Pattern is simple classification methods where in

Pattern recognition is the process of efficiently
detecting any patterns or regularities in the given data. The machine learning
processes can be unsupervised or supervised learning where unsupervised
learning is done without a supervisor and the later with supervisor. Clustering
is an example of unsupervised learning while classification is supervised
learning. The processes can be parametric where in the data is summarised by a
set of parameters. Even if the data set is big it will not produce any change
in the parameters. Linear discriminant analysis and Quadratic discriminant are
parametric classification algorithms. In non parametric the parameters list
grows with the increase of data size.Classification assigns instances to predefined classes based on features.
It analyses and learn association between the features from the training data
to classify the unknown variables. Decision trees, a classification method
divides the search space into subsets using divide and conquer technique.
Giving grades for students is a simple classification problem. In reality
classification is teaching computer to do classification from the derived
knowledge. The classification is predefined and clear. Linear regression is simple
classification methods where in relationship between observed variables are modeled
2.The input data are categorised into training data and test data. Training
data comprises of representative data from a known category and the test data
is unknown data. A feature extractor is used to extract features from input
data. Features are the parameters or explanatory variables most relevant to the
problem extracted from observations. It can be either categorical, ordinal,
integer or real valued and is represented as a vector. When applied in
bioinformatics the vector consist of frequency of nucleotides such as A, T, G,
C or its 2-mer, 3-mer etc. Dimensionality reduction techniques are implemented
to reduce the number of features. 
Feature selection is another pre processing methods used to filter
features to remove unwanted and redundant data and include most relevant or
quality data to produce reliable output. A trainer/classifier, implements any
of the clustering or classification algorithm and maps input to the
corresponding class. The whole process is represented in the diagram given