ABSTRACT
Q&A system plays an increasingly important role in the modern society with information explosion, and knowledge reasoning model (KRM) is the main research content of Q&A system. Existing knowledge reasoning models are mainly divided into text-based information retrieval (IR) and knowledge graph embedding (KGE). KGE is superior to IR in terms of storage and reasoning capabilities for massive factoid information, but lacks the ability to reason non-factoid information, merely focus on the mining of structural information without the semantic information. We proposed a knowledge reasoning model based on non-factoid information enhancement (NFE-KRM) in scenic Q&A. It realizes the KGE integrates semantic information (SIKGE) and the unified semantic embedding space (USES), so that NFE-KRM has the ability to answer both factoid and non-factoid questions. We have used a large number of experiments to prove that SIKGE gets a better performance on Mean Rank and Hits@10. NFE-KRM's F1 score and accuracy on the mixed dataset are both competitive.
- Dettmers, T., Minervini, P., Stenetorp, P., & Riedel, S. 2018. Convolutional 2d knowledge graph embeddings. In Proceedings of the AAAI conference on artificial intelligence (Vol. 32, No. 1).Google ScholarCross Ref
- Yao, L., Mao, C., & Luo, Y. 2019. KG-BERT: BERT for knowledge graph completion. arXiv preprint arXiv:1909.03193.Google Scholar
- Schlichtkrull, M., Kipf, T. N., Bloem, P., Berg, R. V. D., Titov, I., & Welling, M. 2018. Modeling relational data with graph convolutional networks. In European semantic web conference (pp. 593-607). Springer, Cham.Google Scholar
- Zhiqing Sun, Zhi-Hong Deng, Jian-Yun Nie, and Jian Tang. 2019. RotatE: Knowledge Graph Embedding by Relational Rotation in Complex Space. In ICLR.Google Scholar
- Yingjie, G., Xiaolin, G., & Defu, L. 2019. TT-Net: Topic Transfer-Based Neural Network for Conversational Reading Comprehension. IEEE Access, 7, 116696-116705.Google Scholar
- Ye X, Yavuz S, Hashimoto K, 2021. Rng-kbqa: Generation augmented iterative ranking for knowledge base question answering[J]. arXiv preprint arXiv:2109.08678Google Scholar
- Das R, Zaheer M, Thai D, 2021. Case-based reasoning for natural language queries over knowledge bases[J]. arXiv preprint arXiv:2104.08762Google Scholar
- Das R, Godbole A, Naik A, 2022. Knowledge base question answering by case-based reasoning over subgraphs[C]//International Conference on Machine Learning. PMLR, 4777-4793.Google Scholar
- LAN, Yunshi; WANG, Shuohang; and JIANG, Jing. 2019. Knowledge base question answering with topic units. Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence. 5046-5052. Research Collection School Of Computing and Information Systems.Google Scholar
- Devlin, J., Chang, M. W., Lee, K., & Toutanova, K. 2018. Bert: Pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805.Google Scholar
- Gao, T., Yao, X., & Chen, D. 2021. Simcse: Simple contrastive learning of sentence embeddings. arXiv preprint arXiv:2104.08821.Google Scholar
- Gu, Y., & Su, Y. 2022. ArcaneQA: Dynamic Program Induction and Contextualized Encoding for Knowledge Base Question Answering. arXiv preprint arXiv:2204.08109.Google Scholar
Index Terms
- A Knowledge Reasoning Model Based on Non-Factoid Information Enhancement
Recommendations
Using knowledge to facilitate factoid answer pinpointing
COLING '02: Proceedings of the 19th international conference on Computational linguistics - Volume 1In order to answer factoid questions, the Webclopedia QA system employs a range of knowledge resources. These include a QA Typology with answer patterns, WordNet, information about typical numerical answer ranges, and semantic relations identified by a ...
A Knowledge Reasoning Model Based on Rough Set Theory
GCIS '10: Proceedings of the 2010 Second WRI Global Congress on Intelligent Systems - Volume 03With the extensive use of computer and the rapid development of information technology, how to seek for useful knowledge from massive data becomes a difficult problem of knowledge service. According to knowledge reasoning of massive information, this ...
A Contextual Information-Augmented Probabilistic Case-Based Reasoning Model for Knowledge Graph Reasoning
Case-Based Reasoning Research and DevelopmentAbstractKnowledge Graph Reasoning (KGR) is one effective method to improve incompleteness and sparsity problems, which infers new knowledge based on existing knowledge. Although the probabilistic case-based reasoning (CBR) model can predict attributes for ...
Comments