Trjtdtk

Difference between sprint backlog and sprint goal?

I would say: "You are another teacher", but she is a woman and I am a man

How can saying a song's name be a copyright violation?

Why does Optional.map make this assignment work?

Asymptotics of orbits on graphs

Is it canonical bit space?

Would Slavery Reparations be considered Bills of Attainder and hence Illegal?

Forgetting the musical notes while performing in concert

How do I deal with an unproductive colleague in a small company?

How would I stat a creature to be immune to everything but the Magic Missile spell? (just for fun)

Why does ы have a soft sign in it?

Blender 2.8 I can't see vertices, edges or faces in edit mode

What killed these X2 caps?

What is the difference between 仮定 and 想定?

Contemplating leaving job of two years, but a long travel is in the works. Is it professional for me to resign?

Mathematica command that allows it to read my intentions

Is it acceptable for a professor to tell male students to not think that they are smarter than female students?

Short story with a alien planet, government officials must wear exploding medallions

Why is this clock signal connected to a capacitor to gnd?

Aircraft with solar-panels?

Saudi Arabia Transit Visa

Is it inappropriate for a student to attend their mentor's dissertation defense?

What reasons are there for a Capitalist to oppose a 100% inheritance tax?

ssTTsSTtRrriinInnnnNNNIiinngg

Ensemble technique for combining predictions from classification and regression algorithms

1

$begingroup$

Given an anomaly detection problem - A, I have divided the problem into two independent subtasks -A1 and A2 and tackled the tasks using classification and regression algorithms respectively. In order to combine the predictions from the classification and regression algorithms, I have derived an average weighing formula. My question is if this approach can be regarded as ensemble learning? Otherwise, are there better ways to combine such predictions into a common score? I am new here :)

UPDATE
Subtasks A1 and A2 are tackled with One Class SVM (unsupervised) and Gaussian Process Regression (GPR)(supervised) respectively. The scores from both algorithms are added using weighed mean like so:

Final_Score = alpha * OCSVM_Score + (1 - alpha) * GPR_Score

where alpha = 0.5

classification regression anomaly-detection ensemble-learning

share|improve this question

edited Mar 27 at 11:09

SyCode

asked Mar 26 at 14:57

SyCodeSyCode

114

$endgroup$

• 
  
  
  

  
  

  
  
  
  
  
  

  
  $begingroup$
  can you add a bit more details :-) e.g. is this supervised? what's the output of your regression? and how do you average the results?
  $endgroup$
  – oW_
  Mar 26 at 21:34
  
  

  
  
  
  


• 
  
  
  

  
  

  
  
  
  
  
  

  
  $begingroup$
  @oW_ i have updated the question with the information you requested, does it make more sense ?
  $endgroup$
  – SyCode
  Mar 27 at 9:12
  

  
  
  
  

add a comment | 

1

$begingroup$

Given an anomaly detection problem - A, I have divided the problem into two independent subtasks -A1 and A2 and tackled the tasks using classification and regression algorithms respectively. In order to combine the predictions from the classification and regression algorithms, I have derived an average weighing formula. My question is if this approach can be regarded as ensemble learning? Otherwise, are there better ways to combine such predictions into a common score? I am new here :)

UPDATE
Subtasks A1 and A2 are tackled with One Class SVM (unsupervised) and Gaussian Process Regression (GPR)(supervised) respectively. The scores from both algorithms are added using weighed mean like so:

Final_Score = alpha * OCSVM_Score + (1 - alpha) * GPR_Score

where alpha = 0.5

classification regression anomaly-detection ensemble-learning

share|improve this question

edited Mar 27 at 11:09

SyCode

asked Mar 26 at 14:57

SyCodeSyCode

114

$endgroup$

• 
  
  
  

  
  

  
  
  
  
  
  

  
  $begingroup$
  can you add a bit more details :-) e.g. is this supervised? what's the output of your regression? and how do you average the results?
  $endgroup$
  – oW_
  Mar 26 at 21:34
  
  

  
  
  
  


• 
  
  
  

  
  

  
  
  
  
  
  

  
  $begingroup$
  @oW_ i have updated the question with the information you requested, does it make more sense ?
  $endgroup$
  – SyCode
  Mar 27 at 9:12
  

  
  
  
  

add a comment | 

$begingroup$

Given an anomaly detection problem - A, I have divided the problem into two independent subtasks -A1 and A2 and tackled the tasks using classification and regression algorithms respectively. In order to combine the predictions from the classification and regression algorithms, I have derived an average weighing formula. My question is if this approach can be regarded as ensemble learning? Otherwise, are there better ways to combine such predictions into a common score? I am new here :)

UPDATE
Subtasks A1 and A2 are tackled with One Class SVM (unsupervised) and Gaussian Process Regression (GPR)(supervised) respectively. The scores from both algorithms are added using weighed mean like so:

Final_Score = alpha * OCSVM_Score + (1 - alpha) * GPR_Score

where alpha = 0.5

classification regression anomaly-detection ensemble-learning

share|improve this question

edited Mar 27 at 11:09

SyCode

asked Mar 26 at 14:57

SyCodeSyCode

114

$endgroup$

Final_Score = alpha * OCSVM_Score + (1 - alpha) * GPR_Score

share|improve this question

edited Mar 27 at 11:09

SyCode

asked Mar 26 at 14:57

SyCodeSyCode

114

share|improve this question

edited Mar 27 at 11:09

SyCode

asked Mar 26 at 14:57

SyCodeSyCode

114

asked Mar 26 at 14:57

SyCodeSyCode

114

asked Mar 26 at 14:57

SyCodeSyCode

114