Proposing improved meta-heuristic algorithms for clustering and separating users in the recommender systems

Rashidi, Rahim; Khamforoosh, Keyhan; Sheikhahmadi, Amir

doi:10.1007/s10660-021-09478-9

Cited by 12 publications

(6 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although model-based methods allow more accuracy compared to memory-based ones, they still suffer from data sparsity and cold start problems (Rashidi, Khamforoosh, & Sheikhahmadi, 2021). Various methods have been proposed in the literature to solve these problems, which we categorize into several groups.…”

Section: Introductionmentioning

confidence: 99%

TrustDL: Use of trust-based dictionary learning to facilitate recommendation in social networks

Khaledian

Nazari

Khamforoosh

et al. 2023

Expert Systems with Applications

View full text Add to dashboard Cite

Section: Introductionmentioning

confidence: 99%

TrustDL: Use of trust-based dictionary learning to facilitate recommendation in social networks

Khaledian

Nazari

Khamforoosh

et al. 2023

Expert Systems with Applications

View full text Add to dashboard Cite

“…The experimental results of the proposed model OCB-DAE are computed by taking the average RMSE of all modeled clusters using (6). The proposed model uses 80% of dataset for training and 20% for testing.…”

Section: Model Evaluation Resultsmentioning

confidence: 99%

“…Clustering is one of the unsupervised learning techniques, it is widely used in recommender system (RS) to group the users or items based on their similarity. Each group is called a cluster and each cluster contains very similar members by a given data properties [6]. Despite being the most commonly used clustering method, k-means is known to have several disadvantages, such as being influenced by the initial centroids and being sensitive to the initial parameter settings [7].…”

Section: Introductionmentioning

confidence: 99%

Cluster-based denoising autoencoders for rate prediction recommender systems

Al-Asadi

Jasim²

2023

IJEECS

View full text Add to dashboard Cite

Recommender system (RS) is a suitable tool for filtering out items and providing the most relevant and suitable items to each user, based on their individual preferences. Deep learning algorithms achieve great success in several fields including RS. The issue with deep learning-based RS models is that, they ignore the differences of users’ preferences, and they build a model based on all the users’ rates. This paper proposed an optimized clustering-based denoising autoencoder model (OCB-DAE) which trains multiple models instead of one, based on users’ preferences using k-means algorithm combined with a nature-inspired algorithm (NIA) such as artificial fish swarm algorithm to determine the optimal initial centroids to cluster the users based on their similar preferences, and each cluster trains its own denoising autoencoder (DAE) model. The results proved that combining NIA with k-means gives better clustering results comparing with using k-means alone. OCB-DAE was trained and tested with MovieLens 1M dataset where 80% of it is used for training and 20% for testing. Root mean squared error (RMSE) score was used to evaluate the performance of the proposed model which was 0.618. It outperformed the other models that use autoencoder and denoising autoencoder without clustering with 38.5% and 29.5% respectively.

show abstract

“…At present, many clustering methods have been born, and the common clustering algorithms are division-based clustering, hierarchy-based clustering, density-based clustering, grid-based clustering, and modelbased clustering algorithms [1] . Clustering algorithms have a wide range of applications in real life, such as building recommender systems, social media network analysis, and face recognition [2] . It is worth noting that traditional clustering methods are usually designed for data with a single view without considering any other relevant information from other views.…”

Section: Introductionmentioning

confidence: 99%

Incomplete Multi-View Clustering Based on Dynamic Dimensionality Reduction Weighted Graph Learning

Yu,

Sun

2024

IEEE Access

View full text Add to dashboard Cite

In practical applications, the incomplete multi-view clustering problem usually arises due to the omission of data from some views. The existing incomplete multi-view clustering methods often ignore the noise and redundancy of the original data, the valuable information hidden in the missing views, and the different importance of each view. To solve the above problems, this paper proposes incomplete multi-view clustering based on dynamic dimensionality reduction weighted graph learning (ARDGL), which is mainly divided into two parts: dynamic dimensionality reduction weighted graph learning similarity matrix and selfweighted graph fusion. Integrating the two in the same framework helps to reinforce each other. In the process of learning the similarity matrix, the noise, and redundancy of the original data are effectively filtered using dynamic descending weighted graph learning, and the influence of incomplete data on the clustering results is attenuated. In the self-weighted graph fusion process, the quality of the consensus matrix is improved by adding non-negative orthogonal constraints, and the clustering results can be obtained without post-processing. To solve the objective function, an alternating iteration algorithm is proposed. Experiments are conducted on four incomplete multi-view datasets, and the results show that the algorithm has good performance.

show abstract

Proposing improved meta-heuristic algorithms for clustering and separating users in the recommender systems

Cited by 12 publications

References 41 publications

TrustDL: Use of trust-based dictionary learning to facilitate recommendation in social networks

TrustDL: Use of trust-based dictionary learning to facilitate recommendation in social networks

Cluster-based denoising autoencoders for rate prediction recommender systems

Incomplete Multi-View Clustering Based on Dynamic Dimensionality Reduction Weighted Graph Learning

Contact Info

Product

Resources

About