Benchmark and Neural Architecture for Conversational Entity Retrieval from a Knowledge Graph

Mona Zamiri, Yao Qiang, Fedor Nikolaev, Dongxiao Zhu, Alexander Kotov

Keywords: Conversational IR, Entity Retrieval, Knowledge Graphs, Deep Learning, IR Benchmarks

Abstract: This paper introduces a novel information retrieval (IR) task of Conversational Entity Retrieval from a Knowledge Graph (CER-KG). CER-KG extends non-conversational entity retrieval from a knowledge graph (KG) to the conversational scenario. The user queries in CER-KG dialog turns may rely on the results of the preceding turns, which are KG entities. Similar to the conversational document IR, CER-KG can be viewed as a sequence of interrelated ranking tasks. To enable future research on CER-KG, we created QBLink-KG, a publicly available benchmark that was adapted from QBLink, a benchmark for text-based conversational reading comprehension of Wikipedia. In our initial approach to CER-KG, we experimented with Transformer- and LSTM-based dialog context encoders in combination with the Neural Architecture for Conversational Entity Retrieval (NACER), our proposed feature-based neural architecture for entity ranking in CER-KG. NACER computes the ranking score of a candidate KG entity by taking into account a large number of lexical and semantic matching signals between various KG components in its neighborhood, such as entities, categories, and literals, as well as entities in the results of the preceding turns in dialog history. The experimental results for our initial approach to CER-KG reveal the key challenges of the proposed task along with the possible future directions for developing new approaches to it.

Track: Search

Submission Guidelines Scope: Yes

Submission Guidelines Blind: Yes

Submission Guidelines Format: Yes

Submission Guidelines Limit: Yes

Submission Guidelines Authorship: Yes

Student Author: Yes

Submission Number: 2152