Burst-induced Multi-Armed Bandit for Learning Recommendation

Alves, Rodrigo; Ledent, Antoine; Kloft, Marius

doi:10.1145/3460231.3474250

Cited by 2 publications

(5 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This is due to the fact that there is a large number of attributes in the datasets, causing these methods to construct some irrelevant questions regarding the user's current interests. (2) In order to construct questions for the user, most CRS methods employ a policy agent to guide the construction of questions [5,50,67]. However, the training of these policy agents requires a huge and comprehensive corpus so that the learned agents can provide an accurate policy under all circumstances.…”

Section: Zootopia 2016mentioning

confidence: 99%

“…• Direction (1) focuses on applying bandit-based models to CRS in order to balance EE trade-o s for cold-start users [2,36]. Christakopoulou et al [9] rst propose to employ bandit algorithms in CRS to help users better nd their desired restaurants.…”

Section: Conversational Recommendation Systemmentioning

confidence: 99%

“…Li et al [36] propose to seamlessly unify attributes and items in the same arm space and achieve the EE trade-o s automatically using the framework of Thompson Sampling. • In direction (2), the key research issue is to estimate the users' preferences towards attributes.…”

Section: Conversational Recommendation Systemmentioning

confidence: 99%

“…(1) Existing question construction methods propose to construct questions from either the exploitative strategy or the exploitative strategy, which causes that a CRS cannot discover the user's past interests as well as explore the user's potential new interests. (2) In addition, it is di cult to learn an e ective policy agent to guide the construction of questions for most CRS models, as the training of these policy agents requires a huge and comprehensive corpus. Our RFCR model constructs questions for users from the exploitative strategy and the exploratory strategy and proposes a fusion method to take advantage of both strategies.…”

Section: Conversational Recommendation Systemmentioning

confidence: 99%

“…Considering the limited number of users' interactions in CRS, we set 𝑄 to {5, 10, 20}, respectively. There are several schemes to split 𝑄 into 𝑄 1 and 𝑄 2 , we choose the greedy search method (e.g., when 𝑄 = 5, (𝑄 1 , 𝑄 2 ) could be (1, 4), (2, 3), (3,2) or (4, 1)) to nd the optimal splitting scheme. Note that we set neither 𝑄 1 nor 𝑄 2 to 0 in this section, where 0 denotes that either the LQS subsession or the GQS subsession is removed, which we will discuss later in § 6.4.…”

Section: Impact Of DI Erent Spli Ing Schemes Of 𝑄 (Rq 1)mentioning

confidence: 99%

See 4 more Smart Citations

Enhancing Conversational Recommendation Systems with Representation Fusion

et al. 2023

View full text Add to dashboard Cite

Conversational Recommendation Systems (CRSs) aim to improve recommendation performance by utilizing information from a conversation session. A CRS first constructs questions and then asks users for their feedback in each conversation session in order to refine better recommendation lists to users. The key design of CRS is to construct proper questions and obtain users’ feedback in response to these questions so as to effectively capture user preferences. Many CRS works have been proposed; however, they suffer from defects when constructing questions for users to answer: (1) employing a dialogue policy agent for constructing questions is one of the most common choices in CRS, but it needs to be trained with huge corpus; (2) it is not appropriate that constructing questions from a single policy (e.g., A CRS only selects attributes that the user has interacted with) for all users with different preferences. To address these defects, we propose a novel CRS model, namely, a R epresentation F usion-based C onversational R ecommendation model (RFCR), where the whole conversation session is divided into two subsessions (i.e., Local Question Search subsession and Global Question Search subsession) and two different question search methods are proposed to construct questions in the corresponding subsessions without employing policy agents. In particular, in the Local Question Search subsession we adopt a novel graph mining method to find questions, where the paths in the graph between users and attributes can eliminate irrelevant attributes; in the Global Question Search subsession we propose to initialize user preference on items with the user and all item historical rating records and construct questions based on user’s preference. Then, we update the embeddings independently over the two subsessions according to user’s feedback and fuse the final embeddings from the two subsessions for the recommendation. Experiments on three real-world recommendation datasets demonstrate that our proposed method outperforms five state-of-the-art baselines.

show abstract

Section: Zootopia 2016mentioning

confidence: 99%

Section: Conversational Recommendation Systemmentioning

confidence: 99%

Section: Conversational Recommendation Systemmentioning

confidence: 99%

Section: Conversational Recommendation Systemmentioning

confidence: 99%

Section: Impact Of DI Erent Spli Ing Schemes Of 𝑄 (Rq 1)mentioning

confidence: 99%

See 3 more Smart Citations

Enhancing Conversational Recommendation Systems with Representation Fusion

et al. 2023

View full text Add to dashboard Cite

show abstract

Regionalization-Based Collaborative Filtering: Harnessing Geographical Information in Recommenders

Alves

2024

ACM Trans. Spatial Algorithms Syst.

View full text Add to dashboard Cite

Regionalization, also known as spatially constrained clustering, is an unsupervised machine learning technique used to identify and define spatially contiguous regions. In this work, we introduce a methodology to regionalize recommendation systems (RSs) based on a collaborative filtering approach. Two main challenges arise when performing regionalization based on users’ preferences in RSs: (1) unstructured data, as interactions are often scarce and observed on a smaller scale; and (2) the difficulty of evaluation of the quality of the clustering results. To address these challenges, our methodology relies on inductive matrix completion (IMC), a fundamental approach to recover unknown entries of a rating matrix while utilizing region information to extract a region-based feature matrix. With this feature matrix, our method becomes adaptive and seamlessly integrates with various regionalization algorithms to create regionalization candidates. This enables us to derive more accurate recommendations that consider regionalized effects and discover interesting patterns in localized user behavior. We experimentally evaluate our model on synthetic datasets to demonstrate its efficacy in settings where our underlying assumptions are correct. Furthermore, we present a real-world case study illustrating the interpretable information the model can derive in terms of regionalized recommendation relevance.

show abstract

Burst-induced Multi-Armed Bandit for Learning Recommendation

Cited by 2 publications

References 34 publications

Enhancing Conversational Recommendation Systems with Representation Fusion

Enhancing Conversational Recommendation Systems with Representation Fusion

Regionalization-Based Collaborative Filtering: Harnessing Geographical Information in Recommenders

Contact Info

Product

Resources

About