Supervised Learning - Classification Week 7 Challenge
# Question 1
Pros and cons of SVM
Advantages:
* SVM works relatively well when there is a clear margin of separation between classes.
* SVM is more effective in high dimensional spaces.
* SVM is effective in cases where the number of dimensions is greater than the number of samples.
* SVM is relatively memory efficient.
Disadvantages:
* SVM algorithm is not suitable for large data sets.
* SVM does not perform very well when the data set has more noise i.e., target classes are overlapping.
* In cases where the number of features for each data points exceeds the number of training data samples, the SVM will underperform.
* As the support vector classifier works by putting data points, above and below the classifying hyperplane there is no probabilistic explaination for the classification.
# Question 2
SVM: Support Vector Machines is a classification algorithm.
Working:
Scatter plot of petal length vs petal width for Iris Setosa(black) and Iris Versicolor(Blue).
The data can be classified using several different boundary lines. SVM is a method of deciding which boundary line suits the dataset best.
The distance between the closest points (support vectors) to the partitioning line is called the margin. The algorithm will choose the line with the highest margin as the classifier.
For a model with two features, it is a line, with three it is a plane, however with n dimensions, it becomes difficult to visualize it. However, mathematically for n dimensions, the classifier is called a hyperplane.
Gamma: When deciding the hyperplane, if only the closest points are considered, then the model is said to have high gamma. However, if the further away points are also considered, the model is said to have low gamma modes tend to have accuracy issues but less computation cost.
Regularisation: Regularisation refers to how accurately the model should fit the data. A high regularisation can lead to overfitting of the dataset, leading to erroneous predictions. A low regularisation will consist of some misclassified data but will have an overall good accuracy.
Aim: The aim of this challenge is stainless steel plates dataset:
There are 34 fields. The first 27 fields describe some kind of steel plate faults.
* X_Minimum
* X_Maximum
* Y_Minimum
* Y_Maximum
* Pixels_Areas
* X_Perimeter
* Y_Perimeter
* Sum of Luminosity
* Minimum of Luminosity
* Maximum of Luminosity
* Length of conveyer
* Type of steel_A300
* Type of steel_A400
* Steel of plate thickness
* Edges_Index
* Empty_Index
* Square_Index
* Outside x index
* Edges Y index
* Outside Global Index
* Log Of Areas
* Log X Index
* Log Y Index
* Orientation_Index
* Luminousity_Index
* Sigmoid of Areas
The last seven columns are one hot encoded classes, i.e., if the plate fault is classified as "Stains" there will be a 1 in that column and 0's in the other columns. If you are unfamiliar with one-hot encoding, just know that the last seven columns are your class labels.
* Pastry
* Z_Scratch
* K_Scratch
* Stains
* Dirtiness
* Bumps
* Other_Facts