Knowledge Base Question Answering through Recursive Hypergraphs

Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics, pages 448–454 April 19 - 23, 2021. ©2021 Association for Computational Linguistics 448 Knowledge Base Question Answering through Recursive Hypergraphs Naganand Yadati1, Dayanidhi R2, Vaishnavi S2, Indira G2, Srinidhi G2 1 Indian Institute of Science, Bangalore 2 Channabasaveshwara Institute of Technology, Tumkur naganand@iisc.ac.in {dayanidhirs.12, vaishnavisathish98, indiragowtham99, srinidhi.g71}@gmail.com Abstract Knowledge Base Question Answering (KBQA) is the problem of predicting an answer for a factoid question over a given knowledge base (KB). Answering questions typically requires reasoning over multiple links in the given KB. Humans tend to answer questions by grouping different objects to perform reasoning over acquired knowledge. Hypergraphs provide a natural tool to model group relationships. In this work, inspired by typical human intelligence, we propose a new method for KBQA based on hypergraphs. Existing methods for KBQA, though effective, do not explicitly incorporate the recursive relational group structure in the given KB. Our method, which we name RecHyperNet (Recursive Hypergraph Network), exploits a new way of modelling KBs through recursive hypergraphs to organise such group relation- ships in KBs. Experiments on multiple KBQA benchmarks demonstrate the effectiveness of the proposed RecHyperNet. We have released the code. 1 Introduction Knowledge Base Question Answering (KBQA) (Xu et al., 2016)), a task that tests the ability of a machine to understand knowledge like a human, is a challenging, central, and popular task in natu- ral language processing. KBQA is the problem of predicting an answer for a factoid question over a given knowledge base (KB) containing facts such as (Inception, written by, Christopher Nolan). An- swering questions typically requires reasoning over multiple links in the given KB (Zhang et al., 2018). When a typical human answers a question (e.g. What are the genres of movies written by Christo- pher Nolan?), they tend to group objects (e.g. movies written by Christopher Nolan such as In- ception, Interstellar, etc. are grouped together) over their acquired knowledge. Hypergraphs are mathe- matical tools that naturally encode group relation- ships. Directed hypergraphs have been recently used to model KBs for KBQA (Han et al., 2020) but we argue that they fail to naturally model group relationships without loss of information. Spcifi- cally, two KB triples such as (Inception, written by, Christopher Nolan) and (Person of Interest, writ- ten by, Jonathan Nolan) would be modelled by a directed hyperedge {Inception, Person of Interest} -> {Christopher Nolan, Jonathan Nolan} clearly resulting in loss of information (not clear who di- rected which). Though objects of similar types are grouped together and directions are encoded, di- rected hypergraphs fail to model KBs without loss of information. We argue that recursive hypergraphs (Joslyn and Nowak, 2017; Menezes and Roth, 2019) provide a flexible way to model KBs without loss of informa- tion. Specifically, all movies written by Christopher Nolan are grouped together and those written by Jonathan Nolan are separately grouped together. To summarise, we make the following contributions • We model KBs as recursive hypergraphs. To …

Content truncated to 500 words