Algorithms for Personalization Techniques and Recommender Systems

A special issue of Algorithms (ISSN 1999-4893).

Deadline for manuscript submissions: closed (30 April 2020) | Viewed by 22360

Special Issue Editors


E-Mail Website
Guest Editor
Faculty of Computer Science, Free University of Bozen-Bolzano, Bolzano, Italy
Interests: artificial intelligence; recommender systems; multimedia retrieval; active learning; human computer interaction

E-Mail Website
Guest Editor
Faculty of Computer Science, Free University of Bozen-Bolzano, Bolzano, Italy
Interests: recommender systems; affective computing; affective user modeling; personality computing; social signal processing

Special Issue Information

Dear Colleagues,

Recommender Systems are software tools that aim at alleviating the information overload problem by generating personalized recommendations for users. They can support users by providing a shorter and more “personalized” item catalog in order to find items that are more suitable for users’ specific tastes and preferences. Today, most e-commerce and information search and retrieval systems incorporate a recommender system as an essential tool for supporting the user of the system.

Many recommendation approaches have been developed and deployed in a vast range of application domains in the last two decades. The families of algorithms are content-based recommender systems, which suggest items of interest based on the associated content features, collaborative filtering systems, which predict the user ratings based on the previous ratings, demographic recommender systems, which build recommendations by identifying similar users based on demographic information, utility-based recommender systems, which match a user’s need and a set of options available, knowledge-based recommender systems, which attempt to suggest items that are inferred by a reasoning process and form a relationship between a user’s need and possible recommendations, and hybrid recommender systems, which aggregate a number of approaches in order to tackle the specific limitations of an individual approach.

In recent years, however, algorithms that go beyond the matrix completion paradigm have arisen. Examples are (i) algorithms that address specific scenarios, such as sequential recommendations, conversational recommendations, and recommendations in social networks, (ii) user-centric algorithms that take into account explanations, cognitive models, biases, trust, or transparency, (iii) algorithms that take into consideration multiple perspectives, such as multi-stakeholder approaches, cross-domain recommendations, and multisignal recommendations, (iv) and totally novel algorithmic approaches, such as the usage of deep learning.

This Special Issue on “Algorithms for Personalization Techniques and Recommender Systems” aims to form a reference point in this research area, i.e., the models and algorithms for the (more generic) goal of “personalization” and the (more specific) goal of “recommendations”. We invite works that present their latest findings in the state-of-the-art of the related theory and practice, and to contribute to the advancement of this research area. In particular, we would welcome original and unpublished submissions related to the following topics:

  • Content-based filtering;
  • Trust and transparency;
  • Large scale recommendation;
  • Multi-stakeholder approaches;
  • Multisignal recommendations;
  • Cross-domain recommendations;
  • Sequential recommender systems;
  • Collaborative filtering techniques;
  • Side-information for recommendation;
  • Adaptive recommendation algorithms;
  • Visually-aware recommender systems;
  • Conversational recommender systems;
  • Algorithms for group recommendation;
  • Semantic-aware recommender systems;
  • Novel similarity-based recommendation;
  • Recommendation based on deep learning;
  • Recommender systems in social networks;
  • Active learning for recommender systems;
  • Transfer learning in recommender systems;
  • Cross-domain algorithms for recommendation;
  • Exploiting user cognition for recommendation;
  • Recommendation based on matrix factorization;
  • Recommendation based on audiovisual signals;
  • Explanations methods for recommender systems;
  • Generic recommendation models and algorithms;
  • Algorithms for using implicit and explicit preferences;
  • Ranking algorithms for personalization and adaptation;
  • Novelty, diversity, or serendipity in recommender systems;
  • Algorithms for context-aware recommender systems (CARS);
  • Scalability and performance measures in recommender systems;
  • Novel applications areas for recommender systems (e.g., health, education, and fashion);
  • Further relevant topics.

Dr. Mehdi Elahi
Dr. Marko Tkalcic
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Algorithms is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 1600 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

  • recommender systems
  • collaborative filtering
  • content-based filtering
  • rating prediction
  • ranking prediction
  • active learning
  • cold start

Benefits of Publishing in a Special Issue

  • Ease of navigation: Grouping papers by topic helps scholars navigate broad scope journals more efficiently.
  • Greater discoverability: Special Issues support the reach and impact of scientific research. Articles in Special Issues are more discoverable and cited more frequently.
  • Expansion of research network: Special Issues facilitate connections among authors, fostering scientific collaborations.
  • External promotion: Articles in Special Issues are often promoted through the journal's social media, increasing their visibility.
  • e-Book format: Special Issues with more than 10 articles can be published as dedicated e-books, ensuring wide and rapid dissemination.

Further information on MDPI's Special Issue polices can be found here.

