This research proposes a novel transfer learning algorithm, Noise-Label Transfer Learning (NLTL), aiming at exploiting noisy (in terms of labels and features) training data to improve the learning quality. We exploit the information from both accurate and noisy data by transferring the features into common domain and adjust the weights of instances for learning. We experiment on three University of California Irvine (UCI) datasets and one real-world dataset (Plurk) to evaluate the effectiveness of the model.
CITATION STYLE
Lin, W. S., Kuo, T. T., Huang, Y. Y., Lu, W. C., & Lin, S. D. (2014). A transfer-learning approach to exploit noisy information for classification and its application on sentiment detection. Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 8916, 262–273. https://doi.org/10.1007/978-3-319-13987-6_25
Mendeley helps you to discover research relevant for your work.