Explore Courses Blog Tutorials Interview Questions
0 votes
in AI and Deep Learning by (50.2k points)

I am facing a problem in selecting the correct classifier for my data-mining task.

I am labeling webpages using statistical methods and label them using a 1-4 scale,1 being the poorest while 4 being the best.

Previously, I used SVM to train the system since I was using a binary(1,0) label then. But now since I switch to this 4-class label, I need to change classifier, because I think the SVM classifier will only work for two-class classification(Please correct me if I am wrong).

So could you please offer some suggestions here on what kind of classifier is most appropriate here for my classification purpose.

Thanks in advance for suggestions.

1 Answer

0 votes
by (108k points)

Here are some advanced algorithms that can supercharge your data mining jobs. They include the following:

  1. Advanced General-Purpose Machine Learning Algorithms

  • Interactive Trees (C&RT or CART, CHAID)

  • Boosted Tree Classifiers and Regression

  • MARSplines (Multivariate Adaptive Regression Splines)

  • Random Forests for Regression and Classification 

  • Machine Learning—Naïve Bayesian Classifier and Nearest Neighbor 

  • Statistical Learning Theory—Support Vector Machine

  • Sequence, Association, and Link Analysis

  • Independent Components Analysis

  • Kohonen Clustering

  1. Text Mining Algorithms

  2. Quality Control Data Mining and Root Cause Analysis

  • Quality Control Charts

  • Quality Control Charts for Variable Lists

  • Predictive Quality Control

  • Root Cause Analysis

  • Response Optimization for Data Mining Models

  1. Image and Object Data Mining: It contains Visualization and 3D-Medical and Other Scanning Imaging.

There exist multi-class SVMs. LibSVM has an implementation.

Usually, it's better to experiment with several classifiers to find out which one works best on your data. The selection of the classifier type and training algorithm is far less important than your choice of feature set. You could try naïve Bayes, multi-class SVM, MaxEnt, voted perceptrons, or whatever your library offers.

Browse Categories