Multi-Stage Conversational Passage Retrieval: An Approach to Fusing Term Importance Estimation and Neural Query Rewriting

Conversational search plays a vital role in conversational information seeking. As queries in information seeking dialogues are ambiguous for traditional ad hoc information retrieval (IR) systems due to the coreference and omission resolution problems inherent in natural language dialogue, resolving these ambiguities is crucial. In this article, we tackle conversational passage retrieval, an important component of conversational search, by addressing query ambiguities with query reformulation integrated into a multi-stage ad hoc IR system. Specifically, we propose two conversational query reformulation (CQR) methods: (1) term importance estimation and (2) neural query rewriting. For the former, we expand conversational queries using important terms extracted from the conversational context with frequency-based signals. For the latter, we reformulate conversational queries into natural, stand-alone, human-understandable queries with a pretrained sequence-to-sequence model. Detailed analyses of the two CQR methods are provided quantitatively and qualitatively, explaining their advantages, disadvantages, and distinct behaviors. Moreover, to leverage the strengths of both CQR methods, we propose combining their output with reciprocal rank fusion, yielding state-of-the-art retrieval effectiveness, 30% improvement in terms of NDCG@3 compared to the best submission of Text REtrieval Conference (TREC) Conversational Assistant Track (CAsT) 2019.

Use Learnable Knowledge Graph in Dialogue System for Visually Impaired Macro Navigation

Dialogue in natural language is the most important communication method for the visually impaired. Therefore, the dialogue system is the main subsystem in the visually impaired navigation system. The purpose of the dialogue system is to understand the user’s intention, gradually establish context through multiple conversations, and finally provide an accurate destination for the navigation system. We use the knowledge graph as the basis of reasoning in the dialogue system, and then update the knowledge graph so that the system gradually conforms to the user’s background. Based on the experience of using the knowledge graph in the navigation system of the visually impaired, we expect that the same framework can be applied to more fields in order to improve the practicality of natural language dialogue in human–computer interaction.

Context-Sensitive Spatial Interaction and Ambient Control

In this chapter, we first briefly introduce the setting: mobility assistants (the wheelchair Rolland and iWalker) and smart environment control in the Bremen Ambient Assisted Living Lab. In several example scenarios, we then outline our contributions to the state of the art, focussing on spatial knowledge representation, reasoning and spatial interaction (multi-modal, but with special emphasis on natural language dialogue) between three partners: the user, a mobility assistant, and the smart environment.

Developing Natural Language Enabled Games in SCXML

The World Wide Web Consortium (W3C) has a standard in the pipeline – called SCXML – that may turn out to be very suitable for the design and implementation of games, in particular games featuring (possibly multimodal) natural language dialogue. We see three main reasons why SCXML may be a good fit for the game industry: 1) SCXML is all about statecharts – a powerful extension of finite-state machines – and we argue that statecharts has the right kind of expressivity for game design and de-velopment, 2) SCXML is an XML dialect (soon to be) endorsed by the W3C, and will thus become a part of a web infrastructure comprising speech technology and telephony, as well as other useful technologies for building games of certain genres, and 3) SCXML is designed for extensibility and it appears that it would be fairly straightforward – and very worthwhile – to build a game oriented extension (“profile”) around the SCXML core. The paper also presents an experimental implementation of SCXML, accessible from a user-friendly web-interface.

Advances in Education and Consumer Health Informatics

SummaryTo summarize current excellent research in the field of education and consumer health informatics.Synopsis of the articles on education and consumer health informatics selected for the IMIA Yearbook of Medical Informatics 2007.The consumer health informatics papers selected reflect the growing importance of communication and information retrieval systems in health care both for patients and professionals. Sound exemplary studies point out both the benefit for patients as well as the economic advantages of such systems. On the education sector, an intelligent tutoring system for medical students based on natural language dialogue serves as an example for the advancement and refinement of methods.The selected articles demonstrate the potential of advanced communication and information systems in health care. The physician-patient relationship though must not be affected by the introduction of these systems in order to ensure acceptance by both patients and physicians. Therefore these tools should be used in addition to current processes, and not as a replacement.