Top 5 simple and robust Machine Learning Algorithms

Are you looking for Top 5 simple and robust Machine Learning Algorithms, then congratulation, you are in the right place. Today we will discuss Top 5 simple and robust Machine Learning Algorithms

Hello, & Welcome!

In this article, you will learn-

As we know, Machine learning is growing very fast and many people are shifting their careers in ML, moreover, Harvard Business review article titled ‘Data Scientist’ the ‘Sexiest Job of the 21st century’.

As a beginner, you have a question in your mind what are The Top 5 Machine Learning Algorithm.

Therefore, Without wasting your time let’s just discuss The Top 5 Machine Learning algorithms-

Top 5 simple and robust Machine Learning Algorithms

Firstly, understand the types of Machine learning algorithms, before diving into actual algorithms

1. Supervised Machine Learning Algorithm- In a supervised machine learning algorithm, we have training data, which means we train our model on a labeled dataset. We have labeled data for training our model, we provide some new data to predict the result. It is also known as the Classification algorithm.
2. Unsupervised Machine Learning Algorithm- In an unsupervised machine learning algorithm, we don’t have training data, which means a model has to learn by itself without having any prior knowledge. It is also known as the Clustering algorithm.
3. Reinforcement Machine Learning Algorithms- Reinforcement learning use hit and trial, means model learns by its own mistakes and improves their accuracy, just like in video games.

Now we have covered types of machine learning, its time to know about Top 5 simple and robust Machine Learning Algorithms

1. Naïve Bayes Classifier Algorithm –

Supervised or Unsupervised: Supervised

• It is very hard if you have to manually classify data such as documents, emails, or web pages, but Naïve Bayes Classifier does this task very easily. Naïve Bayes Classifier Algorithm is based on Bayes Theorem of Probability, which allocates the population’s element value from one of the categories that are available. Application of Naïve Bayes Classifier is spam filtering and sentiment analysis.
• If you have a large dataset then you should use Naïve Bayes Classifier.

Use Cases

• Spam Filtering: Classifying emails as spam or not spam.
• Sentiment Analysis: Determining the sentiment of text (positive, negative, neutral).

The Naïve Bayes Classifier is a probabilistic model based on Bayes’ Theorem. It assumes independence among features and is particularly effective for large datasets with categorical features.

Python Implementation

import numpy as np
from sklearn.datasets import load_iris
from sklearn.model_selection import train_test_split
from sklearn.naive_bayes import GaussianNB
from sklearn.metrics import classification_report

# Load dataset
iris = load_iris()
X = iris.data
y = iris.target

# Split dataset
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42)

# Initialize and train the Naïve Bayes classifier
model = GaussianNB()
model.fit(X_train, y_train)

# Predict and evaluate
predictions = model.predict(X_test)
print("Naïve Bayes Classifier Report:")
print(classification_report(y_test, predictions))

2. Support Vector Machine Learning Algorithm–

Supervised or Unsupervised: Supervised

• Support Vector Machine (SVM) is a supervised machine learning algorithm. It performs classification and regression problems tasks. In SVM, there is a hyperplane ( a line) which divides the data into different classes.
• SVM tries to find the hyperplane which maximizes the distance between different classes, and this term is called margin maximization that rises the chance of classifying the data more accurately.
• The application of SVM is stock market prediction.

Use Cases

• Stock Market Prediction: Classifying whether the stock price will go up or down.
• Image Classification: Identifying objects in images.

The Support Vector Machine (SVM) algorithm is used for classification and regression tasks. It works by finding the hyperplane that best separates the classes in the feature space.

Python Implementation

import numpy as np
from sklearn.datasets import load_iris
from sklearn.model_selection import train_test_split
from sklearn.svm import SVC
from sklearn.metrics import classification_report

# Load dataset
iris = load_iris()
X = iris.data
y = iris.target

# Split dataset
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42)

# Initialize and train the Support Vector Machine
model = SVC(kernel='linear', random_state=42)
model.fit(X_train, y_train)

# Predict and evaluate
predictions = model.predict(X_test)
print("Support Vector Machine Report:")
print(classification_report(y_test, predictions))

3. K Means Clustering Algorithm –

Supervised or Unsupervised: Unsupervised

• K means clustering algorithms divide each observation into a K cluster.
• It is unsupervised learning. It performs clustering. As it is a type of unsupervised learning so it starts with randomly selected centroids, it selects random K points which are centroids.
• A centroid is a virtual or exact point representing the center of the cluster.
• After that, it assigns every data to the closer centroid. You can use the Euclidean distance formula or use any other distance formula depends upon the distance type.
• K means clustering algorithm again calculate and locate the new centroid for each cluster.
• After that reassigning of data-point is performed based on the new closet centroid, if there any reassignment is performed then they repeat the previous step until the model is ready.
• K means clustering is an iterative process.

