• Journal of the Korea Convergence Society
• /
• v.11 no.11
• /
• pp.33-39
• /
• 2020

In the field of Natural Language Processing, Various research such as Translation, POS Tagging, Q&A, and Sentiment Analysis are globally being carried out. Sentiment Analysis shows high classification performance for English single-domain datasets by pretrained sentence embedding models. In this thesis, the classification performance is compared by Korean E-commerce online dataset with various domain attributes and 6 Neural-Net models are built as BOW (Bag Of Word), LSTM[1], Attention, CNN[2], ELMo[3], and BERT(KoBERT)[4]. It has been confirmed that the performance of pretrained sentence embedding models are higher than word embedding models. In addition, practical Neural-Net model composition is proposed after comparing classification performance on dataset with 17 categories. Furthermore, the way of compressing sentence embedding model is mentioned as future work, considering inference time against model capacity on real-time service.

• Journal of Intelligence and Information Systems
• /
• v.26 no.2
• /
• pp.57-78
• /
• 2020

With the development of the Internet, consumers have had an opportunity to check product information easily through E-Commerce. Product reviews used in the process of purchasing goods are based on user experience, allowing consumers to engage as producers of information as well as refer to information. This can be a way to increase the efficiency of purchasing decisions from the perspective of consumers, and from the seller's point of view, it can help develop products and strengthen their competitiveness. However, it takes a lot of time and effort to understand the overall assessment and assessment dimensions of the products that I think are important in reading the vast amount of product reviews offered by E-Commerce for the products consumers want to compare. This is because product reviews are unstructured information and it is difficult to read sentiment of reviews and assessment dimension immediately. For example, consumers who want to purchase a laptop would like to check the assessment of comparative products at each dimension, such as performance, weight, delivery, speed, and design. Therefore, in this paper, we would like to propose a method to automatically generate multi-dimensional product assessment scores in product reviews that we would like to compare. The methods presented in this study consist largely of two phases. One is the pre-preparation phase and the second is the individual product scoring phase. In the pre-preparation phase, a dimensioned classification model and a sentiment analysis model are created based on a review of the large category product group review. By combining word embedding and association analysis, the dimensioned classification model complements the limitation that word embedding methods for finding relevance between dimensions and words in existing studies see only the distance of words in sentences. Sentiment analysis models generate CNN models by organizing learning data tagged with positives and negatives on a phrase unit for accurate polarity detection. Through this, the individual product scoring phase applies the models pre-prepared for the phrase unit review. Multi-dimensional assessment scores can be obtained by aggregating them by assessment dimension according to the proportion of reviews organized like this, which are grouped among those that are judged to describe a specific dimension for each phrase. In the experiment of this paper, approximately 260,000 reviews of the large category product group are collected to form a dimensioned classification model and a sentiment analysis model. In addition, reviews of the laptops of S and L companies selling at E-Commerce are collected and used as experimental data, respectively. The dimensioned classification model classified individual product reviews broken down into phrases into six assessment dimensions and combined the existing word embedding method with an association analysis indicating frequency between words and dimensions. As a result of combining word embedding and association analysis, the accuracy of the model increased by 13.7%. The sentiment analysis models could be seen to closely analyze the assessment when they were taught in a phrase unit rather than in sentences. As a result, it was confirmed that the accuracy was 29.4% higher than the sentence-based model. Through this study, both sellers and consumers can expect efficient decision making in purchasing and product development, given that they can make multi-dimensional comparisons of products. In addition, text reviews, which are unstructured data, were transformed into objective values such as frequency and morpheme, and they were analysed together using word embedding and association analysis to improve the objectivity aspects of more precise multi-dimensional analysis and research. This will be an attractive analysis model in terms of not only enabling more effective service deployment during the evolving E-Commerce market and fierce competition, but also satisfying both customers.

