Timezone: »

Spotlight
Knowledge Base Question Answering by Case-based Reasoning over Subgraphs
Rajarshi Das · Ameya Godbole · Ankita Rajaram Naik · Elliot Tower · Manzil Zaheer · Hannaneh Hajishirzi · Robin Jia · Andrew McCallum

Tue Jul 19 11:45 AM -- 11:50 AM (PDT) @ Ballroom 1 & 2
Question answering (QA) over knowledge bases (KBs) is challenging because of the diverse, essentially unbounded, types of reasoning patterns needed.However, we hypothesize in a large KB, reasoning patterns required to answer a query type reoccur for various entities in their respective subgraph neighborhoods.Leveraging this structural similarity between local neighborhoods of different subgraphs, we introduce a semiparametric model (CBR-SUBG) with (i) a nonparametric component that for each query, dynamically retrieves other similar $k$-nearest neighbor (KNN) training queries along with query-specific subgraphs and (ii) a parametric component that is trained to identify the (latent) reasoning patterns from the subgraphs of KNN queries and then apply them to the subgraph of the target query. We also propose an adaptive subgraph collection strategy to select a query-specific compact subgraph, allowing us to scale to full Freebase KB containing billions of facts. We show that CBR-SUBG can answer queries requiring subgraph reasoning patterns and performs competitively with the best models on several KBQA benchmarks. Our subgraph collection strategy also produces more compact subgraphs (e.g. 55\% reduction in size for WebQSP while increasing answer recall by 4.85\%)\footnote{Code, model, and subgraphs are available at \url{https://github.com/rajarshd/CBR-SUBG}}.

#### Author Information

##### Ankita Rajaram Naik (University of Massachusetts Amherst)

I'm a final year master student at the University of Massachusetts Amherst with 4+ years of experience working in Natural Language Processing. I am actively looking for full-time positions in Natural Language Processing. My most recent projects are focused on Knowledge Graph Reasoning for Question Answering and Graph Completion; Retrieval models for knowledge bases and tables; and Biomedical Applications of NLP with research publications in leading conference such as AAAI, NAACL and ICML. At UMass Amherst, I primarily worked with Prof. Andrew Mccallum and Rajarshi Das at Information Extraction and Synthesis Lab (IESL). I also interned with the Knowledge Induction Group at IBM Research AI under the mentorship of Michael Glass and Alfio Gloizzo. Moreover, I have also worked with researchers at GE Healthcare and Walmart while receiving my MS in Computer Science.