Quick Look At Classification
0 likes303 views
Classification assigns objects to classes based on their attributes. Naive Bayes uses probabilities to determine the most likely class, while k-nearest neighbor (k-NN) compares attributes of unknown instances to training instances to classify based on the most common class of the nearest k instances. Normalization ensures different attribute units do not impact neighbor selection, and algorithms like naive Bayes use eager learning by generalizing training data into a model beforehand, whereas k-NN uses lazy learning by only generalizing when classifying new instances.
Quick Look At Classification
- 1. Quick Look at CLASSIFICATION
- 2. ClassificationEach object is assigned to precisely one classNaïve Bayes Classifiers Uses the probabilistic theory to find the most likely classNearest neighbor classification Mainly used when all attribute values are continuous. It is also called as (k-Nearest neighbor or k-NN classification)
- 3. Basic K – NN Classification Algorithm Find ‘k’ training instances that are closest to the unknown instanceTake the most commonly occurring classification for these ‘k’ instancesThe neighbors can be weighted to improve classification
- 4. NormalizationLarge magnitudes get more weight while calculating distances and thus nearest neighbors are not properly chosen.Normalization ensures that units chosen don’t affect the selection of nearest neighbors
- 5. Eager & Lazy learningEager learningTraining data is ‘eagerly’ generalized into some representation model without waiting for unknown instances. Eg. Naïve Bayes algorithmLazy learningTraining data is not converted to a representation model until an unknown instance is presented for classification. Eg. Nearest neighbor algorithm
- 6. Visit more self help tutorialsPick a tutorial of your choice and browse through it at your own pace.The tutorials section is free, self-guiding and will not involve any additional support.Visit us at www.dataminingtools.net