#without tears

[jt1674]

Visit Original Link

[i-roll-my-eyes]

I would like to thank all of the SPN creators for ruining all the good songs with pain, suffering and desperation

[tracyorama]

[cryptic-hq]

I cut my tongue for the krillionth time but I don't wanna get my teeth shaved down 😭

[noonetellsyouno]

Okay let's hope I can sort this all out tonight without a breakdown of some sort.

[bahasabunga]

Sajak dalam diam

Aku hanya ingin menangis di tengah rintik hujan. Berlari di padang ilalang yang tingginya menenggelamkan tinggi badanku. Berteriak bersama debur ombak musim kemarau. Merenung di tengah gulitanya malam. Agar tak seorangpun tahu aku sedang menangis, berlari, berteriak , juga merenung. Pada kenyataannya, yang kulakukan hanya menangis tanpa suara dan tergugu tanpa air mata. Aku menangis dalam diam

[deahgge]

Falling deep

I found myself, In an empty room. An empty room, With feelings. An empty room, With emotions. I opened the door, I found nothing. I closed the door, I saw tears. Fear ran through the hallway As it banged on the door. Tears began flooding As it was suffocating me. I swam across, Opening the door. I fall deep, Deep inside. I cry and cry, But never with tears.

[diggdata]

Support Vector Machine without tears

Hello!

Another addition to "without tears" series after Linear Regression without tears. There has been a lot of buzz lately about Support Vector Machine(SVM) in machine learning community after almost two decade since its proposal by Vladimir N. Vapnik in 1993. Now, it has become very important to understand what it is about?

Support Vector Machine

In machine learning, SVMs are supervised learning models with associated learning algorithms that analyze data and recognize patterns. SVMs is one of the top 10 algorithms in data mining. They are used for classification and regression analysis. We have build some understanding of regression analysis in Linear Regression without tears post but we need to first understand what is classification analysis?

Introduction to Classification analysis

If I must say classification is nothing but labeling dataset into multiple classes based on learning from pre-labelled data(which is called Training data). So, basically we are trying to find the equation of the simplest decision boundary possible which can separate the blue dots from green ones with least amount of error.

Therefore, given labeled data points find classification rule(s):

using straight lines for two-dimension feature space or Planes for three dimension and hyper-planes in general

Non-linear decision boundaries can be formed using piecewise linear functions

A simplest introduction to SVM

Implementation in R

https://gist.github.com/ankitksharma/aceef06c278e512a76eb

Important links:

An Idiot’s guide to Support vector machines (SVMs)

Support Vector Machine tutorial

Caltech Lecture on Support Vector Machines