Sammanfattning
search by providing feedback for estimates of search intents.
The estimated intents are visualized for interaction on an Intent
Radar, a novel visual interface that organizes intents onto a radial
layout where relevant intents are close to the center of the visualization
and similar intents have similar angles. The user can give
feedback on the visualized intents, from which the system learns
and visualizes improved intent estimates. We systematically evaluated
the effect of the interactive intent modeling in a mixed-method
task-based information seeking setting with 30 users, where we
compared two interface variants for interactive intent modeling,
namely intent radar and a simpler list-based interface, to a conventional
search system. The results show that interactive intent
modeling significantly improves users’ task performance and the
quality of retrieved information.
| Originalspråk | engelska |
|---|---|
| Titel på gästpublikation | ACM International Conference on Information and Knowledge Management (CIKM 2013) |
| Antal sidor | 6 |
| Utgivningsort | New York |
| Förlag | ACM |
| Utgivningsdatum | 2013 |
| Sidor | 1759-1764 |
| ISBN (tryckt) | 978-1-4503-2263-8 |
| DOI | |
| Status | Publicerad - 2013 |
| MoE-publikationstyp | A4 Artikel i en konferenspublikation |
| Evenemang | ACM international conference on Conference on information & knowledge management - Burlingame, Förenta Staterna (USA) Varaktighet: 27 nov 2013 → 1 dec 2013 Konferensnummer: 22 |
Vetenskapsgrenar
- 113 Data- och informationsvetenskap
Citera det här
}
Directing Exploratory Search with Interactive Intent Modeling. / Ruotsalo, Tuukka; Jaakko, Peltonen; Eugster, Manuel J. A; Glowacka, Dorota; Konyushkova, Ksenia; Athukorala, Kumaripaba; Kosunen, Ilkka; Reijonen, Aki; Myllymäki, Petri; Jacucci, Giulio; Kaski, Samuel.
ACM International Conference on Information and Knowledge Management (CIKM 2013). New York : ACM, 2013. s. 1759-1764.Forskningsoutput: Kapitel i bok/rapport/konferenshandling › Konferensbidrag › Vetenskaplig › Peer review
TY - GEN
T1 - Directing Exploratory Search with Interactive Intent Modeling
AU - Ruotsalo, Tuukka
AU - Jaakko, Peltonen
AU - Eugster, Manuel J. A
AU - Glowacka, Dorota
AU - Konyushkova, Ksenia
AU - Athukorala, Kumaripaba
AU - Kosunen, Ilkka
AU - Reijonen, Aki
AU - Myllymäki, Petri
AU - Jacucci, Giulio
AU - Kaski, Samuel
N1 - Volume: Proceeding volume:
PY - 2013
Y1 - 2013
N2 - We introduce interactive intent modeling, where the user directs exploratorysearch by providing feedback for estimates of search intents.The estimated intents are visualized for interaction on an IntentRadar, a novel visual interface that organizes intents onto a radiallayout where relevant intents are close to the center of the visualizationand similar intents have similar angles. The user can givefeedback on the visualized intents, from which the system learnsand visualizes improved intent estimates. We systematically evaluatedthe effect of the interactive intent modeling in a mixed-methodtask-based information seeking setting with 30 users, where wecompared two interface variants for interactive intent modeling,namely intent radar and a simpler list-based interface, to a conventionalsearch system. The results show that interactive intentmodeling significantly improves users’ task performance and thequality of retrieved information.
AB - We introduce interactive intent modeling, where the user directs exploratorysearch by providing feedback for estimates of search intents.The estimated intents are visualized for interaction on an IntentRadar, a novel visual interface that organizes intents onto a radiallayout where relevant intents are close to the center of the visualizationand similar intents have similar angles. The user can givefeedback on the visualized intents, from which the system learnsand visualizes improved intent estimates. We systematically evaluatedthe effect of the interactive intent modeling in a mixed-methodtask-based information seeking setting with 30 users, where wecompared two interface variants for interactive intent modeling,namely intent radar and a simpler list-based interface, to a conventionalsearch system. The results show that interactive intentmodeling significantly improves users’ task performance and thequality of retrieved information.
KW - 113 Computer and information sciences
U2 - 10.1145/2505515.2505644
DO - 10.1145/2505515.2505644
M3 - Conference contribution
SN - 978-1-4503-2263-8
SP - 1759
EP - 1764
BT - ACM International Conference on Information and Knowledge Management (CIKM 2013)
PB - ACM
CY - New York
ER -