Design of knowledge for conversational recommender system based on product functional requirements

Conversational recommender system (CRS) has been developed to simulate conversation between a customer with a professional sales support. For most customers, specifying needs based on product functional requirements is more natural way rather than technical features, especially for high-tech products. Thus, the CRS requires a knowledge design that is able to generate interactions based on product functional requirements appropriately. This paper proposes a design of knowledge by using ontology to overcome this issue. The nature of ontology structure allows us to map functional requirements with technical features of the product. Based on our user study, most of expert users (familiar with product technical features) and novice users (not familiar with product technical features) perceive that the CRS based on our proposed knowledge design is more useful than that of the recommender system based on the technical features.

[1]  Sergio Escalera Guerrero,et al.  Increasing Retrieval Quality in Conversational Recommenders , 2012, IEEE Transactions on Knowledge and Data Engineering.

[2]  Barry Smyth,et al.  Evaluating Preference-Based Feedback in Recommender Systems , 2002, AICS.

[3]  Dwi H. Widyantoro,et al.  A framework of conversational recommender system based on user functional requirements , 2014, 2014 2nd International Conference on Information and Communication Technology (ICoICT).

[4]  Barry Smyth,et al.  Case-based recommender systems , 2005, The Knowledge Engineering Review.

[5]  Weiru Liu,et al.  History-guided conversational recommendation , 2014, WWW '14 Companion.

[6]  Padraig Cunningham,et al.  A Dynamic Approach to Reducing Dialog in On-Line Decision Guides , 2000, EWCBR.

[7]  Norio Shiratori,et al.  Provision of Thai herbal recommendation based on an ontology , 2010, 3rd International Conference on Human System Interaction.

[8]  Sascha Schmitt,et al.  simVar: A Similarity-Influenced Question Selection Criterion for e-Sales Dialogs , 2002, Artificial Intelligence Review.

[9]  Gerhard Friedrich,et al.  Constraint-Based Recommender Systems , 2015, Recommender Systems Handbook.

[10]  Francesco Ricci,et al.  Feature selection methods for conversational recommender systems , 2005, 2005 IEEE International Conference on e-Technology, e-Commerce and e-Service.

[11]  Hideo.  ExpertClerk : A Conversational Case-Based Reasoning Tool for Developing Salesclerk Agents in E-Commerce , .

[12]  Qun Chen,et al.  A trust-based Top-K recommender system using social tagging network , 2012, 2012 9th International Conference on Fuzzy Systems and Knowledge Discovery.

[13]  Barry Smyth,et al.  On the Dynamic Generation of Compound Critiques in Conversational Recommender Systems , 2004, AH.

[14]  Chang-Shing Lee,et al.  Ontological recommendation multi-agent for Tainan City travel , 2009, Expert Syst. Appl..

[15]  Padraig Cunningham,et al.  A Comparison of Incremental Case-Based Reasoning and Inductive Learning , 1994, EWCBR.

[16]  Barry Smyth,et al.  Compound Critiques for Conversational Recommender Systems , 2004, IEEE/WIC/ACM International Conference on Web Intelligence (WI'04).

[17]  Lorraine McGinty,et al.  On the Evolution of Critiquing Recommenders , 2011, Recommender Systems Handbook.