력 발화를 생성하는 대화 모델을 학습하였고, 그 성능을 측정하였다. CNN에 대해서는 약 12만 발화 쌍을 이용하여 학습하고 1만 발화 쌍으로 실험하였다. 평가 결과 제안 모델이 기존의 RNN 기반 모델에 비해 우수한 결과를 보였다.

• Communications for Statistical Applications and Methods
• /
• v.22 no.2
• /
• pp.181-199
• /
• 2015

In a short history, RNA-seq data have established a revolutionary tool to directly decode various scenarios occurring on whole genome-wide expression profiles in regards with differential expression at gene, transcript, isoform, and exon specific quantification, genetic and genomic mutations, and etc. RNA-seq technique has been rapidly replacing arrays with seq-based platform experimental settings by revealing a couple of advantages such as identification of alternative splicing and allelic specific expression. The remarkable characteristics of high-throughput large-scale expression profile in RNA-seq are lied on expression levels of read counts, structure of correlated samples and genes, larger number of genes compared to sample size, different sampling rates, inevitable systematic RNA-seq biases, and etc. In this study, we will comprehensively review how robust Bayesian and non-parametric methods have a better performance than classical statistical approaches by explicitly incorporating such intrinsic RNA-seq specific features with flexible and more appropriate assumptions and distributions in practice.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.12 no.3
• /
• pp.242-250
• /
• 2019

In this paper, we propose a method to predict the failure of industrial robot using Seq2Seq (Sequence to Sequence) model, which is a model for transforming time series data among Artificial Neural Network models. The proposed method uses the data of the joint current and angular value, which can be measured by the robot itself, without additional sensor for fault diagnosis. After preprocessing the measured data for the model to learn, the Seq2Seq model was trained to convert the current to angle. Abnormal degree for fault diagnosis uses RMSE (Root Mean Squared Error) during unit time between predicted angle and actual angle. The performance evaluation of the proposed method was performed using the test data measured under different conditions of normal and defective condition of the robot. When the Abnormal degree exceed the threshold, it was classified as a fault, and the accuracy of the fault diagnosis was 96.67% from the experiment. The proposed method has the merit that it can perform fault prediction without additional sensor, and it has been confirmed from the experiment that high diagnostic performance and efficiency are available without requiring deep expert knowledge of the robot.