The linear system with missing information is <div>investigated in this paper. New methods are </div><div>introduced to improve the Mean Squared Error (MSE) </div><div>on the test set in comparison to state-of-the-art method</div><div>s, through appropriate tuning of Bias-Variance </div><div>trade-off. The concept is to cluster the data and </div><div>adapt the learning model to each cluster. Hence, </div><div>we set forth a controlled bias into the problem and </div><div>positively utilize it to enhance learning capability on </div><div>the instances considered in some specific </div><div>neighborhood. To deal with missing infrormation, </div><div>we propose a novel algorithm "Missing-SCOP" based </div><div>on SCOP-KMEANS algorithm introduced by Wagstaff,</div><div> et al., utilizing the missing pattern of the dataset for </div><div>construction of a soft-constraint matrix and clustering </div><div>in missing scenario. It is shown that controlled </div><div>over-fitting suggested by our algorithm improves </div><div>prediction accuracy in various cases. </div><div>Numerical experiments approve the efficacy of our</div><div> proposed algorithm in enhancing the prediction</div><div> accuracy.</div>
The linear system with missing information is <div>investigated in this paper. New methods are </div><div>introduced to improve the Mean Squared Error (MSE) </div><div>on the test set in comparison to state-of-the-art method</div><div>s, through appropriate tuning of Bias-Variance </div><div>trade-off. The concept is to cluster the data and </div><div>adapt the learning model to each cluster. Hence, </div><div>we set forth a controlled bias into the problem and </div><div>positively utilize it to enhance learning capability on </div><div>the instances considered in some specific </div><div>neighborhood. To deal with missing infrormation, </div><div>we propose a novel algorithm "Missing-SCOP" based </div><div>on SCOP-KMEANS algorithm introduced by Wagstaff,</div><div> et al., utilizing the missing pattern of the dataset for </div><div>construction of a soft-constraint matrix and clustering </div><div>in missing scenario. It is shown that controlled </div><div>over-fitting suggested by our algorithm improves </div><div>prediction accuracy in various cases. </div><div>Numerical experiments approve the efficacy of our</div><div> proposed algorithm in enhancing the prediction</div><div> accuracy.</div>
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.