Lately, a large amount of textual data have been poured out of the Internet and the technology to refine them is needed. Most of these data are long text and often have no title. Therefore, in this paper, we propose a technique to combine the sequence-to-sequence model of RNN and the REINFORCE algorithm to generate the title of the long text automatically. In addition, the TextRank algorithm was applied to extract a summarized text to minimize information loss in order to protect the shortcomings of the sequence-to-sequence model in which an information is lost when long texts are used. Through the experiment, the techniques proposed in this study are shown to be superior to the existing ones.
목차
Abstract 1. 서론 2. 관련 연구 3. RNN과 강화 학습을 결합한 제목 생성 모델 3.1 추출 요약 모델 3.2 사전 학습 3.3 강화 학습을 적용한 모델 3.4 보상함수 4. 실험 5.결과 및 성능 평가 6. 결론 및 추후 연구 References
키워드
Title GenerationDeep LearningReinforcement Learning
저자
조성민 [ Sung-Min Cho | Graduate School of Computer Science, Kwangwoon University ]
First Author
김우생 [ Wooseng Kim | Professor, School of Software, Kwangwoon University ]
Corresponding Author