machine learning - Data complexity measure -


as strive explain accuracy of machine learning algorithms, many authors suggest start degree of complexity in data.

i working in data complexity measure like: class separability,overlapping , outlier measure, affect classifier performance.

for example if classes more separable, accuracy of classifier increased , classifier takes less time.

i want calculate outlier measure of each data point in dataset , combine these measure , make 1 measure outlier in dataset. how can calculate outlier using k nearest neighbour or using k means clustering technique. thanx in advance..


Comments

Popular posts from this blog

c# - Binding a comma separated list to a List<int> in asp.net web api -

how to prompt save As Box in Excel Interlop c# MVC 4 -

xslt 1.0 - How to access or retrieve mets content of an item from another item? -