A Transfer-Learning Approach to Exploit Noisy Information for Classification and Its Application on Sentiment Detection - Technologies and Applications of Artificial Intelligence - page 272

6 Conclusion

In this paper, we propose a novel prediction problem together with a transfer learning
algorithm to solve it. We serve the objects which have multiple labels as a bridge and
transfer knowledge from different data domains. We update instance weights and
transfer features by comparing labels and features in high-quality domain and low-
quality domain simultaneously. The experiment result shows NLTL consistently out-
performs the competitors. Furthermore, we propose a real-world task of sentiment
diffusion prediction that can benefit from our framework. Our experiments demon-
strate how such problem can be formulated into a noisy-label prediction task that can
be solved using NLTL.

Acknowledgement. This work is primarily supported by a grant from Telecommuni-
cation Laboratories, Chunghwa Telecom Co., Ltd under the contract No. TL-103-
8201.

References

1. Sáez, J.A., Galar, M., Luengo, J., Herrera, F.: Tackling the Problem of Classification with
Noisy Data Using Multiple Classifier Systems: Analysis of the Performance and Robust-
ness. In: Information Science (2013)
2. Pan, S.J., Yang, Q.: A Survey on Transfer Learning. IEEE Transactions on Knowledge and
Data Engineering 22, 1345-1359 (2010)
3. Dai, W., Yang, Q., Xue, G., Yu, Y.: Boosting for Transfer Learning. In: Proceedings of the
24th International Conference on Machine Learning, pp. 193-200 (2007)
4. Blitzer, J., McDonald, R., Pereira, F.: Domain Adaptation with Structural Correspondence
Learning. In: Proceedings of the 2006 Conference on Empirical Methods in Natural
Language Processing, pp. 120-128 (2006)
5. Bache, K., Lichman, M.: UCI Machine Learning Repository. University of California,
School of Information and Computer Science, Irvine (2013)
6. Meng, X., Wei, F., Liu, X., Zhou, M., Li, S., Wang, H.: Entity-Centric Topic-Oriented
Opinion Summarization in Twitter. In: Proceedings of the 18th ACM SIGKDD Internation-
al Conference on Knowledge Discovery and Data Mining, pp. 379-387 (2012)
7. Kuo, T.-T., Hung, S.-C., Lin, W.-S., Peng, N., Lin, S.-D., Lin, W.-F.: Exploiting Latent
Information to Predict Diffusions of Novel Topics on Social Network. In: Proceedings of
the 50th Annual Meeting of the Association for Computational Linguistics, pp. 344-348
(2012)
8. Kuo, T.-T., Hung, S.-C., Lin, W.-S., Lin, S.-D., Peng, T.-C., Shih, C.-C.: Assessing the
Quality of Diffusion Models Using Real-World Social Network Data. Technologies and
Applications of Artificial Intelligence (TAAI) 2011, 200-205 (2011)
9. Kohavi, R., John, G.H.: Wrappers for feature subset selection. Artificial Intelligence 97(1),
273-324 (1997)