A Transfer-Learning Approach to Exploit Noisy Information for Classification and Its Application on Sentiment Detection

Autor:	Wei-Shih Lin, Wan-Chen Lu, Shou-De Lin, Tsung-Ting Kuo, Yu-Yang Huang
Rok vydání:	2014
Předmět:	Training set Exploit Computer science business.industry media_common.quotation_subject Machine learning computer.software_genre Domain (software engineering) Quality (business) Noise (video) Artificial intelligence Data mining Transfer of learning business computer Noisy data media_common
Zdroj:	Technologies and Applications of Artificial Intelligence ISBN: 9783319139869
DOI:	10.1007/978-3-319-13987-6_25
Popis:	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.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::b98d53138611ed359b5c5c9df0d01347 https://doi.org/10.1007/978-3-319-13987-6_25 Zobrazit plný text záznamu