Quick Review of Machine Learning Algorithms

Quick Review of Machine Learning Algorithms

There are 3 types of Machine Learning Algorithms:

[1] Supervised Learning: target_variable = f(independent_variables)

o   Given target/dependent variable

o   Independent variables

o   Training is run until desired level of accuracy

Examples of Supervised Learning:

·         Regression

·         Decision Tree

·         Random Forest

·         KNN

·         Logistic Regression

[2] Unsupervised Learning: clustering population in different groups

o   Target/independent variable is not given

Examples of Unsupervised Learning:

·         Apriori algorithm

·         K-means

[3] Reinforcement Learning: semi-supervised learning

The machine trains itself continually using trial and error to an exposed environment.

Example of Reinforcement Learning:

·         Markov Decision Process

                           Used Library for different machine learning algorithms

v  Linear Regression

from sklearn.linear_model import LinearRegression

v  Logistic Regression

from sklearn.linear_model import LogisticRegression

v  Decision Tree

from sklearn.tree import DecisionTreeClassifier

v  SVM

from sklearn.svm import SVC

v  Naive Bayes

from sklearn.naive_bayes import GaussianNB

v  KNN

from sklearn.neighbors import KNeighborsClassifier

v  K-Means

from sklearn.cluster import KMeans

v  Random Forest

from sklearn.ensemble import RandomForestClassifier

v  Dimensionality Reduction Algorithms

from sklearn.decomposition import PCA

v  Gradient Boosting algorithms

Ø  GBM

from sklearn.ensemble import GradientBoostingClassifier

Ø  XGBoost

from xgboost import XGBClassifier

Ø  LightGBM

Ø  CatBoost

from catboost import CatBoostRegressor