Loading...
| Friend's email | |
| Your name | |
| Your email | |
| enter code | |
This page was sent successfuly
- Type of Document: M.Sc. Thesis
- Language: Farsi
- Document No: 41479 (19)
- University: Sharif University of Technology
- Department: Computer Engineering
- Advisor(s): Beigy, Hamid
- Abstract:
- Classification is one of the most important problems in machine learning area. Reliable and successful classification is essential for diagnosing patients for further treatment. In many applications such as bioinformatics unlabeled data is abundant and available. However labeling data is much more difficult and expensive to obtain. This dissertation presents a novel transductive approach for the development of robust microarray data classification. The transduction problem is to estimate the value of classification function at the given points in the working set. This contrasts with the standard inductive learning problem of estimating the classification method at all possible values and then using the fixed function to deduce the classes of the working dataset. In such situation we do not care about the general decision function and we do not have enough information to estimate data distribution well enough. This leads to poor generalization. Our goal here is to classify set of example with as few errors as possible. Dealing with gene expression datasets has many challenges such as large number of features usually thousands of features, relatively insufficient number of labeled data, noise and etc. In this research we try to overcome insufficient training data with transductive approach
- Keywords:
- Machine Learning ; Microarray ; Bioinformatics ; Gene Expression Data ; Transductive Learning
Digital Object List-
محتواي پايان نامه - view
Bookmark