• The Journal of the Korea institute of electronic communication sciences
• /
• v.17 no.5
• /
• pp.801-808
• /
• 2022

Research on how to embed knowledge in large-scale Linked Data and apply neural network models for entity matching is relatively scarce. The most fundamental problem with this is that different labels lead to lexical heterogeneity. In this paper, we propose an extended GCN (Graph Convolutional Network) model that combines re-align structure to solve this lexical heterogeneity problem. The proposed model improved the performance by 53% and 40%, respectively, compared to the existing embedded-based MTransE and BootEA models, and improved the performance by 5.1% compared to the GCN-based RDGCN model.

• International Journal of Computer Science & Network Security
• /
• v.23 no.11
• /
• pp.21-31
• /
• 2023

In this paper, we focused on the problem of evaluating multi-class classification accuracy and simulation of multiple classifier performance metrics. Multi-class classifiers for sentiment analysis involved many challenges, whereas previous research narrowed to the binary classification model since it provides higher accuracy when dealing with text data. Thus, we take inspiration from the non-linear Support Vector Machine to modify the algorithm by embedding dynamic hyperplanes representing multiple class labels. Then we analyzed the performance of multi-class classifiers using macro-accuracy, micro-accuracy and several other metrics to justify the significance of our algorithm enhancement. Furthermore, we hybridized Enhanced Convolution Neural Network (ECNN) with Dynamic Support Vector Machine (DSVM) to demonstrate the effectiveness and efficiency of the classifier towards multi-class text data. We performed experiments on three hybrid classifiers, which are ECNN with Binary SVM (ECNN-BSVM), and ECNN with linear Multi-Class SVM (ECNN-MCSVM) and our proposed algorithm (ECNNDSVM). Comparative experiments of hybrid algorithms yielded 85.12 % for single metric accuracy; 86.95 % for multiple metrics on average. As for our modified algorithm of the ECNN-DSVM classifier, we reached 98.29 % micro-accuracy results with an f-score value of 98 % at most. For the future direction of this research, we are aiming for hyperplane optimization analysis.

• Annual Conference on Human and Language Technology
• /
• 2017.10a
• /
• pp.137-142
• /
• 2017

원격 지도 학습은 자동으로 매우 큰 코퍼스와 지식베이스 간의 주석 데이터를 생성하여 기계 학습에 필요한 학습 데이터를 사람의 손을 빌리지 않고 저렴한 비용으로 만들 수 있어, 많은 연구들이 관계 추출 문제를 해결하기 위해 원격 지도 학습 방법을 적용하고 있다. 그러나 기존 연구들에서는 모델 학습의 입력으로 사용되는 단어 임베딩에서 단어의 동형이의어 성질을 반영하지 못한다는 단점이 있다. 때문에 서로 다른 의미를 가진 동형이의어가 하나의 임베딩 값을 가지다 보니, 단어의 의미를 정확히 파악하지 못한채 관계 추출 모델을 학습한다고 볼 수 있다. 본 논문에서는 원격 지도 학습 기반 관계 추출 모델에 다중-어의 단어 임베딩을 적용한 모델을 제안한다. 다중-어의 단어 임베딩 학습을 위해 어의 중의성 해소 모듈을 활용하였으며, 관계 추출 모델은 문장 내 주요 특징을 효율적으로 파악하는 모델인 CNN과 PCNN을 활용하였다. 본 논문에서 제안하는 다중-어의 단어 임베딩 적용 관계추출 모델의 성능을 평가하기 위해 추가적으로 2가지 방식의 단어 임베딩을 학습하여 비교 평가를 수행하였고, 그 결과 어의 중의성 해소 모듈을 활용한 단어 임베딩을 활용하였을 때 관계추출 모델의 성능이 향상된 결과를 보였다.

• Journal of KIISE:Computer Systems and Theory
• /
• v.29 no.1
• /
• pp.22-34
• /
• 2002

In this paper, we consider ring embeding problem in faulty (n,k) star graphs which is recently proposed as an alternative interconnection network topology, By effectively utilizing such strategies as series of dimension expansions and even distribution of faulty nodes into sub-stars in graph itself. we prove that it is possible to construct a maximal fault-free ring excluding only faulty nodes when the number of faults is no more than n-3 and $n-k{\geq}2$, and also propose an algorithm which can embed the corresponding ring in (n.k)-star graphs This results will be applied into the multicasting applications that the underlying cycle properties on the multi-computer system.

• 한국어정보학회:학술대회논문집
• /
• 2017.10a
• /
• pp.137-142
• /
• 2017

원격 지도 학습은 자동으로 매우 큰 코퍼스와 지식베이스 간의 주석 데이터를 생성하여 기계 학습에 필요한 학습 데이터를 사람의 손을 빌리지 않고 저렴한 비용으로 만들 수 있어, 많은 연구들이 관계 추출 문제를 해결하기 위해 원격 지도 학습 방법을 적용하고 있다. 그러나 기존 연구들에서는 모델 학습의 입력으로 사용되는 단어 임베딩에서 단어의 동형이의어 성질을 반영하지 못한다는 단점이 있다. 때문에 서로 다른 의미를 가진 동형이의어가 하나의 임베딩 값을 가지다 보니, 단어의 의미를 정확히 파악하지 못한 채 관계 추출 모델을 학습한다고 볼 수 있다. 본 논문에서는 원격 지도 학습 기반 관계 추출 모델에 다중-어의 단어 임베딩을 적용한 모델을 제안한다. 다중-어의 단어 임베딩 학습을 위해 어의 중의성 해소 모듈을 활용하였으며, 관계 추출 모델은 문장 내 주요 특징을 효율적으로 파악하는 모델인 CNN과 PCNN을 활용하였다. 본 논문에서 제안하는 다중-어의 단어 임베딩 적용 관계추출 모델의 성능을 평가하기 위해 추가적으로 2가지 방식의 단어 임베딩을 학습하여 비교 평가를 수행하였고, 그 결과 어의 중의성 해소 모듈을 활용한 단어 임베딩을 활용하였을 때 관계추출 모델의 성능이 향상된 결과를 보였다.

