@article { author = {Miri, Malihe and Sadeghi, Mohammad Taghi and Abootalebi, Vahid}, title = {Two Stages Sparse Representation-based Classifier and its Application for Cancer Classification}, journal = {Iranian Journal of Biomedical Engineering}, volume = {8}, number = {1}, pages = {45-56}, year = {2014}, publisher = {Iranian Society for Biomedical Engineering}, issn = {5869-2008}, eissn = {9685-8006}, doi = {10.22041/ijbme.2014.13555}, abstract = {Successful outcomes of Sparse Representation-based Classifier (SRC) and Sparse Subspace Clustering (SSC) in many applications motivated us to combine these methods and propose a hierarchical classifier. The main idea behind the SRC and SSC algorithms is to represent a data using a sparse linear combination of elementary signals so that those elementary signals which are similar to the data contribute mainly in the representation. In this paper, the performance of a sparse representation based classifier is improved by pre-clustering of training samples using the SSC algorithm. A twostage SRC is then designed using the resulting clusters. A test data is classified by first determining the most similar cluster. The data label is subsequently found using the second stage classifier. The performance of the proposed method is evaluated considering cancer classification problem using the 14-Tumors microarray dataset. Due to low number of data samples per each class and high dimensionality of the data, this is a challenging problem. Curse of dimensionality, overfitting of the classifier to the training data and computational complexity are the possible related problems. Our experimental results show that the proposed method outperforms some other state of the art classifiers.}, keywords = {Sparse Subspace Clustering,Microarray data,Cancer classification,Hierarchical classifier,Sparse Representation-based Classification,sparse representation}, title_fa = {طبقه‌بندی کننده دومرحله‌ای مبتنی بر نمایش تنک و کاربرد آن در تشخیص سرطان}, abstract_fa = {ﺑﺎ ﺗﻮﺟﻪ ﺑﻪ ﻧﺘﺎﻳﺞ ﻣﻮﻓﻘﻴﺖﺁﻣﻴﺰ ﻃﺒﻘﻪﺑﻨﺪیﻛﻨﻨﺪﻩ ﻣﺒﺘﻨﻰ ﺑﺮ ﻧﻤﺎﻳﺶ ﺗﻨﮏ (SRC) ﻭ ﺧﻮﺷﻪﺑﻨﺪی ﺯﻳﺮﻓﻀﺎی ﺗﻨﮏ (SSC) ﺩﺭ ﻛﺎﺭﺑﺮﺩﻫﺎی ﻣﺨﺘﻠﻒ، ﺩﺭ ﺍﻳﻦ ﻣﻘﺎﻟﻪ ﺑﺎ ﺗﺮﻛﻴﺐ ﺍﻳﻦ ﺩﻭ ﺭﻭﺵ، ﻳﮏ ﺭﻭﺵ ﻃﺒﻘﻪﺑﻨﺪی ﺳﻠﺴﻠﻪ ﻣﺮﺍﺗﺒﻰ ﺍﺭﺍﺋﻪ ﻣﻰﺷﻮﺩ. ﺍﻳﺪﻩ ﺍﺻﻠﻰ ﺩﺭ ﺭﻭﺵﻫﺎی ﻃﺒﻘﻪﺑﻨﺪی ﻭ ﺧﻮﺷﻪﺑﻨﺪی ﻣﺒﺘﻨﻰ ﺑﺮ ﻧﻤﺎﻳﺶ ﺗﻨﮏ، ﻧﻤﺎﻳﺶ ﻫﺮ ﺩﺍﺩﻩ ﺑﻪ ﺻﻮﺭﺕ ﺗﺮﻛﻴﺐ ﺧﻄﻰ ﺗﻨﮏ ﺍﺯ ﺳﺎﻳﺮ ﺩﺍﺩﻩﻫﺎ ﺍﺳﺖ ﺑﻪ ﮔﻮﻧﻪﺍی ﻛﻪ ﺩﺍﺩﻩﻫﺎی ﻣﺸﺎﺑﻪ ﺑﺎ ﺩﺍﺩﻩ ﻣﻮﺭﺩ ﻧﻈﺮ ﺩﺭ ﺍﻳﻦ ﺗﺮﻛﻴﺐ ﺧﻄﻰ ﺑﻴﺸﺘﺮﻳﻦ ﻭﺯﻥ ﺭﺍ ﺑﻪ ﺧﻮﺩ ﺍﺧﺘﺼﺎﺹ ﺩﻫﻨﺪ. ﺩﺭ ﺭﻭﺵ ﭘﻴﺸﻨﻬﺎﺩی، ﺑﻪ ﻣﻨﻈﻮﺭ ﺩﺳﺖﻳﺎﺑﻰ ﺑﻪ ﺻﺤﺖ ﻃﺒﻘﻪﺑﻨﺪی ﺑﻴﺸﺘﺮ، ﺍﺑﺘﺪﺍ ﺩﺍﺩﻩﻫﺎی ﺁﻣﻮﺯﺷﻰ ﺑﺎ ﺍﺳﺘﻔﺎﺩﻩ ﺍﺯ ﺭﻭﺵ ﺧﻮﺷﻪﺑﻨﺪی ﺯﻳﺮﻓﻀﺎی ﺗﻨﮏ ﺑﺨﺶﺑﻨﺪی ﻣﻰﺷﻮﻧﺪ. ﺳﭙﺲ ﺑﺎ ﺍﺳﺘﻔﺎﺩﻩ ﺍﺯ ﺷﻴﻮﮤ ﺑﻜﺎﺭ ﮔﺮﻓﺘﻪ ﺷﺪﻩ ﺩﺭ ﻃﺒﻘﻪﺑﻨﺪیﻛﻨﻨﺪﻩ ﻣﺒﺘﻨﻰ ﺑﺮ ﻧﻤﺎﻳﺶ ﺗﻨﮏ، ﻃﺒﻘﻪﺑﻨﺪیﻛﻨﻨﺪﻩﺍی ﺩﻭ ﻣﺮﺣﻠﻪﺍی ﻃﺮﺍﺣﻰ ﻣﻰﺷﻮﺩ. ﺩﺭ ﻣﺮﺣﻠﺔ ﺍﻭﻝ، ﺧﻮﺷﻪﺍی ﻛﻪ ﺩﺍﺩﻩ ﻭﺭﻭﺩی ﺑﻴﺸﺘﺮﻳﻦ ﺷﺒﺎﻫﺖ ﺭﺍ ﺑﺎ ﺁﻥ ﺩﺍﺭﺩ ﺗﻌﻴﻴﻦ ﺷﺪﻩ ﻭ ﺩﺭ ﻣﺮﺣﻠﻪ ﺑﻌﺪ ﻃﺒﻘﺔ ﻣﺮﺑﻮﻃﻪ ﺑﺮﭼﺴﺐ ﺩﺍﺩﻩ) ﺗﻌﻴﻴﻦ ﻣﻰﺷﻮﺩ. ﺑﺮﺍی ﺍﺭﺯﻳﺎﺑﻰ ﺭﻭﺵ ﭘﻴﺸﻨﻬﺎﺩی ﺍﺯ ﺩﺍﺩﮔﺎﻥ ﺭﻳﺰﺁﺭﺍﻳﻪ Tumors-14 -ﻛﻪ ﺣﺎﻭی ﺍﻃﻼﻋﺎﺕ ﻣﺮﺑﻮﻁ ﺑﻪ ﻧﻮﻉ ﺳﺮﻃﺎﻥ ﻣﺨﺘﻠﻒ ﺍﺳﺖ- ﺍﺳﺘﻔﺎﺩﻩ ﺷﺪﻩ ﺍﺳﺖ. ﺍﺯ ﺟﻤﻠﻪ ﻭﻳﮋﮔﻰﻫﺎی ﺍﻳﻦ ﺩﺍﺩﮔﺎﻥ ﺗﻌﺪﺍﺩ ﺯﻳﺎﺩ ﺑﻌﺪ ﺩﺭ ﻣﻘﺎﺑﻞ ﺗﻌﺪﺍﺩ ﻛﻢ ﻧﻤﻮﻧﻪ ﺩﺭ ﺍﺳﺖ ﻛﻪ ﻋﻤﻞ ﻃﺒﻘﻪﺑﻨﺪی ﺁﻥﻫﺎ ﺭﺍ ﺑﻪ ﻣﺴﺄﻟﻪﺍی ﭼﺎﻟﺶﺑﺮﺍﻧﮕﻴﺰ ﺗﺒﺪﻳﻞ ﻣﻰﻛﻨﺪ. ﺍﺑﻌﺎﺩ ﺯﻳﺎﺩ ﺩﺍﺩﻩﻫﺎ ﻧﻪ ﺗﻨﻬﺎ ﻣﺸﻜﻼﺗﻰ ﺍﺯ ﺟﻤﻠﻪ ﻧﻔﺮﻳﻦ ﺍﺑﻌﺎﺩ ﻭ ﺑﻴﺶ ﺍﻧﻄﺒﺎﻕ ﻃﺒﻘﻪﺑﻨﺪیﻛﻨﻨﺪﻩ ﺑﻪ ﺩﺍﺩﻩﻫﺎی ﺁﻣﻮﺯﺷﻰ ﺭﺍ ﺑﻪ ﺩﻧﺒﺎﻝ ﺩﺍﺭﺩ، ﺑﻠﻜﻪ ﺑﺎﻋﺚ ﺍﻓﺰﺍﻳﺶ ﭘﻴﭽﻴﺪﮔﻰ ﻣﺤﺎﺳﺒﺎﺗﻰ ﺷﺪﻩ؛ ﺯﻣﺎﻥ ﻻﺯﻡ ﺭﺍ ﺑﺮﺍی ﺍﺟﺮﺍی ﺍﻟﮕﻮﺭﻳﺘﻢﻫﺎ ﺍﻓﺰﺍﻳﺶ ﻣﻰﺩﻫﺪ. ﺁﺯﻣﺎﻳﺶﻫﺎی ﺍﻧﺠﺎﻡ ﺷﺪﻩ ﺑﺮ ﺍﻳﻦ ﺩﺍﺩﮔﺎﻥ ﺑﺎ ﺍﺳﺘﻔﺎﺩﻩ ﺍﺯ ﺭﻭﺵ ﭘﻴﺸﻨﻬﺎﺩی ﻧﺸﺎﻥ ﻣﻰﺩﻫﺪ ﻛﻪ ﺩﺭ ﻣﻘﺎﻳﺴﻪ ﺑﺎ ﺳﺎﻳﺮ ﺭﻭﺵﻫﺎی ﻃﺒﻘﻪﺑﻨﺪی، ﺍﻳﻦ ﺭﻭﺵ ﺑﻪ ﻧﺘﺎﻳﺞ ﺑﻬﺘﺮی ﻣﻨﺠﺮ ﻣﻰﺷﻮﺩ.}, keywords_fa = {ﺧﻮﺷﻪﺑﻨﺪی ﺯﻳﺮﻓﻀﺎی ﺗﻨﮏ,ﺩﺍﺩﮔﺎﻥ ﺭﻳﺰﺁﺭﺍﻳﻪ,ﻃﺒﻘﻪﺑﻨﺪی ﺳﺮﻃﺎﻥ,ﻃﺒﻘﻪﺑﻨﺪیﻛﻨﻨﺪﻩ ﺳﻠﺴﻠﻪ ﻣﺮﺍﺗﺒﻰ,ﻃﺒﻘﻪﺑﻨﺪیﻛﻨﻨﺪﻩ ﻣﺒﺘﻨﻰ ﺑﺮ ﻧﻤﺎﻳﺶ ﺗﻨﮏ,ﻧﻤﺎﻳﺶ ﺗﻨﮏ}, url = {https://www.ijbme.org/article_13555.html}, eprint = {https://www.ijbme.org/article_13555_ea49dec81a974a3363aed0f68158f594.pdf} }