Critiquing-based recommenders: survey and emerging trends

By Li Chen, Pearl Huan Z. Pu. User modeling and user-adapted interaction, 2012

313 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Information Systems, Sociology and Political Science

View Paper: https://www.semanticscholar.org/paper/5b0295e92ac6f493f23f46efcd6fdfbbca74ac48

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Information Systems

In Sociology and Political Science

Citations in the corpus

A Survey on Conversational Recommender Systems (2020, background): “s or determine the next question to ask, and (ii) additional, supporting tasks. 5.1 Main Tasks Broadly speaking, CRS carry out four general types of tasks (or: system actions) during the conversation [26, 96]: Request, Recommend, Explain, and Respond, see Figure4. However, not every CRS necessarily ”
A multi-disciplinary review of knowledge acquisition methods: From human to autonomous eliciting agents (2016, methodology): “A recently emerging approach, thoroughly reviewed in [41], applicable in conjunction with any of the methods and family of methods described above, is the Elicitation by Critiquing, in which third party participants in the elicitation process are asked to critique SMEs actions or the resultant produ”
Advances and Challenges in Conversational Recommender Systems: A Survey (2021, methodology): “Generally, interactive recommendations [231, 175, 204, 40], critiquing methods [23, 187, 107, 106], and CRSs focusing on the multi-turn conversation strategy [32, 31, 88, 89, 95] are prone to use the pre-annotated input and rule-based or templatebased output; dialogue systems [200, 220, 51] and CRSs”
Explaining the user experience of recommender systems (2012, background): “…recommenders let users rate items, (e.g. youtube.com, see also McNee et al. 2002; Cena et al. 2010; Gena et al. 2011), critique items (see also Chen and Pu 2012; Viappiani et al. 2006), assign weights to item attributes (see also Häubl et al. 2004), or indicate their specific needs (e.g.…”
Recommender Systems: Introduction and Challenges (2015, unknown): “recommender systems are case-based [15, 19, 45].”
Collaborative Filtering beyond the User-Item Matrix (2014, methodology): “Collaborative Filtering beyond the User-Item Matrix 3:31 by allowing them to adjust the recommendations for this user [Chen and Pu 2012].”
"I LOVE THIS SITE!" vs. "It's a little girly": Perceptions of and Initial User Experience with Pinterest (2015, background): “For example, a “critiquing” approach [3] would let users select contacts to import and follow, then update their content feed in real time.”
Towards Conversational Recommender Systems (2016, background): “There have been many works (critique-based [7], constraint-based [10], dialog, utility-based recommenders [19]) emphasizing the importance of interactivity in recommenders so that the user has a more active role over the recommendations.”
Towards Knowledge-Based Recommender Dialog System (2019, unknown): “Related literature can also be found in Thompson et al. (2004), Mahmood and Ricci (2009), Chen and Pu (2012), Widyantoro and Baizal (2014) and Liao et al. (2019).”
Evaluating Recommender Systems with User Experiments (2015, background): “Substantial user-centric work in this area (as summarized in [19]) shows that example critiquing systems save cognitive effort and increase decision accuracy.”
Recommender systems: from algorithms to user experience (2012, background): “The FindMe systems spawned a wide range of more interactive, exploratory recommender systems, including many of the systems now known as critiquing-based systems (Chen and Pu 2012).”
Inspectability and control in social recommenders (2012, background): “Researchers have implemented various aspects of control in recommender systems, ranging from simple preference elicitation at recommendation time [6] to more complex iterative processes such as dynamic critiquing which allows users to tweak ordered numerical attributes during the recommendation phas”
The Personalization of Conversational Agents in Health Care: Systematic Review (2019, background): “Personalization in CAs can be achieved implicitly by processing past interactions with users [11,13] or explicitly by user-entered information at the set-up time [8] or using ongoing confirmation style input [14].”
Recommender Systems (2012, background): “…users tend to visit very popular items and select similar items, while old users usuallyhavemorespecificinterests[53,54],andusersbehavemuchdifferentlybetweenlow-risk(e.g.,collectingbookmarks, downloading music, etc.) and high-risk (e.g., buying a computer, renting a house, etc.) activities [55,56].”
Recommender Systems (2016, unknown): “Surveys of preference elicitation methods and critiquing may be found in [148, 149].”
RecSim: A Configurable Simulation Platform for Recommender Systems (2019, methodology): “CIRs often use conversational methods [Vinyals and Le, 2015, Ghazvininejad et al., 2018], example critiquing or preference elicitation [Chen and Pu, 2012, Christakopoulou et al., 2016], bandit-based exploration [Li et al., 2010, 2016, Christakopoulou and Banerjee, 2018], or reinforcement learning…”