• KIPS Transactions on Software and Data Engineering
• /
• v.9 no.2
• /
• pp.61-68
• /
• 2020

E-sports has grown steadily in recent years and has become a popular sport in the world. In this paper, we propose a win-loss prediction model of League of Legends at the start of the game. In League of Legends, the combination of a champion statistics of the team that is made through each player's selection affects the win-loss of the game. The proposed model is a deep learning model based on Bidirectional LSTM embedding which considers a combination of champion statistics for each team without any domain knowledge. Compared with other prediction models, the highest prediction accuracy of 58.07% was evaluated in the proposed model considering a combination of champion statistics for each team.

• International Journal of Advanced Culture Technology
• /
• v.5 no.1
• /
• pp.83-89
• /
• 2017

In this paper, we propose an extended method of one-hot Long Short-Term Memory (LSTM) and evaluate the performance on spam filtering task. Most of traditional methods proposed for spam filtering task use word occurrences to represent spam or non-spam messages and all syntactic and semantic information are ignored. Major issue appears when both spam and non-spam messages share many common words and noise words. Therefore, it becomes challenging to the system to filter correct labels between spam and non-spam. Unlike previous studies on information filtering task, instead of using only word occurrence and word context as in probabilistic models, we apply a neural network-based approach to train the system filter for a better performance. In addition to one-hot representation, using term weight with attention mechanism allows classifier to focus on potential words which most likely appear in spam and non-spam collection. As a result, we obtained some improvement over the performances of the previous methods. We find out using region embedding and pooling features on the top of LSTM along with attention mechanism allows system to explore a better document representation for filtering task in general.

• Journal of Electrical Engineering and information Science
• /
• v.2 no.5
• /
• pp.17-26
• /
• 1997

Interconnect characterization on a wafer level was performed. Test patterns for single, two-coupled, and triple-coupled lines ere designed by using 0.5$\mu\textrm{m}$ CMOS process. Then interconnect capacitances and resistances were experimentally extracted by using tow port network measurements, Particularly to eliminate parasitic effects, the Y-parameter de-embedding was performed with specially designed de-embedding patterns. Also, for the purpose of comparisons, capacitance matrices were calculated by using the existing CAD model and field-solver-based commercial simulator, METAL and MEDICI. This work experimentally verifies that existing CAD models or parameter extraction may have large deviation from real values. The signal transient simulation with the experimental data and other methodologies such as field-solver-based simulation and existing model was performed. as expected, the significantly affect on the signal delay and crosstalk. The signal delay due to interconnects dominates the sub-micron-based a gate delay (e.g., inverter). Particularly, coupling capacitance deviation is so large (about more than 45% in the worst case) that signal integrity cannot e guaranteed with the existing methodologies. The characterization methodologies of this paper can be very usefully employed for the signal integrity verification or he electrical design rule establishments of IC interconnects in the industry.

• International Journal of Contents
• /
• v.17 no.4
• /
• pp.1-15
• /
• 2021

Named entity recognition (NER) is a crucial task for NLP, which aims to extract information from texts. To build NER systems, deep learning (DL) models are learned with dictionary features by mapping each word in the dataset to dictionary features and generating a unique index. However, this technique might generate noisy labels, which pose significant challenges for the NER task. In this paper, we proposed DL-dictionary features, and evaluated them on two datasets, including the OntoNotes 5.0 dataset and our new infectious disease outbreak dataset named GFID. We used (1) a Bidirectional Long Short-Term Memory (BiLSTM) character and (2) pre-trained embedding to concatenate with (3) our proposed features, named the Convolutional Neural Network (CNN), BiLSTM, and self-attention dictionaries, respectively. The combined features (1-3) were fed through BiLSTM - Conditional Random Field (CRF) to predict named entity classes as outputs. We compared these outputs with other predictions of the BiLSTM character, pre-trained embedding, and dictionary features from previous research, which used the exact matching and partial matching dictionary technique. The findings showed that the model employing our dictionary features outperformed other models that used existing dictionary features. We also computed the F1 score with the GFID dataset to apply this technique to extract medical or healthcare information.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.18 no.5
• /
• pp.217-223
• /
• 2023

In this paper, we propose a new model for the conditional generation of music, considering key and rhythm, fundamental elements of music. MIDI sheet music is converted into a WAV format, which is then transformed into a Mel Spectrogram using the Short-Time Fourier Transform (STFT). Using this information, key and rhythm details are classified by passing through two Convolutional Neural Networks (CNNs), and this information is again fed into the Music Transformer. The key and rhythm details are combined by differentially multiplying the weights and the embedding vectors of the MIDI events. Several experiments are conducted, including a process for determining the optimal weights. This research represents a new effort to integrate essential elements into music generation and explains the detailed structure and operating principles of the model, verifying its effects and potentials through experiments. In this study, the accuracy for rhythm classification reached 94.7%, the accuracy for key classification reached 92.1%, and the Negative Likelihood based on the weights of the embedding vector resulted in 3.01.