Use Cases

• Customer Segmentation: Grouping customers based on purchasing behavior.
• Image Compression: Reducing the number of colors in an image.

K-Means Clustering is an unsupervised learning algorithm that partitions data into K clusters. It iteratively updates cluster centers to minimize variance within each cluster.

Python Implementation

import numpy as np
from sklearn.datasets import load_wine
from sklearn.cluster import KMeans
import matplotlib.pyplot as plt

# Load dataset
wine = load_wine()
X_wine = wine.data

# Initialize and fit K-Means
kmeans = KMeans(n_clusters=3, random_state=42)
kmeans.fit(X_wine)
clusters = kmeans.predict(X_wine)

# Display results
print("K-Means Clustering:")
print("Cluster Centers:\n", kmeans.cluster_centers_)
print("Cluster Labels (first 10):\n", clusters[:10])

# Optional: Plot clusters
plt.scatter(X_wine[:, 0], X_wine[:, 1], c=clusters, cmap='viridis')
plt.scatter(kmeans.cluster_centers_[:, 0], kmeans.cluster_centers_[:, 1], s=300, c='red', marker='X')
plt.xlabel('Feature 1')
plt.ylabel('Feature 2')
plt.title('K-Means Clustering')
plt.show()

4. Apriori Algorithm-

Supervised or Unsupervised: Unsupervised

• Apriori algorithm is based on association rules. For example, if someone who bought milk also bought bread. Another name of the Apriori algorithm is association rule learning.
• Marketing is the main application of the Apriori algorithm because it allows looking at certain patterns in buying items. For example, if someone buys milk and bread together so put milk and bread together in a supermarket to increase the sale of both items. By putting together both items, a person who comes to buy only milk might purchase bread too.
• Apriori algorithm work on three terms- Support, Confidence and lift.
  • Support (in terms of milk and bread Example) = customer who bought milk/ total no of customer
  • Confidence = customer who bought milk & bread both/ total customer
  • Lift = Confidence/ support

Use Cases

• Market Basket Analysis: Identifying items frequently bought together.
• Recommendation Systems: Suggesting products based on previous purchases.

The Apriori Algorithm is used for association rule learning, discovering frequent itemsets and relationships between items in large transactional datasets.

Python Implementation

import pandas as pd
from mlxtend.frequent_patterns import apriori, association_rules

# Create a sample transactional dataset
dataset = pd.DataFrame({
    'Milk': [1, 1, 0, 1, 0, 1, 1],
    'Bread': [1, 1, 0, 0, 0, 1, 1],
    'Butter': [0, 1, 1, 0, 0, 1, 0],
})

# Apply Apriori algorithm
frequent_itemsets = apriori(dataset, min_support=0.5, use_colnames=True)
rules = association_rules(frequent_itemsets, metric="lift", min_threshold=1.0)

# Display results
print("Apriori Algorithm Rules:")
print(rules)

5. Decision Tree Algorithm-

Supervised or Unsupervised: Supervised

• Decision Tree is used in two cases- Classification and Regression.
• As the name suggests Decision Tree, it is a tree-like structure start from the root node to the leaf node.
• The internal or non-leaf node of the decision tree represents the test on a feature, the leaf node act as a class label.
• The decision tree checks the condition in each node and proceeds down.

Use Cases

• Medical Diagnosis: Predicting diseases based on patient data.
• Credit Scoring: Evaluating creditworthiness of individuals.

The Decision Tree Algorithm is used for both classification and regression tasks. It creates a tree-like model of decisions, making it easy to visualize and interpret.

Python Implementation

import numpy as np
from sklearn.datasets import load_iris
from sklearn.model_selection import train_test_split
from sklearn.tree import DecisionTreeClassifier
from sklearn.metrics import classification_report

# Load dataset
iris = load_iris()
X = iris.data
y = iris.target

# Split dataset
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42)

# Initialize and train the Decision Tree classifier
model = DecisionTreeClassifier(random_state=42)
model.fit(X_train, y_train)

# Predict and evaluate
predictions = model.predict(X_test)
print("Decision Tree Classifier Report:")
print(classification_report(y_test, predictions))

Are you ML Beginner and confused, from where to start ML, then read our BLOG – How do I learn Machine Learning?

Enjoy Machine Learning

All the Best!

FAQ-

Learn Basics of ML. Learn here.

Though of the Day…

‘Tell me and I forget. Teach me and I remember. Involve me and I learn.’ 

– Benjamin Franklin