On the Importance of Subtext in Recommender Systems

Eliciting Nuanced Preferences Using a Speech-based Conversational Interface

Peter Grasch 1  and Alexander Felfernig 1
  • 1 Graz University of Technology, Institute for Software Technology, Graz, Austria
Peter Grasch
  • Corresponding author
  • Email
  • Further information
  • Peter Grasch is a master’s student at Graz University of Technology, majoring in Computer Science with a special focus on Algorithms and Software Technology. He received his Bachelor’s degree in the fall of 2012 from the same university. Next to his studies, Peter Grasch has been involved in numerous research projects revolving around speech recognition, including the EU project Astromobile. His research interests include human-computer interaction, recommender systems, and speech recognition.
  • Search for other articles:
  • degruyter.comGoogle Scholar
and Prof. Alexander Felfernig
  • Email
  • Further information
  • Alexander Felfernig is a full professor at the Graz University of Technology (Austria) since March 2009 where he directs the Applied Software Engineering (ASE) research group at the Institute for Software Technology (IST). He received his PhD in Computer Science from the University of Klagenfurt. Alexander Felfernig has published numerous papers in renowned international conferences and journals, co-authored the book on “Recommender Systems” and acted as an organizer of international conferences and workshops. He is also a member of the Editorial Board of Applied Intelligence and the Journal of Intelligent Information Systems. His research interests include configuration systems, recommender systems, model-based diagnosis, software requirements engineering, different aspects of human decision making, and knowledge acquisition methods.
  • Search for other articles:
  • degruyter.comGoogle Scholar

Abstract

Conversational recommender systems have been shown capable of allowing users to navigate even complex and unknown application domains effectively. However, optimizing preference elicitation remains a largely unsolved problem. In this paper we introduce SPEECHREC, a speech-enabled, knowledge-based recommender system, that engages the user in a natural-language dialog, identifying not only purely factual constraints from the users’ input, but also integrating nuanced lexical qualifiers and paralinguistic information into the recommendation strategy. In order to assess the viability of this concept, we present the results of an empirical study where we compare SPEECHREC to a traditional knowledge-based recommender system and show how incorporating more granular user preferences in the recommendation strategy can increase recommendation quality, while reducing median session length by 46 %.

  • [1]

    Alina Andreevskaia, Sabine Bergler, and Monica Urseanu. All blogs are not made equal: Exploring genre differences in sentiment tagging of blogs. In Intl. Conf. on Weblogs and Social Media, Boulder, CO, USA, 2007.

  • [2]

    John Brooke. SUS - a quick and dirty usability scale. Usability evaluation in industry, 189:194, 1996.

  • [3]

    Robin Burke. Integrating knowledge-based and collaborative-filtering recommender systems. In Proc. of the Worksh. on AI and Electr. Commerce, pages 69–72, 1999.

  • [4]

    Robin Burke. Knowledge-based recommender systems. Encyclopedia of library and information systems, 69(Supplement 32):175–186, 2000.

  • [5]

    Robin D. Burke, Kristian J. Hammond, and Benjamin C. Young. Knowledge-based navigation of complex information spaces. In Proc. of the 13th Natl. Conf. on AI - Volume 1, AAAI’96, pages 462–468. AAAI Press, 1996.

  • [6]

    Li Chen and Pearl Pu. Survey of preference elicitation methods. Technical report, Swiss Federal Inst. of Techn. in Lausanne (EPFL, 2004).

  • [7]

    Li Chen and Pearl Pu. Evaluating critiquing-based recommender agents. In Proc. of the 21st Natl. Conf. on AI - Volume 1, AAAI’06, pages 157–162. AAAI Press, 2006.

  • [8]

    Alexander Clark. Combining distributional and morphological information for part of speech induction. In Proc. of the Tenth Conf. on European Chapter of the Association for Computational Linguistics - Volume 1, EACL ’03, pages 59–66, Stroudsburg, PA, USA, 2003. Association for Computational Linguistics.

  • [9]

    Ruihai Dong, Markus Schaal, MichaelP. O ahony, Kevin McCarthy, and Barry Smyth. Opinionated product recommendation. In SarahJane Delany and Santiago Ontañón, editors, Case-Based Reasoning Research and Development, volume 7969 of Lecture Notes in Computer Science, pages 44–58. Springer Berlin Heidelberg, 2013.

  • [10]

    F. Eyben, M. Wollmer, and B. Schuller. OpenEAR - introducing the munich open-source emotion and affect recognition toolkit. In Affective Computing and Intelligent Interaction and Workshops, 2009. ACII 2009. 3rd Intl. Conf. on, pages 1–6, Sept 2009.

  • [11]

    Florian Eyben, Felix Weninger, Florian Gross, and Björn Schuller. Recent developments in openSMILE, the munich open-source multimedia feature extractor. In Proc. of the 21st ACM Intl. Conf. on Multimedia, MM ’13, pages 835–838, New York, NY, USA, 2013. ACM.

  • [12]

    A. Felfernig and R. Burke. Constraint-based recommender systems: Technologies and research issues. In Proc. of the 10th Intl. Conf. on Electronic Commerce, ICEC ’08, pages 3:1–3:10, New York, NY, USA, 2008. ACM.

  • [13]

    A. Felfernig, K. Isak, K. Szabo, and P. Zachar. The VITA financial services sales support environment. In Proc. of the 19th Natl. Conf. on Innovative Applications of AI - Volume 2, IAAI’07, pages 1692–1699. AAAI Press, 2007.

