How can a svm classifier do multi-label classification, except OVR, binary relevance, and SVM_rank?

07 July 2018 3 4K Report

Hello genius,

As described in the question, here are details about my problem.

What kind of training process do I expect?

Currently, the data set has n*3 dimensional inputs and n*4 dimensional outputs. I am expecting to train a svm classifier that can find out an entry's label(s) SIMULTANEOUSLY when inputs are given (like ANNs).

What have I tried and found?

I have tried OVR in scikit learn and svm_rank. However, OVR in scikit learn does not allow the training with multiple outputs. svm_rank requires query ID and target value, which are difficult for me to obtain. Since I can not tell which label is better. Binary relevance does not seem to be a good idea for my limited data set.

I have searched SVC as well, but I am a bit concerns about the reality of SVC prediction.

What do I need?

I need a SVM classifier that can find out an entry's label(s) SIMULTANEOUSLY when inputs are given (like ANNs).

Build my own svm is the last thing that I want to do. Thank you so much for your sharing.

Best regards,

A desperate non-CS student

Mohamed AbdElAziz Khamis

https://ukm.pure.elsevier.com/en/publications/binary-relevance-br-method-classifier-of-multi-label-classificati

http://scikit.ml/api/classify.html

http://www.cs.cmu.edu/~sgopal1/papers/MLJ11.pdf

http://tuprints.ulb.tu-darmstadt.de/3226/7/loza12diss.pdf

Article A Literature Survey on Algorithms for Multi-label Learning

Rahmad Sadli

Hi Ying Hong,

Have you looked at LibSVM?

It is an open source library developed by Chih-Chung Chang and Chih-Jen Lin since the year 2000. It can be used to classify two-class and multi-class cases. Based on its literature, libSVM uses the one-versus-one for multi-class cases.

Good luck..

https://www.csie.ntu.edu.tw/~cjlin/libsvm/

Mohsen Naji

Hi Ying,

You can use the one against one (OAO) and one against all (OAA) techniques. Multiple SVMs and data re-labeling is needed for those approaches.

What is the standardised processes of multiple groups analysis SEM?

How to improve CFI by using Amos?

Why should we choose deep learning to machine learning? Why is CNN better for image classification? Can I get Typical Answer?

What are some of the research gaps in machine learning and artificial intelligence in africa?

How to increase a model accuracy ?

Can NFL theorem be valid in infinite search space in ML?

Is there any research regarding how training and development is going to change in the era of Fourth Industrial Revolution?

Which group formation theory do you recommend for projects that aim to bring together entrepreneurs for training for regional development?

What is the difference between L1 and L2 regularization in machine learning?

Can anyone provide the Support vector regression code structure of LIBSVM MATLAB or python ?

Is there any Machine Learning/AI algorithm that lets you predict with incomplete features without missing value imputation process?

How can I represent SDS-PAGE results for machine learning analysis?