A Semantic Parsing and Reasoning-Based Approach to Knowledge Base Question Answering

Abstract

Knowledge Base Question Answering (KBQA) is a task where existing techniques have faced significant challenges, such as the need for complex question understanding, reasoning, and large training datasets. In this work, we demonstrate Deep Thinking Question Answering (DTQA), a semantic parsing and reasoning-based KBQA system. DTQA (1) integrates multiple, reusable modules that are trained specifically for their individual tasks (e.g. semantic parsing, entity linking, and relationship linking), eliminating the need for end-to-end KBQA training data; (2) leverages semantic parsing and a reasoner for improved question understanding. DTQA is a system of systems that achieves state-of-the-art performance on two popular KBQA datasets.

Boosting Question Answering Systems with Multi-Label Classification Techniques

“…Suppose the SPARQL of the question has only one set of RDF triples. In that case, the accuracy rate is 73.83%, i.e., the number of questions correctly 9 presents the End-to-End performance of the proposed system on the LC-QuAD test set and compares it with two other QA systems, i.e., QAMP [30] and DTQA [31], respectively.…”

Enhancing SPARQL Query Performance With Recurrent Neural Networks

“…Table 7 presents the End-to-End performance of the proposed system on the LC-QuAD test set and compares it with two other QA systems, i.e., QAMP [33] and DTQA [34], respectively. Table 9 summarizes the number of questions in the SPARQL syntax of 604 questions in one, two, and three RDF triples, and the number of questions correctly predicted using the Ensemble BR model.…”

“…The QALD metrics (Precision, Recall, F-measure) are used to measure the performances of QALD-7, QALD-8, and QALD-9 that are shown in Tables 8. The proposed system was compared to other systems, i.e., gAnswer2 [35], Wdaqua [11], QAwizard [14], Light-QAwizard [15], and DTQA [34]. gAnswer2 and Wdaqua have won the QALD competition in the past.…”