earticle

논문검색

Text Classification Using Parallel Word-level and Character-level Embeddings in Convolutional Neural Networks

원문정보

초록

영어

Deep learning techniques such as Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) show superior performance in text classification than traditional approaches such as Support Vector Machines (SVMs) and Naïve Bayesian approaches. When using CNNs for text classification tasks, word embedding or character embedding is a step to transform words or characters to fixed size vectors before feeding them into convolutional layers. In this paper, we propose a parallel word-level and character-level embedding approach in CNNs for text classification. The proposed approach can capture word-level and character-level patterns concurrently in CNNs. To show the usefulness of proposed approach, we perform experiments with two English and three Korean text datasets. The experimental results show that character-level embedding works better in Korean and word-level embedding performs well in English. Also the experimental results reveal that the proposed approach provides better performance than traditional CNNs with word-level embedding or character-level embedding in both Korean and English documents. From more detail investigation, we find that the proposed approach tends to perform better when there is relatively small amount of data comparing to the traditional embedding approaches.

목차

ABSTRACT
Ⅰ. Introduction
Ⅱ. Related Work
2.1. Text Classification
2.2. Deep Learning in Text Mining
2.3. Convolutional Neural Networks (CNNs) for Text Classification
Ⅲ. Proposed Approach
3.1. Hyperparameters Configuration
3.2. Word Vector and Character Vector
3.3. Regularization and Normalization
Ⅳ. Experimental Design and Datasets
4.1. Comparing Models
4.2. Datasets
Ⅴ. Results and Discussion
5.1. Comlementary Effect
5.2. Size Effect
5.3. Possibility of Improvement through Hyperparameter and Embedding optimization
5.4. Implications
Ⅵ. Conclusion and Future Work
Acknowledgements

저자정보

  • Geonu Kim Undergraduate Student, School of Business, Hanyang University, Korea
  • Jungyeon Jang Manager, Hyundai Motor Company, Korea
  • Juwon Lee Analyst, Korea Ratings, Korea
  • Kitae Kim Researcher, Hana Institute of Finance, KEB Hana Bank, Korea
  • Woonyoung Yeo M.S. Student, Business Informatics from Graduate School, Hanyang University, Korea
  • Jong Woo Kim Professor, School of Business, Hanyang University, Korea

참고문헌

자료제공 : 네이버학술정보

    함께 이용한 논문

      ※ 기관로그인 시 무료 이용이 가능합니다.

      • 5,200원

      0개의 논문이 장바구니에 담겼습니다.