Published Papers (3 papers)

Order results
Result details
Select all
Export citation of selected articles as:

Research

27 pages, 2513 KiB  
Article
Towards Cognitive Recommender Systems
by Amin Beheshti, Shahpar Yakhchi, Salman Mousaeirad, Seyed Mohssen Ghafari, Srinivasa Reddy Goluguri and Mohammad Amin Edrisi
Algorithms 2020, 13(8), 176; https://doi.org/10.3390/a13080176 - 22 Jul 2020
Cited by 63 | Viewed by 10491
Abstract
Intelligence is the ability to learn from experience and use domain experts’ knowledge to adapt to new situations. In this context, an intelligent Recommender System should be able to learn from domain experts’ knowledge and experience, as it is vital to know the [...] Read more.
Intelligence is the ability to learn from experience and use domain experts’ knowledge to adapt to new situations. In this context, an intelligent Recommender System should be able to learn from domain experts’ knowledge and experience, as it is vital to know the domain that the items will be recommended. Traditionally, Recommender Systems have been recognized as playlist generators for video/music services (e.g., Netflix and Spotify), e-commerce product recommenders (e.g., Amazon and eBay), or social content recommenders (e.g., Facebook and Twitter). However, Recommender Systems in modern enterprises are highly data-/knowledge-driven and may rely on users’ cognitive aspects such as personality, behavior, and attitude. In this paper, we survey and summarize previously published studies on Recommender Systems to help readers understand our method’s contributions to the field in this context. We discuss the current limitations of the state of the art approaches in Recommender Systems and the need for our new approach: A vision and a general framework for a new type of data-driven, knowledge-driven, and cognition-driven Recommender Systems, namely, Cognitive Recommender Systems. Cognitive Recommender Systems will be the new type of intelligent Recommender Systems that understand the user’s preferences, detect changes in user preferences over time, predict user’s unknown favorites, and explore adaptive mechanisms to enable intelligent actions within the compound and changing environments. We present a motivating scenario in banking and argue that existing Recommender Systems: (i) do not use domain experts’ knowledge to adapt to new situations; (ii) may not be able to predict the ratings or preferences a customer would give to a product (e.g., loan, deposit, or trust service); and (iii) do not support data capture and analytics around customers’ cognitive activities and use it to provide intelligent and time-aware recommendations. Full article
(This article belongs to the Special Issue Algorithms for Personalization Techniques and Recommender Systems)
Show Figures

Figure 1

25 pages, 6642 KiB  
Article
An Algorithm for Density Enrichment of Sparse Collaborative Filtering Datasets Using Robust Predictions as Derived Ratings
by Dionisis Margaris, Dimitris Spiliotopoulos, Gregory Karagiorgos and Costas Vassilakis
Algorithms 2020, 13(7), 174; https://doi.org/10.3390/a13070174 - 17 Jul 2020
Cited by 11 | Viewed by 3941
Abstract
Collaborative filtering algorithms formulate personalized recommendations for a user, first by analysing already entered ratings to identify other users with similar tastes to the user (termed as near neighbours), and then using the opinions of the near neighbours to predict which items the [...] Read more.
Collaborative filtering algorithms formulate personalized recommendations for a user, first by analysing already entered ratings to identify other users with similar tastes to the user (termed as near neighbours), and then using the opinions of the near neighbours to predict which items the target user would like. However, in sparse datasets, too few near neighbours can be identified, resulting in low accuracy predictions and even a total inability to formulate personalized predictions. This paper addresses the sparsity problem by presenting an algorithm that uses robust predictions, that is predictions deemed as highly probable to be accurate, as derived ratings. Thus, the density of sparse datasets increases, and improved rating prediction coverage and accuracy are achieved. The proposed algorithm, termed as CFDR, is extensively evaluated using (1) seven widely-used collaborative filtering datasets, (2) the two most widely-used correlation metrics in collaborative filtering research, namely the Pearson correlation coefficient and the cosine similarity, and (3) the two most widely-used error metrics in collaborative filtering, namely the mean absolute error and the root mean square error. The evaluation results show that, by successfully increasing the density of the datasets, the capacity of collaborative filtering systems to formulate personalized and accurate recommendations is considerably improved. Full article
(This article belongs to the Special Issue Algorithms for Personalization Techniques and Recommender Systems)
Show Figures

Graphical abstract

19 pages, 2718 KiB  
Article
A Hybrid Ontology-Based Recommendation System in e-Commerce
by Márcio Guia, Rodrigo Rocha Silva and Jorge Bernardino
Algorithms 2019, 12(11), 239; https://doi.org/10.3390/a12110239 - 8 Nov 2019
Cited by 23 | Viewed by 6503
Abstract
The growth of the Internet has increased the amount of data and information available to any person at any time. Recommendation Systems help users find the items that meet their preferences, among the large number of items available. Techniques such as collaborative filtering [...] Read more.
The growth of the Internet has increased the amount of data and information available to any person at any time. Recommendation Systems help users find the items that meet their preferences, among the large number of items available. Techniques such as collaborative filtering and content-based recommenders have played an important role in the implementation of recommendation systems. In the last few years, other techniques, such as, ontology-based recommenders, have gained significance when reffering better active user recommendations; however, building an ontology-based recommender is an expensive process, which requires considerable skills in Knowledge Engineering. This paper presents a new hybrid approach that combines the simplicity of collaborative filtering with the efficiency of the ontology-based recommenders. The experimental evaluation demonstrates that the proposed approach presents higher quality recommendations when compared to collaborative filtering. The main improvement is verified on the results regarding the products, which, in spite of belonging to unknown categories to the users, still match their preferences and become recommended. Full article
(This article belongs to the Special Issue Algorithms for Personalization Techniques and Recommender Systems)
Show Figures

Figure 1

Back to TopTop