Joint Spatio-Textual Reasoning for Answering Tourism Questions

TL;DR

This paper introduces a novel joint spatio-textual reasoning model that effectively combines geographic and textual information to improve tourism question answering and POI recommendation accuracy.

Contribution

It presents the first integrated model that combines spatial and textual reasoning for tourism questions, enhancing answer accuracy over existing methods.

Findings

Significant performance improvements over baseline models.

Effective integration of geo-spatial and textual data.

Validated on real-world POI recommendation tasks.

Abstract

Our goal is to answer real-world tourism questions that seek Points-of-Interest (POI) recommendations. Such questions express various kinds of spatial and non-spatial constraints, necessitating a combination of textual and spatial reasoning. In response, we develop the first joint spatio-textual reasoning model, which combines geo-spatial knowledge with information in textual corpora to answer questions. We first develop a modular spatial-reasoning network that uses geo-coordinates of location names mentioned in a question, and of candidate answer POIs, to reason over only spatial constraints. We then combine our spatial-reasoner with a textual reasoner in a joint model and present experiments on a real world POI recommendation task. We report substantial improvements over existing models with-out joint spatio-textual reasoning.