One fundamental idea behind this approach is that there exist patterns among different users’ preferences. And we propose a linear regression model to characterize the inner relationships among different users’ rating habits.

Aug 14, 2023 · The simple linear regression approach. We approach the recommendation engine as a linear regression problem. We can formalize the problem in a linear regression model, y = wx+b.

Dec 15, 2017 · We propose a content-based filtering algorithm based on a multiattribute network. Network analysis can consider similarities among indirectly-connected items. The proposed method addresses the data sparsity and over-specialization problems.

Feb 12, 2018 · In this paper, from the perspective of regression, we propose a cross-domain collaborative filtering algorithm based on Feature Construction and Locally Weighted Linear Regression (FCLWLR).

SGD is the main training algorithm for many current machine learning methods including deep learning. The key advantage of LMS is that it can be used on-line and used adaptively. Each LMS iteration takes a new data sample xl and produces a prediction based on the current model parameter wl as.

Aug 18, 2023 · We can use these features to recommend movies/items for a user. So recommending the film we can use a linear regression model (wx+b) with a bias component. i.e wx. we can rewrite the equation...

To overcome this drawback, we propose a cross-domain collaborative filtering algorithm based on Feature Construction and Locally Weighted Linear Regression (FCLWLR). We first construct features in different domains and use these features to represent different auxiliary domains.

May 4, 2019 · This paper attempts to solve the problems with linear regression-based collaborative filtering recommendation algorithm, namely, the difficulty in extracting eigenvalues, the low accuracy and the poor interpretability.

To address this issue, inspired by the censored regression model and the maximum likelihood criterion, this paper proposes a novel robust recursive widely linear adaptive algorithm, called the widely linear maximum likelihood censored regression (WL-ML-CR).

Jul 15, 2022 · The final results show that, according to the multiple linear regression model between reader satisfaction and its influencing factors, in order to improve library service quality and build a...