• 지식경영연구
• /
• 제21권1호
• /
• pp.99-116
• /
• 2020

Text is the most widely used means of exchanging or expressing knowledge and information in the real world. Recently, researches on structuring unstructured text data for text analysis have been actively performed. One of the most representative document embedding method (i.e. doc2Vec) generates a single vector for each document using the whole corpus included in the document. This causes a limitation that the document vector is affected by not only core words but also other miscellaneous words. Additionally, the traditional document embedding algorithms map each document into only one vector. Therefore, it is not easy to represent a complex document with interdisciplinary subjects into a single vector properly by the traditional approach. In this paper, we introduce a multi-vector document embedding method to overcome these limitations of the traditional document embedding methods. After introducing the previous study on multi-vector document embedding, we visually analyze the effects of the multi-vector document embedding method. Firstly, the new method vectorizes the document using only predefined keywords instead of the entire words. Secondly, the new method decomposes various subjects included in the document and generates multiple vectors for each document. The experiments for about three thousands of academic papers revealed that the single vector-based traditional approach cannot properly map complex documents because of interference among subjects in each vector. With the multi-vector based method, we ascertained that the information and knowledge in complex documents can be represented more accurately by eliminating the interference among subjects.

• 대한한의학원전학회지
• /
• 제32권1호
• /
• pp.61-74
• /
• 2019

Objectives : The purpose of this study is to help select an appropriate word embedding method when analyzing East Asian traditional medicine texts as data. Methods : Based on prescription data that imply traditional methods in traditional East Asian medicine, we have examined 4 count-based word embedding and 2 prediction-based word embedding methods. In order to intuitively compare these word embedding methods, we proposed a "prescription generating game" and compared its results with those from the application of the 6 methods. Results : When the adjacent vectors are extracted, the count-based word embedding method derives the main herbs that are frequently used in conjunction with each other. On the other hand, in the prediction-based word embedding method, the synonyms of the herbs were derived. Conclusions : Counting based word embedding methods seems to be more effective than prediction-based word embedding methods in analyzing the use of domesticated herbs. Among count-based word embedding methods, the TF-vector method tends to exaggerate the frequency effect, and hence the TF-IDF vector or co-word vector may be a more reasonable choice. Also, the t-score vector may be recommended in search for unusual information that could not be found in frequency. On the other hand, prediction-based embedding seems to be effective when deriving the bases of similar meanings in context.

• 지능정보연구
• /
• 제25권3호
• /
• pp.19-41
• /
• 2019

텍스트 데이터에 대한 다양한 분석을 위해 최근 비정형 텍스트 데이터를 구조화하는 방안에 대한 연구가 활발하게 이루어지고 있다. doc2Vec으로 대표되는 기존 문서 임베딩 방법은 문서가 포함한 모든 단어를 사용하여 벡터를 만들기 때문에, 문서 벡터가 핵심 단어뿐 아니라 주변 단어의 영향도 함께 받는다는 한계가 있다. 또한 기존 문서 임베딩 방법은 하나의 문서가 하나의 벡터로 표현되기 때문에, 다양한 주제를 복합적으로 갖는 복합 문서를 정확하게 사상하기 어렵다는 한계를 갖는다. 본 논문에서는 기존의 문서 임베딩이 갖는 이러한 두 가지 한계를 극복하기 위해 다중 벡터 문서 임베딩 방법론을 새롭게 제안한다. 구체적으로 제안 방법론은 전체 단어가 아닌 핵심 단어만 이용하여 문서를 벡터화하고, 문서가 포함하는 다양한 주제를 분해하여 하나의 문서를 여러 벡터의 집합으로 표현한다. KISS에서 수집한 총 3,147개의 논문에 대한 실험을 통해 복합 문서를 단일 벡터로 표현하는 경우의 벡터 왜곡 현상을 확인하였으며, 복합 문서를 의미적으로 분해하여 다중 벡터로 나타내는 제안 방법론에 의해 이러한 왜곡 현상을 보정하고 각 문서를 더욱 정확하게 임베딩할 수 있음을 확인하였다.

