Skip to main content
Disciplines
Automotive Business IT + Informatics Construction + Real Estate Electrical Engineering + Electronics Energy + Sustainability Insurance + Risk Finance + Banking Management + Leadership Marketing + Sales Mechanical Engineering + Materials
Events
DE
EN
Books
Journals
Topic Page
Marketing
Start single access now
Access for companies
EXTENDED SEARCH
Log in
Top
Published in:
A Character Expression Model Affecting Spoken Dialogue Behaviors Read chapter Read first chapter

2021 | OriginalPaper | Chapter

Packing, Stacking, and Tracking: An Empirical Study of Online User Adaptation

Authors : Jean-Sébastien Laperrière, Darryl Lam, Kotaro Funakoshi

Published in: Conversational Dialogue Systems for the Next Decade

Publisher: Springer Singapore

Log in

Activate our intelligent search to find suitable subject content or patents.

search-config
loading …

Abstract

This paper explores the application of expert tracking to online user adaptation based on a set of basic predictors in order to classify input in multimodal interaction settings. We compare the performances of this approach to other common approaches that aggregate multiple predictors, like stacking and voting. To realistically assess the performances of algorithms that require feedback, we added noise to feedback to simulate an imperfect system. Using two datasets, we obtained inconsistent results. With one dataset, expert tracking was the best option for short interactions, but with the other dataset, it was outperformed by other algorithms. In contrast, voting worked surprisingly well. On the basis of these results, we discuss implications and future directions.

Dont have a licence yet? Then find out more about our products and how to get one now:

Springer Professional "Wirtschaft+Technik"

Online-Abonnement

Mit Springer Professional "Wirtschaft+Technik" erhalten Sie Zugriff auf:

  • über 102.000 Bücher
  • über 537 Zeitschriften

aus folgenden Fachgebieten:

  • Automobil + Motoren
  • Bauwesen + Immobilien
  • Business IT + Informatik
  • Elektrotechnik + Elektronik
  • Energie + Nachhaltigkeit
  • Finance + Banking
  • Management + Führung
  • Marketing + Vertrieb
  • Maschinenbau + Werkstoffe
  • Versicherung + Risiko

Jetzt Wissensvorsprung sichern!

inform now

Springer Professional "Technik"

Online-Abonnement

Mit Springer Professional "Technik" erhalten Sie Zugriff auf:

  • über 67.000 Bücher
  • über 390 Zeitschriften

aus folgenden Fachgebieten:

  • Automobil + Motoren
  • Bauwesen + Immobilien
  • Business IT + Informatik
  • Elektrotechnik + Elektronik
  • Energie + Nachhaltigkeit
  • Maschinenbau + Werkstoffe




 

Jetzt Wissensvorsprung sichern!

inform now

Springer Professional "Wirtschaft"

Online-Abonnement

Mit Springer Professional "Wirtschaft" erhalten Sie Zugriff auf:

  • über 67.000 Bücher
  • über 340 Zeitschriften

aus folgenden Fachgebieten:

  • Bauwesen + Immobilien
  • Business IT + Informatik
  • Finance + Banking
  • Management + Führung
  • Marketing + Vertrieb
  • Versicherung + Risiko




Jetzt Wissensvorsprung sichern!

inform now
previous chapter Response Generation to Out-of-Database Questions for Example-Based Dialogue Systems
next chapter On the Use of Phonotactic Vector Representations with FastText for Language Identification
Footnotes
1
We tested another recent tracking algorithm, CBCE  [10]. In a simple simulated situation, we confirmed the superiority of CBCE to Bousquet’s, as was claimed. However, in none of the settings examined in this paper did CBCE outperform Bousquet’s. Therefore, we omit the results with CBCE due to space limitations.
 
2
While the dialogue act feature requires costly annotation work, the contribution of this feature to the overall performance is limited. It is the second least contributing one among the seven feature sets investigated in the ablation study in [21].
 
3
The corpus is now officially named as Hazumi corpus.
 
4
[8] uses an extended version of the features used in this work. The description of the feature extraction process in [8] is mostly applicable to the features used in this work.
 
6
A model is trained with the data of one subject and is tested with all of the subjects including itself individually.
 
7
One may deploy automatic recognition modules for natural reactions from users as discussed in [6] or may adopt any specially designed interaction devices so that the users can provide feedback precisely but easily.
 
