Natural language processing and Chinese computing : 8th CCF International Conference, NLPCC 2019, Dunhuang, China, October 9-14, 2019, Proceedings. Part I / Jie Tang, Min-Yen Kan, Dongyan Zhao, Sujian Li, Hongying Zan (eds.).

International conference proceedings.

Includes author index.

Online resource; title from PDF title page (SpringerLink, viewed October 11, 2019).

Intro; Preface; Organization; Contents -- Part I; Contents -- Part II; Conversational Bot/QA/IR; Variational Attention for Commonsense Knowledge Aware Conversation Generation; 1 Introduction; 2 Related Work; 3 Model Description; 3.1 Task Definition; 3.2 Background: Knowledge Aware Framework; 3.3 Variational Attention for Knowledge Incorporation; 4 Experiments; 4.1 Data; 4.2 Settings; 4.3 Baselines; 4.4 Automatic Evaluation; 4.5 Manual Evaluation; 4.6 Extra Accurate Incorporation Evaluation; 4.7 Study Case; 5 Conclusion; References; Improving Question Answering by Commonsense-Based Pre-training

1 Introduction2 Tasks and Datasets; 3 Commonsense Knowledge; 4 Approach Overview; 5 Commonsense-Based Model; 6 Experiment; 6.1 Model Comparisons and Analysis; 6.2 Error Analysis and Discussion; 7 Related Work; 8 Conclusion; References; Multi-strategies Method for Cold-Start Stage Question Matching of rQA Task; 1 Introduction; 2 Related Work; 3 Methodology; 3.1 Densely-Connected Fusion Attentive Network; 3.2 Strategies; 4 Experiments; 4.1 Datasets; 4.2 Results of the DFAN Model; 4.3 Comparisons Between Strategies; 5 Analysis and Discussion; 5.1 Ablation Study on DFAN; 5.2 Effect of Data Size

6 Conclusion and Future WorkReferences; Multilingual Dialogue Generation with Shared-Private Memory; 1 Introduction; 2 Related Works; 2.1 Dialogue Systems; 2.2 Memory Networks; 2.3 Multi-task Learning; 3 Model; 3.1 Preliminary Background Knowledge; 3.2 Key-Value Memory Augmented Seq2Seq; 3.3 Seq2Seq with Shared-Private Memory; 4 Experimental Settings; 4.1 Datasets; 4.2 Evaluation Metrics; 4.3 Implementation Details; 4.4 Comparisons; 5 Results and Analysis; 5.1 Monolingual Models; 5.2 Multilingual Models; 5.3 Model Analysis; 6 Conclusion; References

Learning Personalized End-to-End Task-Oriented Dialogue Generation1 Introduction; 2 Related Work; 3 Method; 3.1 Problem Definition; 3.2 Framework Structure; 3.3 Training Objective; 4 Experimental Setup; 4.1 Experimental Data; 4.2 Model Configurations; 4.3 Baseline Methods; 4.4 Evaluation Metrics; 5 Experimental Results; 5.1 Ablation Study; 6 Conclusion and Future Work; References; SMART: A Stratified Machine Reading Test; 1 Introduction; 2 Related Work; 2.1 Related English Machine Comprehension Data Sets; 2.2 Related Chinese Machine Comprehension Data Sets; 3 Constructing the SMART Data Set

3.1 Source Data Preparation3.2 Stratified Question and Answer Design; 3.3 Key Statistics of the Data Set; 4 Establishing a Baseline; 5 Experiments; 6 Conclusion and Future Work; References; How Question Generation Can Help Question Answering over Knowledge Base; 1 Introduction; 2 Our Approach; 2.1 Dual Learning; 2.2 Fine Tuning; 3 Models; 3.1 QA Model; 3.2 QG Model; 4 Experiment; 4.1 Setup; 4.2 Relation Detection Results; 4.3 Comparison Results of Dual Learning; 4.4 QG Performance; 4.5 KBQA End-Task Results; 4.6 Case Study; 5 Related Work; 6 Conclusion; References

This two-volume set of LNAI 11838 and LNAI 11839 constitutes the refereed proceedings of the 8th CCF Conference on Natural Language Processing and Chinese Computing, NLPCC 2019, held in Dunhuang, China, in October 2019. The 85 full papers and 56 short papers presented were carefully reviewed and selected from 492 submissions. They are organized in the following topical sections: Conversational Bot/QA/IR; Knowledge graph/IE; Machine Learning for NLP; Machine Translation; NLP Applications; NLP for Social Network; NLP Fundamentals; Text Mining; Short Papers; Explainable AI Workshop; Student Workshop: Evaluation Workshop. -- Provided by publisher.