Understanding how people use natural language to ask for recommendations

Jie Kang; Kyle Condiff; Shuo Chang; Joseph A. Konstan; Loren Terveen; F. Maxwell Harper

doi:10.1145/3109859.3109873

Understanding how people use natural language to ask for recommendations

Jie Kang, Kyle Condiff, Shuo Chang, Joseph A. Konstan, Loren Terveen, F. Maxwell Harper

Computer Science and Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

52 Scopus citations

Abstract

The technical barriers for conversing with recommender systems using natural language are vanishing. Already, there are commercial systems that facilitate interactions with an AI agent. For instance, it is possible to say "what should I watch" to an Apple TV remote to get recommendations. In this research, we investigate how users initially interact with a new natural language recommender to deepen our understanding of the range of inputs that these technologies can expect. We deploy a natural language interface to a recommender system, we observe users' first interactions and follow-up queries, and we measure the differences between speaking- and typing-based interfaces.We employ qualitative methods to derive a categorization of users' first queries (objective, subjective, and navigation) and follow-up queries (refine, reformulate, start over). We employ quantitative methods to determine the differences between speech and text, finding that speech inputs are typically longer and more conversational.

Original language	English (US)
Title of host publication	RecSys 2017 - Proceedings of the 11th ACM Conference on Recommender Systems
Publisher	Association for Computing Machinery, Inc
Pages	229-237
Number of pages	9
ISBN (Electronic)	9781450346528
DOIs	https://doi.org/10.1145/3109859.3109873
State	Published - Aug 27 2017
Event	11th ACM Conference on Recommender Systems, RecSys 2017 - Como, Italy Duration: Aug 27 2017 → Aug 31 2017

Publication series

Name	RecSys 2017 - Proceedings of the 11th ACM Conference on Recommender Systems

Conference

Conference	11th ACM Conference on Recommender Systems, RecSys 2017
Country/Territory	Italy
City	Como
Period	8/27/17 → 8/31/17

Bibliographical note

Funding Information:
This material is based on work supported by the National Science Foundation under grants IIS-0964695, IIS-1017697, IIS-1111201, IIS-1210863, and IIS-1218826, and by a grant from Google.

Publisher Copyright:
© 2017 Association for Computing Machinery.

Keywords

Critiquing
Natural language recommenders
Qualitative methods
Recommender systems
Speech
User study
Virtual assistants
Voice

Access

10.1145/3109859.3109873

OpenUrl availability

Full text

Cite this

Kang, J., Condiff, K., Chang, S., Konstan, J. A., Terveen, L., & Harper, F. M. (2017). Understanding how people use natural language to ask for recommendations. In RecSys 2017 - Proceedings of the 11th ACM Conference on Recommender Systems (pp. 229-237). (RecSys 2017 - Proceedings of the 11th ACM Conference on Recommender Systems). Association for Computing Machinery, Inc. https://doi.org/10.1145/3109859.3109873

Understanding how people use natural language to ask for recommendations. / Kang, Jie; Condiff, Kyle; Chang, Shuo et al.
RecSys 2017 - Proceedings of the 11th ACM Conference on Recommender Systems. Association for Computing Machinery, Inc, 2017. p. 229-237 (RecSys 2017 - Proceedings of the 11th ACM Conference on Recommender Systems).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Kang, J, Condiff, K, Chang, S, Konstan, JA , Terveen, L & Harper, FM 2017, Understanding how people use natural language to ask for recommendations. in RecSys 2017 - Proceedings of the 11th ACM Conference on Recommender Systems. RecSys 2017 - Proceedings of the 11th ACM Conference on Recommender Systems, Association for Computing Machinery, Inc, pp. 229-237, 11th ACM Conference on Recommender Systems, RecSys 2017, Como, Italy, 8/27/17. https://doi.org/10.1145/3109859.3109873

Kang J, Condiff K, Chang S, Konstan JA , Terveen L, Harper FM. Understanding how people use natural language to ask for recommendations. In RecSys 2017 - Proceedings of the 11th ACM Conference on Recommender Systems. Association for Computing Machinery, Inc. 2017. p. 229-237. (RecSys 2017 - Proceedings of the 11th ACM Conference on Recommender Systems). doi: 10.1145/3109859.3109873

