22-05-2012, 01:18 PM
Homework of Evolutionary Computing
Homework of Evolutionary Computing.doc (Size: 52 KB / Downloads: 29)
Binary classification examples:
- Cancer classification using microarray experiments.
- Cancer classification using epigenetic measurements.
- Cancer classification using mass spec data.
- Protein marker identification and classification.
- Protein localization.
- QSAR Analysis.
Multiple classification examples:
- Secondary structure prediction.
- micro array data for gene function analysis
2) Find out types of domain specific kernel functions present in bio-informatics?
- Chi-square kernal
- String kernel
- Fisher kernel
- Mismatch kernel
- Graph kernel
- Histogram intersection kernal
3) What is a linear hyperplane in terms of kernel functions?
A hyperplane is a concept in geometry. It is a generalization of the plane into a different number of dimensions.
A hyperplane of an n-dimensional space is a flat subset with dimension n − 1. By its nature, it separates the space into two half spaces.kernal function help us to get the solve n-dimentional problem without projecting it to higher dimentions.
4) How do you find out best kernel function using Genetic Algorithm?
5) Write type of kernel functions used for multiple- class SVM?
- Polynomial kernel.
- Rational quadratic kernel.
- Multi quadratic kernel.
- Exponential kernel.
- Laplacian kernel.
- Bessel kernel.
- Radial basis function kernel.
- Sigmoid kernel.
6) List out types of Nuclear compartments present in a cell?
- Cajal bodies.
- Polymorphic interphase karyosomal association.
- Promyelocytic leukemia.
- Para speckles.
- Splicing speckles.
- Nucleolus.
- Chromosomes.
- Gemini of coiled bodies.
7) Give examples for single class SVM used in bio-informatics?
- protein-protein interactions
- predication of mi-RNA
8) How do you make query proteins of unequal length into equal length?
Convert the sequence into single amino acid frequency or di-peptide frequency.
9) List out motif’s finding algorithms available?
- Randomized quick sort.
- Word-based algorithms
- Greedy profile motif search.
- Gibb’s sampler.
- Random projections.
- Probabilistic algorithms
- Machine learning techniques (FMGA)
- Ensemble algorithm