Leveraging Abstract Meaning Representation for Knowledge Base Question Answering

Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021, pages 3884–3894 August 1–6, 2021. ©2021 Association for Computational Linguistics 3884 Leveraging Abstract Meaning Representation for Knowledge Base Question Answering Pavan Kapanipathi∗, Ibrahim Abdelaziz∗, Srinivas Ravishankar∗, Salim Roukos, Alexander Gray, Ramon Astudillo, Maria Chang, Cristina Cornelio, Saswati Dana, Achille Fokoue, Dinesh Garg, Alfio Gliozzo, Sairam Gurajada, Hima Karanam, Naweed Khan, Dinesh Khandelwal, Young-Suk Lee, Yunyao Li, Francois Luus, Ndivhuwo Makondo, Nandana Mihindukulasooriya, Tahira Naseem, Sumit Neelam, Lucian Popa, Revanth Reddy, Ryan Riegel, Gaetano Rossiello, Udit Sharma, G P Shrivatsa Bhargav, Mo Yu IBM Research Abstract Knowledge base question answering (KBQA) is an important task in Natural Language Pro- cessing. Existing approaches face significant challenges including complex question under- standing, necessity for reasoning, and lack of large end-to-end training datasets. In this work, we propose Neuro-Symbolic Question Answering (NSQA), a modular KBQA system, that leverages (1) Abstract Meaning Repre- sentation (AMR) parses for task-independent question understanding; (2) a simple yet ef- fective graph transformation approach to con- vert AMR parses into candidate logical queries that are aligned to the KB; (3) a pipeline-based approach which integrates multiple, reusable modules that are trained specifically for their individual tasks (semantic parser, entity and relationship linkers, and neuro-symbolic rea- soner) and do not require end-to-end training data. NSQA achieves state-of-the-art per- formance on two prominent KBQA datasets based on DBpedia (QALD-9 and LC-QuAD 1.0). Furthermore, our analysis emphasizes that AMR is a powerful tool for KBQA sys- tems. 1 Introduction Knowledge base question answering (KBQA) is a sub-field within Question Answering with de- sirable characteristics for real-world applications. KBQA requires a system to answer a natural lan- guage question based on facts available in a Knowl- edge Base (KB) (Zou et al., 2014; Vakulenko et al., 2019; Diefenbach et al., 2020; Abdelaziz et al., 2021). Facts are retrieved from a KB through structured queries (in a query language such as SPARQL), which often contain multiple triples that ∗Equal contribution, correspondence to Pavan Ka- panipathi (kapanipa@us.ibm.com), Ibrahim Abdelaziz (ibrahim.abdelaziz1@ibm.com), Srinivas Ravishankar (srini@ibm.com) represent the steps or antecedents required for ob- taining the answer. This enables a transparent and self-explanatory form of QA, meaning that inter- mediate symbolic representations capture some of the steps from natural language question to answer. With the rise of neural networks in NLP, various KBQA models approach the task in an end-to-end manner. Many of these approaches formulate text- to-query-language as sequence-to-sequence prob- lem, and thus require sufficient examples of paired natural language and target representation pairs. However, labeling large amounts of data for KBQA is challenging, either due to the requirement of expert knowledge (Usbeck et al., 2017), or arti- facts introduced during automated creation (Trivedi et al., 2017). Real-world scenarios require solving complex multi-hop questions i.e. secondary un- knowns within a main question and questions em- ploying unusual expressions. Pipeline approaches can delegate language understanding to pre-trained semantic parsers, which mitigates the data prob- lem, but are considered to suffer from error prop- agation. However, the performance of semantic parsers …

Content truncated to 500 words