@inproceedings{f4bde4d988714b519c0c60d1575ba485,

title = "Understanding how people use natural language to ask for recommendations",

abstract = "The technical barriers for conversing with recommender systems using natural language are vanishing. Already, there are commercial systems that facilitate interactions with an AI agent. For instance, it is possible to say {"}what should I watch{"} to an Apple TV remote to get recommendations. In this research, we investigate how users initially interact with a new natural language recommender to deepen our understanding of the range of inputs that these technologies can expect. We deploy a natural language interface to a recommender system, we observe users' first interactions and follow-up queries, and we measure the differences between speaking- and typing-based interfaces.We employ qualitative methods to derive a categorization of users' first queries (objective, subjective, and navigation) and follow-up queries (refine, reformulate, start over). We employ quantitative methods to determine the differences between speech and text, finding that speech inputs are typically longer and more conversational.",

keywords = "Critiquing, Natural language recommenders, Qualitative methods, Recommender systems, Speech, User study, Virtual assistants, Voice",

author = "Jie Kang and Kyle Condiff and Shuo Chang and Konstan, {Joseph A.} and Loren Terveen and Harper, {F. Maxwell}",

note = "Funding Information: This material is based on work supported by the National Science Foundation under grants IIS-0964695, IIS-1017697, IIS-1111201, IIS-1210863, and IIS-1218826, and by a grant from Google. Publisher Copyright: {\textcopyright} 2017 Association for Computing Machinery.; 11th ACM Conference on Recommender Systems, RecSys 2017 ; Conference date: 27-08-2017 Through 31-08-2017",

year = "2017",

month = aug,

day = "27",

doi = "10.1145/3109859.3109873",

language = "English (US)",

series = "RecSys 2017 - Proceedings of the 11th ACM Conference on Recommender Systems",

publisher = "Association for Computing Machinery, Inc",

pages = "229--237",

booktitle = "RecSys 2017 - Proceedings of the 11th ACM Conference on Recommender Systems",

}

TY - GEN

T1 - Understanding how people use natural language to ask for recommendations

AU - Kang, Jie

AU - Condiff, Kyle

AU - Chang, Shuo

AU - Konstan, Joseph A.

AU - Terveen, Loren

AU - Harper, F. Maxwell

N1 - Funding Information: This material is based on work supported by the National Science Foundation under grants IIS-0964695, IIS-1017697, IIS-1111201, IIS-1210863, and IIS-1218826, and by a grant from Google. Publisher Copyright: © 2017 Association for Computing Machinery.

PY - 2017/8/27

Y1 - 2017/8/27

N2 - The technical barriers for conversing with recommender systems using natural language are vanishing. Already, there are commercial systems that facilitate interactions with an AI agent. For instance, it is possible to say "what should I watch" to an Apple TV remote to get recommendations. In this research, we investigate how users initially interact with a new natural language recommender to deepen our understanding of the range of inputs that these technologies can expect. We deploy a natural language interface to a recommender system, we observe users' first interactions and follow-up queries, and we measure the differences between speaking- and typing-based interfaces.We employ qualitative methods to derive a categorization of users' first queries (objective, subjective, and navigation) and follow-up queries (refine, reformulate, start over). We employ quantitative methods to determine the differences between speech and text, finding that speech inputs are typically longer and more conversational.

AB - The technical barriers for conversing with recommender systems using natural language are vanishing. Already, there are commercial systems that facilitate interactions with an AI agent. For instance, it is possible to say "what should I watch" to an Apple TV remote to get recommendations. In this research, we investigate how users initially interact with a new natural language recommender to deepen our understanding of the range of inputs that these technologies can expect. We deploy a natural language interface to a recommender system, we observe users' first interactions and follow-up queries, and we measure the differences between speaking- and typing-based interfaces.We employ qualitative methods to derive a categorization of users' first queries (objective, subjective, and navigation) and follow-up queries (refine, reformulate, start over). We employ quantitative methods to determine the differences between speech and text, finding that speech inputs are typically longer and more conversational.

KW - Critiquing

KW - Natural language recommenders

KW - Qualitative methods

KW - Recommender systems

KW - Speech

KW - User study

KW - Virtual assistants

KW - Voice

UR - http://www.scopus.com/inward/record.url?scp=85030449678&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85030449678&partnerID=8YFLogxK

U2 - 10.1145/3109859.3109873

DO - 10.1145/3109859.3109873

M3 - Conference contribution

AN - SCOPUS:85030449678

T3 - RecSys 2017 - Proceedings of the 11th ACM Conference on Recommender Systems

SP - 229

EP - 237

BT - RecSys 2017 - Proceedings of the 11th ACM Conference on Recommender Systems

PB - Association for Computing Machinery, Inc

T2 - 11th ACM Conference on Recommender Systems, RecSys 2017

Y2 - 27 August 2017 through 31 August 2017

ER -

Understanding how people use natural language to ask for recommendations

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access

OpenUrl availability

Other files and links

Fingerprint

Cite this