Literature
1.
Araki M, Tomimasu S, Nakano M, Komatani K, Okada S, Fujie S, Sugiyama H (2018) Collection of multimodal dialog data and analysis of the result of annotation of users’ interests. In: Proceedings of language resources and evaluation conference (LREC), pp 1584–1588
2.
Bousquet O, Warmuth MK (2002) Tracking a small set of experts by mixing past posteriors. J Mach Learn Res 3(Nov):363–396MathSciNetMATH
3.
Criminisi A, Shotton J, Konukoglu E (2011) Decision forests for classification, regression, density estimation, manifold learning and semi-supervised learning. Tech. rep, Microsoft Research
4.
Daumé III H (2007) Frustratingly easy domain adaptation. In: Proceedings of the 45th annual meeting of the association of computational linguistics, pp 256–263
5.
Fiscus JG (1999) A post-processing system to yield reduced word error rates: recognizer output voting error reduction (rover). In: Proceedings of IEEE workshop on automatic speech recognition and understanding
6.
Funakoshi K (2018) A multimodal multiparty human-robot dialogue corpus for real world interaction. In: LREC 2018 special speech sessions, pp 35–39
7.
Henderson M, Thomson B, Williams JD (2014) The second dialog state tracking challenge. In: Proceedings of the 15th annual meeting of the special interest group on discourse and dialogue (SIGDIAL), pp 263–272
8.
Hirano Y, Okada S, Nishimoto H, Komatani K (2019) Multitask prediction of exchange-level annotations for multimodal dialogue systems. In: Proceedings of 2019 international conference on multimodal interaction (ICMI), pp 85–94
9.
Japkowicz N, Stephen S (2002) The class imbalance problem: a systematic study. Intell Data Anal 6(5):429–449CrossRef
10.
Jun KS, Orabona F, Wright S, Willett R (2017) Improved strongly adaptive online learning using coin betting. In: Proceedings of the 20th international conference on artificial intelligence and statistics, pp 943–951
11.
Komatani K, Okada S, Nishimoto H, Araki M, Nakano M (2019) Multimodal dialogue data collection and analysis of annotation disagreement. In: Proceedings of international workshop on spoken dialogue systems (IWSDS)
12.
Lee A, Nakamura K, Nisimura R, Saruwatari H, Shikano K (2004) Noise robust real world spoken dialogue system using GMM based rejection of unintended inputs. In: Proceedings of Interspeech, pp 173–176
13.
14.
Malmasi S, Dras M (2018) Native language identification with classifier stacking and ensembles. Comput Linguist 44:403–446CrossRef
15.
Nakano Y, Baba N, Huang HH, Hayashi Y (2013) Implementation and evaluation of a multimodal addressee identification mechanism for multiparty conversation systems. In: Proceedings of ACM international conference on multimodal interaction (ICMI), pp 35–42
16.
Nishimoto H, Takeda R, Komatani K (2018) Predicting user’s interest level in dialogues with multimodal features. In: Proceedings of the 32nd annual conference of the japanese society for artificial intelligence, vol 3C2-OS-14b-04. (in Japanese)
17.
Pentland A (2007) Social signal processing. IEEE Signal Process Mag 24(4):108–111CrossRef
18.
Saffari A, Leistner C, Santner J, Godec M, Bischof H (2009) On-line random forests. In: Proceedings of 3rd IEEE ICCV workshop on on-line computer vision (2009)
19.
Schuller B, Steid S, Batline A (2009) The interspeech 2009 emotion challenge. In: Proceedings of 10th annual conference of the international speech communication association (INTERSPEECH), pp 312–315
20.
Segev N, Harel M, Mannor S, Crammer K, El-Yaniv R (2017) Learn on source, refine on target: a model transfer learning framework with random forests. IEEE Trans Pattern Anal Mach Intell 39(9):1811–1823CrossRef
21.
Sugiyama T, Funakoshi K, Nakano M, Komatani K (2015) Estimating response obligation in multi-party human-robot dialogues. In: Proceedings of 2015 IEEE-RAS 15th international conference on humanoid robots (Humanoids), pp 166–172
22.
Sztyler T, Stuckenschmidt H (2017) Online personalization of cross-subjects based activity recognition models on wearable devices. In: Proceedings of IEEE international conference on pervasive computing and communications (PerCom), pp 180–189
23.
Vovk V (1990) Aggregating strategies. In: Proceedings of the 3rd annual workshop on computational learning theory, pp 371—383
24.
Wang SQ, Yang J, Chou KC (2006) Using stacked generalization to predict membrane protein types based on pseudo-amino acid composition. J Theor Biol 242(4):941–946MathSciNetCrossRef
25.
26.
Metadata
Title
Packing, Stacking, and Tracking: An Empirical Study of Online User Adaptation
Authors
Jean-Sébastien Laperrière
Darryl Lam
Kotaro Funakoshi
Copyright Year
2021
Publisher
Springer Singapore
Book
Conversational Dialogue Systems for the Next Decade

Print ISBN: 978-981-15-8394-0

Electronic ISBN: 978-981-15-8395-7

Copyright Year: 2021

DOI
https://doi.org/10.1007/978-981-15-8395-7_24