MLPQ: A Dataset for Path Question Answering over Multilingual Knowledge Graphs

Abstract

Knowledge Graph-based Multilingual Question Answering (KG-MLQA), as one of the essential subtasks in Knowledge Graph-based Question Answering (KGQA), emphasizes that questions on the KGQA task can be expressed in different languages to solve the lexical gap between questions and knowledge graph(s). However, the existing KG-MLQA works mainly focus on the semantic parsing of multilingual questions but ignore the questions that require integrating information from cross-lingual knowledge graphs (CLKG). This paper extends KG-MLQA to Cross-lingual KG-based multilingual Question Answering (CLKGQA) and constructs the first CLKGQA dataset over multilingual DBpedia named MLPQ, which contains 300K questions in English, Chinese, and French. We further propose a novel KG sampling algorithm for KG construction, making the MLPQ support the research of different types of methods. To evaluate the dataset, we put forward a general question answering workflow whose core idea is to transform CLKGQA into KG-MLQA. We first use the Entity Alignment (EA) model to merge CLKG into a single KG and get the answer to the question by the Multi-hop QA model combined with the Multilingual pre-training model. By instantiating the above QA workflow, we establish two baseline models for MLPQ, one of which uses Google translation to obtain alignment entities, and the other adopts the recent EA model. Experiments show that the baseline models are insufficient to obtain the ideal performances on CLKGQA. Moreover, the availability of our benchmark contributes to the community of question answering and entity alignment.