Which Machine Learning Algorithm Should You Use By Problem Type?

When I was beginning my way in data science, I often faced the problem of choosing the most appropriate algorithm for my specific problem. If you’re like me, when you open some article about machine learning algorithms, you see dozens of detailed descriptions. The paradox is that they don’t ease the choice.

Well, to not let you feel out of the track, I would suggest you to have a good understanding of the implementation and mathematical intuition behind several supervised and unsupervised Machine Learning Algorithms like -

• Linear regression
• Logistic regression
• Decision tree
• Naive Bayes
• Support vector machine
• Random forest
• AdaBoost
• Gradient-boosting trees
• Simple neural network
• Hierarchical clustering
• Gaussian mixture model
• Convolutional neural network
• Recurrent neural network
• Recommender system

Remember, the list of Machine Learning Algorithms I mentioned are the ones that are mandatory to have a good knowledge of , while you are a beginner in Machine/Deep Learning !

Now that we have some intuition about types of machine learning tasks, let’s explore the most popular algorithms with their applications in real life, based on their problem statements !

Try to work on each of these problem statements after getting to the end of this blog ! I can assure you would learn a lot, a hell lot!

Problem Statement 1 -

To Predict the Housing Prices

Machine Learning Algorithm(s) to solve the problem —

• Advanced regression techniques like random forest and gradient boosting

Problem Statement 2 -

Explore customer demographic data to identify patterns

Machine Learning Algorithm(s) to solve the problem —

• Clustering (elbow method)

Problem Statement 3 -

Predicitng Loan Repayment

Machine Learning Algorithm(s) to solve the problem —

• Classification Algorithms for imbalanced dataset

Problem Statement 4 -

Predict if a skin lesion is benign or malignant based on its characteristics (size, shape, color, etc)

Machine Learning Algorithm(s) to solve the problem —

• Convolutional Neural Network ( U-Net being the best for segmentation stuffs)

Problem Statement 5 -

Predict client churn

Machine Learning Algorithm(s) to solve the problem —

• Linear discriminant analysis (LDA) or Quadratic discriminant analysis (QDA)

( particularly popular because it is both a classifier and a dimensionality reduction technique)

Problem Statement 6 -

Provide a decision framework for hiring new employees

Machine Learning Algorithm(s) to solve the problem —

• Decision Tree is a pro gamer here

Problem Statement 7 -

Understand and predict product attributes that make a product most likely to be purchased

Machine Learning Algorithm(s) to solve the problem —

• Logistic Regression
• Decision Tree

Problem Statement 8 -

Analyze sentiment to assess product perception in the market.

Machine Learning Algorithm(s) to solve the problem —

• Naive Bayes — Support Vector Machines (NBSVM)

Problem Statement 9 -

Create classification system to filter out spam emails

Machine Learning Algorithm(s) to solve the problem —

• Classification Algorithms —

Naive Bayes, SVM , Multilayer Perceptron Neural Networks (MLPNNs) and Radial Base Function Neural Networks (RBFNN) suggested.

Problem Statement 10 -

Predict how likely someone is to click on an online ad

Machine Learning Algorithm(s) to solve the problem —

• Logistic Regression
• Support Vector Machines

Problem Statement 11 -

Detect fraudulent activity in credit-card transactions.

Machine Learning Algorithm(s) to solve the problem —

• Adaboost
• Isolation Forest
• Random Forest

Problem Statement 12 -

Predict the price of cars based on their characteristics

Machine Learning Algorithm(s) to solve the problem —

• Gradient-boosting trees are best at this.

Problem Statement 13 -

Predict the probability that a patient joins a healthcare program

Machine Learning Algorithm(s) to solve the problem —

• Simple neural networks

Problem Statement 14 -

Predict whether registered users will be willing or not to pay a particular price for a product.

Machine Learning Algorithm(s) to solve the problem —

• Neural Networks

Problem Statement 15 -

Segment customers into groups by distinct charateristics (eg, age group)

Machine Learning Algorithm(s) to solve the problem —

• K-means clustering

Problem Statement 16 -

Feature extraction from speech data for use in speech recognition systems

Machine Learning Algorithm(s) to solve the problem —

• Gaussian mixture model

Problem Statement 17 -

Object tracking of multiple objects, where the number of mixture components and their means predict object locations at each frame in a video sequence.

Machine Learning Algorithm(s) to solve the problem —

• Gaussian mixture model

Problem Statement 18 -

Organizing the genes and samples from a set of microarray experiments so as to reveal biologically interesting patterns.

Machine Learning Algorithm(s) to solve the problem —

• Hierarchical clustering algorithms

Problem Statement 19 -

Recommend what movies consumers should view based on preferences of other customers with similar attributes.

Machine Learning Algorithm(s) to solve the problem —

• Recommender system

Problem Statement 20 -

Recommend news articles a reader might want to read based on the article she or he is reading.

Machine Learning Algorithm(s) to solve the problem —

• Recommender system

Problem Statement 21 -

Recommend news articles a reader might want to read based on the article she or he is reading.

Machine Learning Algorithm(s) to solve the problem —

• Recommender system

Problem Statement 22 -

Optimize the driving behavior of self-driving cars

Machine Learning Algorithm(s) to solve the problem —

• Reinforcement Learning

Problem Statement 23 -

Diagnose health diseases from medical scans.

Machine Learning Algorithm(s) to solve the problem —

• Convolutional Neural Networks

Problem Statement 24 -

Balance the load of electricity grids in varying demand cycles

Machine Learning Algorithm(s) to solve the problem —

• Reinforcement Learning

Problem Statement 25 -

When you are working with time-series data or sequences (eg, audio recordings or text)

Machine Learning Algorithm(s) to solve the problem —

• Recurrent neural network
• LSTM

Problem Statement 26 -

Provide language translation

Machine Learning Algorithm(s) to solve the problem —

• Recurrent neural network

Problem Statement 27 -

Generate captions for images

Machine Learning Algorithm(s) to solve the problem —

• Recurrent neural network

Problem Statement 28 -

Power chatbots that can address more nuanced customer needs and inquiries

Machine Learning Algorithm(s) to solve the problem —

• Recurrent neural network

I hope that I could explain to you common perceptions of the most used machine learning algorithms and give intuition on how to choose one for your specific problem.

If you are a beginner in Data Science and Machine Learning and have some specific queries with regard to Data Science/ML-AI, guidance for Career Transition to Data Science, Interview/Resume Preparation or even want to get a Mock Interview before your D-Day, feel free to book a 1:1 call here. I will be happy to help!

Happy Machine Learning ! :)

Until next time..!

Contact me via LinkedIn