• Journal of the Korean Society for Library and Information Science
• /
• v.53 no.2
• /
• pp.179-199
• /
• 2019

In this paper, we propose optimal methodologies for classifying patent literature by examining various feature extraction methods, machine learning and deep learning models, and provide optimal performance through experiments. We compared the traditional BoW method and a distributed representation method (word embedding vector) as a feature extraction, and compared the morphological analysis and multi gram as the method of constructing the document collection. In addition, classification performance was verified using traditional machine learning model and deep learning model. Experimental results show that the best performance is achieved when we apply the deep learning model with distributed representation and morphological analysis based feature extraction. In Section, Class and Subclass classification experiments, We improved the performance by 5.71%, 18.84% and 21.53%, respectively, compared with traditional classification methods.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.38 no.4
• /
• pp.621-625
• /
• 2018

This paper proposes a method for bridge damage factor recognition from inspection reports using deep learning. Bridge inspection reports contains inspection results including identified damages and causal analysis results. However, collecting such information from inspection reports manually is limited due to their considerable amount. Therefore, this paper proposes a model for recognizing bridge damage factor from inspection reports applying Named Entity Recognition (NER) using deep learning. Named Entity Recognition, Word Embedding, Recurrent Neural Network, one of deep learning methods, were applied to construct the proposed model. Experimental results showed that the proposed model has abilities to 1) recognize damage and damage factor included in a training data, 2) distinguish a specific word as a damage or a damage factor, depending on its context, and 3) recognize new damage words not included in a training data.

• Journal of Intelligence and Information Systems
• /
• v.28 no.2
• /
• pp.127-146
• /
• 2022

Recently, as word embedding has shown excellent performance in various tasks of deep learning-based natural language processing, researches on the advancement and application of word, sentence, and document embedding are being actively conducted. Among them, cross-language transfer, which enables semantic exchange between different languages, is growing simultaneously with the development of embedding models. Academia's interests in vector alignment are growing with the expectation that it can be applied to various embedding-based analysis. In particular, vector alignment is expected to be applied to mapping between specialized domains and generalized domains. In other words, it is expected that it will be possible to map the vocabulary of specialized fields such as R&D, medicine, and law into the space of the pre-trained language model learned with huge volume of general-purpose documents, or provide a clue for mapping vocabulary between mutually different specialized fields. However, since linear-based vector alignment which has been mainly studied in academia basically assumes statistical linearity, it tends to simplify the vector space. This essentially assumes that different types of vector spaces are geometrically similar, which yields a limitation that it causes inevitable distortion in the alignment process. To overcome this limitation, we propose a deep learning-based vector alignment methodology that effectively learns the nonlinearity of data. The proposed methodology consists of sequential learning of a skip-connected autoencoder and a regression model to align the specialized word embedding expressed in each space to the general embedding space. Finally, through the inference of the two trained models, the specialized vocabulary can be aligned in the general space. To verify the performance of the proposed methodology, an experiment was performed on a total of 77,578 documents in the field of 'health care' among national R&D tasks performed from 2011 to 2020. As a result, it was confirmed that the proposed methodology showed superior performance in terms of cosine similarity compared to the existing linear vector alignment.

• Annual Conference on Human and Language Technology
• /
• 2019.10a
• /
• pp.543-545
• /
• 2019

본 논문은 최신 한국어 의존 구문 분석 모델(Korean dependency parsing model)들과 다양한 앙상블 모델(ensemble model)들을 결합하여 그 성능을 분석한다. 단어 표현은 미리 학습된 워드 임베딩 모델(word embedding model)과 ELMo(Embedding from Language Model), Bert(Bidirectional Encoder Representations from Transformer) 그리고 다양한 추가 자질들을 사용한다. 또한 사용된 의존 구문 분석 모델로는 Stack Pointer Network Model, Deep Biaffine Attention Parser와 Left to Right Pointer Parser를 이용한다. 최종적으로 각 모델의 분석 결과를 앙상블 모델인 Bagging 기법과 XGBoost(Extreme Gradient Boosting) 이용하여 최적의 모델을 제안한다.

