Which Machine Learning Algorithm Should I Use?
Now that we’ve reviewed the main types of machine learning
algorithms, it’s only natural to wonder, “Which algorithm should
I use?” The answer to the question varies depending on many
factors, including:
The size, quality and nature of data.
The available computational time.
The urgency of the task.
What you want to do with the data.
Even an experienced data scientist cannot tell which algorithm
will perform the best before trying different algorithms. Starting
with these popular machine learning algorithms is always a good
first step, especially for those new to machine learning.
Here is a list of 15 prominent machine learning algorithms,
spanning both supervised and unsupervised learning paradigms:
Linear Regression: A supervised algorithm used for predicting
continuous target variables based on a linear relationship with
input features.
Logistic Regression: A supervised algorithm primarily used for
binary classification problems, predicting the probability of an
event occurring.
Decision Trees: A supervised algorithm that creates a tree-like
model of decisions and their possible consequences, used for both
classification and regression.
Random Forest: An ensemble supervised algorithm that builds
multiple decision trees and merges their predictions to improve
accuracy and reduce overfitting.
Support Vector Machines (SVM): A supervised algorithm for
classification and regression, finding an optimal hyperplane to
separate data points into different classes.
K-Nearest Neighbors (KNN): A non-parametric supervised
algorithm used for classification and regression, where the
classification of a data point is determined by the majority class
of its 'k' nearest neighbors.
Naive Bayes: A probabilistic supervised algorithm based on
Bayes' theorem, commonly used for classification tasks,
especially in text classification.
Gradient Boosting Machines (GBM): An ensemble supervised
algorithm that builds a strong predictive model by sequentially
adding weaker models (typically decision trees) and correcting
the errors of previous models.
XGBoost: A highly optimized and efficient implementation of
gradient boosting, widely used for its speed and performance in
various machine learning competitions.
LightGBM: Another gradient boosting framework known for its
speed and efficiency, particularly effective with large datasets.
CatBoost: A gradient boosting library that handles categorical
features automatically and effectively, often providing good
performance.
K-Means Clustering: An unsupervised algorithm for
partitioning a dataset into 'k' distinct clusters, where each data
point belongs to the cluster with the nearest mean.
Hierarchical Clustering: An unsupervised algorithm that builds
a hierarchy of clusters, either by starting with individual data
points and merging them (agglomerative) or by starting with one
large cluster and splitting it (divisive).
DBSCAN (Density-Based Spatial Clustering of Applications
with Noise): An unsupervised algorithm for discovering clusters
of varying shapes and sizes in a dataset, based on the density of
data points.
Principal Component Analysis (PCA): An unsupervised
dimensionality reduction technique that transforms high-
dimensional data into a lower-dimensional representation while
retaining most of the variance.