Informing the Design of Spoken Conversational Search

Trippas, Johanne R.; Spina, Damiano; Cavedon, Lawrence; Joho, Hideo; Sanderson, Mark

doi:10.1145/3176349.3176387

Cited by 110 publications

(79 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For the strategy of "direct replacement" where the participants wanted to replace one name entity with another name entity, it would not be a problem if the system is a real human being (the wizard), as in the settings of this study and in the study by Trippas et al (2018). Among the three main strategies, "specific operation" and "partial repeat" could clearly tell the system how to modify the existing query.…”

Section: Patterns Of Partial Query Modificationmentioning

confidence: 96%

“…In that case, in the next turn the participant had to speak the complete query. For the strategy of "direct replacement" where the participants wanted to replace one name entity with another name entity, it would not be a problem if the system is a real human being (the wizard), as in the settings of this study and in the study by Trippas et al (2018). It may be challenging to a computer system when the two name entities are weakly linked.…”

Section: Patterns Of Partial Query Modificationmentioning

confidence: 97%

“…Voice queries were more likely to be about local service, audio-video sites, and question-answer sites, and less likely to be about sensitive subjects, such as adult site, or interaction intensive tasks, such as social network site (Guy, 2016;Schalkwyk et al, 2010). Trippas, Spina, Cavedon, Joho, and Sanderson (2018) explored user search behavior in an audio only search system, where the users performed search and retrieved the results by talking with an intermediary. They found that users preferred to issue a spoken query when they had insufficient access to a keyboard.…”

Section: Query Behavior and Query Reformulation With Voice Inputmentioning

confidence: 99%

See 2 more Smart Citations

Examining users' partial query modification patterns in voice search

Yuan

2019

Asso for Info Science & Tech

View full text Add to dashboard Cite

This article investigates how to improve the effectiveness of voice search systems. Earlier research found that participants employed voice search much less frequently than keyboard search. The main reasons that participants disliked voice search are system mistakes and the inability to modify queries. In keyboard search, query reformulation is facilitated by partial query modification, which is not supported by most of the current voice search systems. Consequently, users need to speak the complete query in voice search even with only minor changes. This article focuses on examining partial query modification during voice search through a Wizard of Oz user experiment. It examines if users would prefer partial query modification and how they perform it in voice search. Thirty-two participants participated in the experiment. Results indicated that when given the opportunity, the users performed more partial query modifications than complete queries. Common partial query modification strategies and patterns emerged from the experiment. The results can be used to improve the voice search system design and benefit the research community in general. System implications and future work were discussed.

show abstract

Section: Patterns Of Partial Query Modificationmentioning

confidence: 96%

Section: Patterns Of Partial Query Modificationmentioning

confidence: 97%

Section: Query Behavior and Query Reformulation With Voice Inputmentioning

confidence: 99%

See 1 more Smart Citation

Examining users' partial query modification patterns in voice search

Yuan

2019

Asso for Info Science & Tech

View full text Add to dashboard Cite

show abstract

“…Trippas et al [53] suggested that existing information seeking models do not suffice for the increased interactivity, complexity, or the agency of SCS systems.…”

Section: Introductionmentioning

confidence: 99%

Towards a model for spoken conversational search

Trippas

Spina

Thomas³

et al. 2020

Information Processing & Management

Self Cite

View full text Add to dashboard Cite

Conversation is the natural mode for information exchange in daily life, a spoken conversational interaction for search input and output is a logical format for information seeking. However, the conceptualisation of user-system interactions or information exchange in spoken conversational search (SCS) has not been explored. The first step in conceptualising SCS is to understand the conversational moves used in an audio-only communication channel for search. This paper explores conversational actions for the task of search. We define a qualitative methodology for creating conversational datasets, propose analysis protocols, and develop the SCSdata. Furthermore, we use the SCSdata to create the first annotation schema for SCS: the SCoSAS, enabling us to investigate interactivity in SCS. We further establish that SCS needs to incorporate interactivity and pro-activity to overcome the complexity that the information seeking process in an audio-only channel poses. In summary, this exploratory study unpacks the breadth of SCS. Our results highlight the need for integrating discourse in future SCS models and contributes the advancement in the formalisation of SCS models and the design of SCS systems.

show abstract

“…The tasks can range from as simple as setting an alarm to more complex cases like health advice. Conversational assistants have been employed for information seeking [35] and recommendation [34] among other information systems applications. Moreover, such systems can be used as home assistants, e.g., Alexa and Google Home; or be integrated in a smartphone [1,2] or wearable devices, e.g., Apple Siri.…”

Section: Introductionmentioning

confidence: 99%

Harnessing Evolution of Multi-Turn Conversations for Effective Answer Retrieval

Aliannejadi

Chakraborty

Ríssola

et al. 2020

Proceedings of the 2020 Conference on Human Information Interaction and Retrieval

View full text Add to dashboard Cite

With the improvements in speech recognition and voice generation technologies over the last years, a lot of companies have sought to develop conversation understanding systems that run on mobile phones or smart home devices through natural language interfaces. Conversational assistants, such as Google Assistant™ and Microsoft Cortana™, can help users to complete various types of tasks. This requires an accurate understanding of the user's information need as the conversation evolves into multiple turns. Finding relevant context in a conversation's history is challenging because of the complexity of natural language and the evolution of a user's information need. In this work, we present an extensive analysis of language, relevance, dependency of user utterances in a multi-turn information-seeking conversation. To this aim, we have annotated relevant utterances in the conversations released by the TREC CaST 2019 track. The annotation labels determine which of the previous utterances in a conversation can be used to improve the current one. Furthermore, we propose a neural utterance relevance model based on BERT fine-tuning, outperforming competitive baselines. We study and compare the performance of multiple retrieval models, utilizing different strategies to incorporate the user's context. The experimental results on both classification and retrieval tasks show that our proposed approach can effectively identify and incorporate the conversation context. We show that processing the current utterance using the predicted relevant utterance leads to a 38% relative improvement in terms of nDCG@20. Finally, to foster research in this area, we have released the dataset of the annotations.

show abstract

Informing the Design of Spoken Conversational Search

Cited by 110 publications

References 35 publications

Examining users' partial query modification patterns in voice search

Examining users' partial query modification patterns in voice search

Towards a model for spoken conversational search

Harnessing Evolution of Multi-Turn Conversations for Effective Answer Retrieval

Contact Info

Product

Resources

About