• Annual Conference on Human and Language Technology
• /
• 2018.10a
• /
• pp.522-526
• /
• 2018

본 연구는 수많은 챗봇이 생성될 수 있는 챗봇 빌더 시스템에서 저비용 컴퓨팅 파워에서도 구동 가능한 가벼운 문장 분류 시스템을 제안하며, 미등록어 처리를 위해 워드 임베딩 기법인 GloVe를 이용하여 문장 벡터를 생성하고 이를 추가 자질로 사용하는 방법을 소개한다. 제안한 방법으로 자체 구축한 테스트 말뭉치를 이용하여 성능을 평가해본 결과 최대 93.06% 성능을 보였으며, 자체 보유한 CNN 모델과의 비교 평가 결과 성능은 2.5% 낮지만, 모델 학습 속도는 25배, 학습 시 메모리 사용량은 6배, 생성된 모델 파일 크기는 302배나 효율성 있음을 보였다.

• Proceedings of The KACE
• /
• 2017.08a
• /
• pp.221-224
• /
• 2017

본 논문은 사용자 수준에 적합한 맞춤형 학습코스를 추천하여 학습효과를 향상시킬 수 있는 추천모델을 개발하고, 효과분석을 위한 방안을 제시한다. 학습자 개개인의 학습수준이나 학습내용 등에 따라 적합한 학습주제를 선정하여 제공하는 것은 중요하나, 일반적인 추천은 전문가 그룹을 활용한 사람중심의 추천으로 시간이 오래 걸리는 등 자원의 비효율적 한계점[1]을 가지고 있다. 이를 극복하기 위해, TF-IDF를 이용해 단어별 가중치를 계산하여 고빈도 단어를 추출하여 벡터 공간에 배치시키고, Cosine Similarity 기법을 이용해 벡터간의 유사도를 측정하였다. 학습자 프로파일을 분석하고, 학습스킬간의 연관성을 고려하여 맞춤형 학습코스를 추천하기 위해, 워드 임베딩 기법을 적용하였고, 이를 위해 오픈소스 Gensim[2]을 이용하였다. 맞춤형 학습코스 추천 모델의 효과를 분석하기 위한 실험을 설계하고 평가 문항지를 개발하였다.

• Journal of Digital Convergence
• /
• v.17 no.2
• /
• pp.163-170
• /
• 2019

Over the past decade, the development of the Web explosively increased the data. Feature selection step is an important step in extracting valuable data from a large amount of data. This study proposes a novel opinion mining model based on combining feature selection (FS) methods with Word embedding to vector (Word2vec) and BOW (Bag-of-words). FS methods adopted for this study are CFS (Correlation based FS) and IG (Information Gain). To select an optimal FS method, a number of classifiers ranging from LR (logistic regression), NN (neural network), NBN (naive Bayesian network) to RF (random forest), RS (random subspace), ST (stacking). Empirical results with electronics and kitchen datasets showed that LR and ST classifiers combined with IG applied to BOW features yield best performance in opinion mining. Results with laptop and restaurant datasets revealed that the RF classifier using IG applied to Word2vec features represents best performance in opinion mining.

• Journal of Korea Society of Industrial Information Systems
• /
• v.28 no.4
• /
• pp.11-19
• /
• 2023

The paper proposes a smart mirror system that recommends fragrances based on user emotion analysis. This paper combines natural language processing techniques such as embedding techniques (CounterVectorizer and TF-IDF) and machine learning classification models (DecisionTree, SVM, RandomForest, SGD Classifier) to build a model and compares the results. After the comparison, the paper constructs a personal emotion-based fragrance recommendation mirror model based on the SVM and word embedding pipeline-based emotion classifier model with the highest performance. The proposed system implements a personalized fragrance recommendation mirror based on emotion analysis, providing web services using the Flask web framework. This paper uses the Google Speech Cloud API to recognize users' voices and use speech-to-text (STT) to convert voice-transcribed text data. The proposed system provides users with information about weather, humidity, location, quotes, time, and schedule management.