Biomarker Prediction for Cancer Using Genetic Algorithms and Neural Networks
碩士 === 元智大學 === 資訊工程學系 === 95 === In this thesis, a method to classify proteomic data and to find biomarker for cancer prediction is proposed. The proposed method includes two stages: feature extraction and classification. In the stage of feature extraction, Genetic Algorithm (GA) is used to search...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Others |
| Language: | zh-TW |
| Published: |
2007
|
| Online Access: | http://ndltd.ncl.edu.tw/handle/62316939198056994939 |
| Summary: | 碩士 === 元智大學 === 資訊工程學系 === 95 === In this thesis, a method to classify proteomic data and to find biomarker for cancer prediction is proposed. The proposed method includes two stages: feature extraction and classification. In the stage of feature extraction, Genetic Algorithm (GA) is used to search all possible subsets of a large set of candidate features (M/Z ratios) and the multilayer feedforward neural network is used as a measure of fitness for these subsets, which is in turn used by the GA to develop subsets of salient features. In the stage of classification, the selected features by GA/ANN are applied to test data to show that they can successfully distinguish cancer symptom.
Various experiments have been conducted to prove that the proposed GA/ANN method has almost 100% classification accuracy.
|
|---|