  • [14]

    Alexander Felfernig, Michael Jeran, Gerald Ninaus, Florian Reinfrank, Stefan Reiterer, and Martin Stettinger. Basic approaches in recommendation systems. In Recommendation Systems in Software Engineering, pages 15–37. Springer, 2014.

  • [15]

    Alexander Felfernig, Stefan Reiterer, Martin Stettinger, and Michael Jeran. An overview of direct diagnosis and repair techniques in the WeeVis recommendation environment. In Knowledge-based Configuration: From Research to Business Cases, pages 297–307. Elsevier, 2014.

  • [16]

    Peter Grasch, Alexander Felfernig, and Florian Reinfrank. ReComment: Towards critiquing-based recommendation with speech interaction. In Proc. of the 7th ACM conf. on Recommender systems, pages 157–164. ACM, 2013.

  • [17]

    Lorraine McGinty and James Reilly. On the evolution of critiquing recommenders. In Recommender Systems Handbook, pages 419–453. Springer, 2011.

  • [18]

    Lorraine McGinty and Barry Smyth. On the role of diversity in conversational recommender systems. In Proc. of the 5th Intl. Conf. on Case-based Reasoning: Research and Development, ICCBR’03, pages 276–290. Springer, 2003.

  • [19]

    Gary McKeown, Michel François Valstar, Roderick Cowie, and Maja Pantic. The SEMAINE corpus of emotionally coloured character interactions. In Multimedia and Expo (ICME), 2010 IEEE Intl. Conf. on, pages 1079–1084. IEEE, 2010.

  • [20]

    George A. Miller. WordNet: A lexical database for english. Commun. ACM, 38(11):39–41, November 1995.

  • [21]

    Samaneh Moghaddam and Martin Ester. Opinion Digger: An unsupervised opinion miner from unstructured product reviews. In Proc. of the 19th ACM Intl. Conf. on Information and Knowledge Management, CIKM ’10, pages 1825–1828, New York, NY, USA, 2010. ACM.

  • [22]

    Daniel Naber. OpenThesaurus: Building a thesaurus with a web community. Retrieved January, 3:2005, 2004.

  • [23]

    Michael J. Pazzani and Daniel Billsus. Content-based recommendation systems. In Peter Brusilovsky, Alfred Kobsa, and Wolfgang Nejdl, editors, The Adaptive Web, pages 325–341. Springer, 2007.

  • [24]

    Pearl Huan Z. Pu and Pratyush Kumar. Evaluating example-based search tools. In Proc. of the 5th ACM Conf. on Electronic Commerce, EC ’04, pages 208–217, New York, NY, USA, 2004. ACM.

  • [25]

    Lingyun Qiu and Izak Benbasat. An investigation into the effects of text-to-speech voice and 3d avatars on the perception of presence and flow of live help in electronic commerce. ACM Trans. Comput.-Hum. Interact., 12(4):329–355, 2005.

  • [26]

    James A Russell. A circumplex model of affect. Journal of personality and social psychology, 39(6):1161, 1980.

  • [27]

    Barbara Schuppler, Martine Adda-Decker, and Juan A. Morales-Cordovilla. Pronunciation variation in read and conversational Austrian German. In Interspeech 2014, pages 1453–1457, Singapore, 2014.

  • [28]

    Rico Sennrich, Gerold Schneider, Martin Volk, and Martin Warin. A new hybrid dependency parser for german. Proc. of the German Society for Comp. Linguistics and Language Techn., pages 115–124, 2009.

  • [29]

    Rico Sennrich, Martin Volk, and Gerold Schneider. Exploiting synergies between open resources for german dependency parsing, POS-tagging, and morphological analysis. In RANLP, pages 601–609, 2013.

  • [30]

    Mostafa Al Shaikh, Helmut Prendinger, and Ishizuka Mitsuru. Assessing sentiment of text by semantic dependency and contextual valence analysis. In Proc. of the 2Nd Intl. Conf. on Affective Computing and Intelligent Interaction, ACII ’07, pages 191–202. Springer, 2007.

  • [31]

    Hideo Shimazu. ExpertClerk: Navigating shoppers’ buying process with the combination of asking and proposing. In Proc. of the 17th Intl. Joint Conf. on AI - Volume 2, IJCAI’01, pages 1443–1448, San Francisco, CA, USA, 2001. Morgan Kaufmann Publishers Inc.

  • [32]

    Elizabeth Shriberg. Toerrrr’is human: ecology and acoustics of speech disfluencies. Journal of the Intl. Phonetic Association, 31(1):153–169, 2001.

  • [33]

    Gabriel Skantze. Exploring human error handling strategies: Implications for spoken dialogue systems. In ISCA Tutorial and Research Workshop on Error Handling in Spoken Dialogue Systems, 2003.

  • [34]

    Cynthia A. Thompson, Mehmet H. Göker, and Pat Langley. A personalized system for conversational recommendations. J. Artif. Int. Res., 21(1):393–428, March 2004.

  • [35]

    Pontus Wärnestål. User evaluation of a conversational recommender system. In Proc. of the 4th Workshop on Knowledge and Reasoning in Practical Dialogue Systems, 2005.

Purchase article
Get instant unlimited access to the article.
$42.00
Log in
Already have access? Please log in.


or
Log in with your institution

Journal + Issues

i-com - Journal of Interactive Media is devoted to human-computer interaction, media design, usability, engineering and systems evaluation, software ergonomics, cooperative systems, e-learning, mobile and ubiquitous systems, user-adaptive systems, agent development tools and methods for media in different application fields, barrier-free systems design, and the social aspects of information and communication technologies.

Search