Integrating Active Learning Strategies in Model Based Recommender Systems

In this era of information overload and abundant choices, recommender systems play an important role in assisting users by providing personalized suggestions and enhancing user experiences across various domains. Traditional recommender systems encounter several challenges, specifically caused by the cold start problem and data sparsity. This article presents an innovative approach to dealing with these limitations by utilizing active learning strategies based recommender system. Emphasizing the effectiveness of active learning, the method strategically applies various query strategies within the movie recommendation domain. The query strategies aim to balance query quantity and feedback quality. Strategies such as random, popularity, similarity to profile, highest predicted, binary predicted, gini, popgini, error, and poperror enhance collaborative filtering-based recommendations, catering to varied user preferences. The study applies these strategies and matrix factorization to the MovieLens100K and MovieLens1M datasets, offering practical insights for diverse movie recommendation scenarios. The active learning system demonstrates its potential for improving recommender systems in real-world environments.

• Publication year

  2024

• Venue

  Proceedings of the 7th International Conference on Networking, Intelligent Systems and Security

• Publication date

  2024-04-18

• Fields of study

  Not labeled

• Identifiers
  DOI 10.1145/3659677.3659838
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Simulating Users’ Interactions with Recommender Systems

  2022cited by this paper
• FeatureMF: An Item Feature Enriched Matrix Factorization Model for Item Recommendation

  2021cited by this paper
• A systematic literature review of sparsity issues in recommender systems

  2020cited by this paper
• Product Recommendation for e-Commerce System based on Ontology

  2019cited by this paper
• Collaborative filtering embeddings for memory-based recommender systems

  2019cited by this paper
• Improved SVD-based Initialization for Nonnegative Matrix Factorization using Low-Rank Correction

  2018cited by this paper
• The MovieLens Datasets: History and Context

  2016cited by this paper
• A survey of active learning in collaborative filtering recommender systems

  2016cited by this paper
• Active Learning in Collaborative Filtering Recommender Systems

  2014cited by this paper
• Active learning

  2013cited by this paper
• Active learning strategies for rating elicitation in collaborative filtering

  2013cited by this paper
• Recommendation System based on user interaction data applied to intelligent electronic books

  2011cited by this paper
• Collaborative Filtering Recommender Systems

  2011cited by this paper
• Addressing cold-start problem in recommendation systems

  2008cited by this paper
• Interview with Simon Funk

  2007cited by this paper
• Item-based top-N recommendation algorithms

  2004cited by this paper
• Getting to know you: learning new user preferences in recommender systems

  2002cited by this paper
• Using collaborative filtering to weave an information tapestry

  1992cited by this paper

• No citing papers are available for this paper.