• Journal of Multimedia Information System
• /
• 제9권2호
• /
• pp.93-102
• /
• 2022

Dependency parsing is a decision problem of the syntactic relation between words in a sentence. Recently, deep learning models are used for dependency parsing based on the word representations in a continuous vector space. However, it causes a mislabeled tagging problem for the proper nouns that rarely appear in the training corpus because it is difficult to express out-of-vocabulary (OOV) words in a continuous vector space. To solve the OOV problem in dependency parsing, we explored the proper noun embedding method according to the embedding unit. Before representing words in a continuous vector space, we replace the proper nouns with a special token and train them for the contextual features by using the multi-layer bidirectional LSTM. Two models of the syllable-based and morpheme-based unit are proposed for proper noun embedding and the performance of the dependency parsing is more improved in the ensemble model than each syllable and morpheme embedding model. The experimental results showed that our ensemble model improved 1.69%p in UAS and 2.17%p in LAS than the same arc-eager approach-based Malt parser.

• Journal of Information Processing Systems
• /
• 제16권6호
• /
• pp.1407-1423
• /
• 2020

A graph is a data structure consisting of nodes and edges between these nodes. Graph embedding is to generate a low dimensional vector for a given graph that best represents the characteristics of the graph. Recently, there have been studies on graph embedding, especially using deep learning techniques. However, until now, most deep learning-based graph embedding techniques have focused on unweighted graphs. Therefore, in this paper, we propose a graph embedding technique for weighted graphs based on long short-term memory (LSTM) autoencoders. Given weighted graphs, we traverse each graph to extract node-weight sequences from the graph. Each node-weight sequence represents a path in the graph consisting of nodes and the weights between these nodes. We then train an LSTM autoencoder on the extracted node-weight sequences and encode each nodeweight sequence into a fixed-length vector using the trained LSTM autoencoder. Finally, for each graph, we collect the encoding vectors obtained from the graph and combine them to generate the final embedding vector for the graph. These embedding vectors can be used to classify weighted graphs or to search for similar weighted graphs. The experiments on synthetic and real datasets show that the proposed method is effective in measuring the similarity between weighted graphs.

• 스마트미디어저널
• /
• 제7권4호
• /
• pp.24-29
• /
• 2018

딥러닝에서 자연어 처리를 위한 텍스트 분석 기법은 워드 임베딩을 통해 단어를 벡터 형태로 표현한다. 본 논문에서는 워드 임베딩 기법과 딥러닝 기법을 이용하여 SMS 문자 메시지를 문서 벡터로 구성하고 이를 스팸 문자 메시지와 정상적인 문자 메시지로 분류하는 방법을 제안하였다. 유사한 문맥을 가진 단어들은 벡터 공간에서 인접한 벡터 공간에 표현되도록 하기 위해 전처리 과정으로 자동 띄어쓰기를 적용하고 스팸 문자 메시지로 차단되는 것을 피하기 위한 목적으로 음절의 자모를 특수기호로 왜곡하여 맞춤법이 파괴된 상태로 단어 벡터와 문장 벡터를 생성하였다. 또한 문장 벡터 생성 시 CBOW와 skip gram이라는 두 가지 워드 임베딩 알고리즘을 적용하여 문장 벡터를 표현하였으며, 딥러닝을 이용한 스팸 문자 메시지 필터링의 성능 평가를 위해 SVM Light와 정확도를 비교 측정하였다.

• 정보과학회 논문지
• /
• 제43권6호
• /
• pp.678-685
• /
• 2016

한국어 개체명 인식에 다양한 연구가 있었지만, 영어 개체명 인식에 비해 자질이 부족한 문제를 가지고 있다. 본 논문에서는 한국어 개체명 인식의 자질 부족 문제를 해결하기 위해 word embedding 자질을 개체명 인식에 사용하는 방법을 제안한다. CBOW(Continuous Bag-of-Words) 모델을 이용하여 word vector를 생성하고, word vector로부터 K-means 알고리즘을 이용하여 군집 정보를 생성한다. word vector와 군집 정보를 word embedding 자질로써 CRFs(Conditional Random Fields)에 사용한다. 실험 결과 TV 도메인과 Sports 도메인, IT 도메인에서 기본 시스템보다 각각 1.17%, 0.61%, 1.19% 성능이 향상되었다. 또한 제안 방법이 다른 개체명 인식 및 분류 시스템보다 성능이 향상되는 것을 보여 그 효용성을 입증했다.

• 한국음향학회지
• /
• 제39권2호
• /
• pp.137-142
• /
• 2020

문장 종속 짧은 발화에서 문장 독립 긴 발화까지 다양한 환경에서 I-vector 특징에 기반을 둔 많은 연구가 수행되었다. 본 논문에서는 원거리 잡음 환경에서 녹음한 데이터에서 Probabilistic Linear Discriminant Analysis(PLDA)를 적용한 I-vector와 주의 집중 기법을 접목한 Long Short Term Memory(LSTM) 기반의 화자 임베딩을 추출하여 결합한 화자 검증 알고리즘을 소개한다. LSTM 모델의 Equal Error Rate(EER)이 15.52 %, Attention-LSTM 모델이 8.46 %로 7.06 % 성능이 향상되었다. 이로써 본 논문에서 제안한 기법이 임베딩을 휴리스틱 하게 정의하여 사용하는 기존 추출방법의 문제점을 해결할 수 있는 것을 확인하였다. PLDA를 적용한 I-vector의 EER이 6.18 %로 결합 전 가장 좋은 성능을 보였다. Attention-LSTM 기반 임베딩과 결합하였을 때 EER이 2.57 %로 기존보다 3.61 % 감소하여 상대적으로 58.41 % 성능이 향상되었다.

• 대한수학회보
• /
• 제39권2호
• /
• pp.289-297
• /
• 2002

It was shown by L. Polterovich ([3]) that if L is a totally real submanifold of a symplectic manifold $(M,\omega)$ and L is parallelizable then L is normal. So we try to find an answer to the question of whether there is a compatible almost complex structure J on the symplectic vector bundle $TM$\mid$_{L}$ such that $TL{\cap}JTL=0$ assuming L is normal and parallelizable. Although we could not reach an answer, we observed that the claim holds at the vector space level. And related to the question, we showed that for a symplectic vector bundle $(M,\omega)$ of rank 2n and $E=E_1{\bigoplus}E_2$, where $E=E_1,E_2$are Lagrangian subbundles of E, there is an almost complex structure J on E compatible with ${\omega}$ and $JE_1=E_2$. And finally we provide a necessary and sufficient condition for a given embedding into a symplectic manifold to be normal.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권7호
• /
• pp.3475-3496
• /
• 2018

Most of current watermarking algorithms for GIS vector data embed copyright information by means of modifying the coordinate values, which will do harm to its quality and accuracy. To preserve the fidelity of vector line data and protect its copyright at the same time, a lossless watermarking algorithm is proposed based on storage feature in this paper. Firstly, the superiority of embedding watermark based on storage feature is demonstrated theoretically and technically. Then, the basic concepts and operations on storage feature have been defined including length and angle of the polyline feature. In the process of embedding watermark, the watermark information is embedded into directions of polyline feature by the quantitative mechanism, while the positions of embedding watermark are determined by the feature length. Hence, the watermark can be extracted by the same geometric features without original data or watermark. Finally, experiments have been conducted to show that coordinate values remain unchanged after embedding watermark. Moreover, experimental results are presented to illustrate the effectiveness of the method.