Subscribe to DSC Newsletter

Which machine learning / deep learning algorithm to use by problem type

I like to approach algorithms from the perspective of problem solving. I created this list from a Mc Kinsey document (link below). It’s a good indicative approach. In practice,

Predict housing prices

Regression(supervised)

explore customer demographic data to identify patterns

Unsupervised learning

 

Understand product-sales drivers such as competition prices, distribution, advertisement, etc

Linear regression

 

Classify customers based on how likely they are to repay a loan

Logistic regression

 

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

Logistic regression

 

Predict client churn

Linear/quadratic discriminant analysis

Predict a sales lead’s likelihood of closing

Linear/quadratic discriminant analysis

Provide a decision framework for hiring new employees

Decision tree

 

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

Decision tree

 

eg, if an email contains theword “money,” then the probability of it being spam is high

Naive Bayes

 

Analyze sentiment to assess product perception in the market

Naive Bayes

 

Create classifiers to filter spam emails

Naive Bayes

 

Predict how many patients a hospital will need to serve in a time period

Support vector machine

 

Predict how likely someone is to click on an online ad

Support vector machine

 

Predict call volume in call centers for staffing decisions

Random forest

Predict power usage in an electrical- distribution grid

Random forest

Detect fraudulent activity in credit-card transactions.

AdaBoost

 

Simple, low-cost way to classify images (eg, recognize land usage from satellite images for climate-change models).

AdaBoost

 

Forecast product demand and inventory levels

Gradient-boosting trees

 

Predict the price of cars based on their characteristics (eg, age and mileage)

Gradient-boosting trees

 

Predict the probability that a patient joins a healthcare program

Simple neural network

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

 

Simple neural network

Segment customers into groups by distinct charateristics (eg, age group)— for instance, to better assign marketing campaigns or prevent churn

K-means clustering

 

Segment customers to better assign marketing campaigns using less-distinct customer characteristics (eg, product preferences)

Gaussian mixture model

 

Segment employees based on likelihood of attrition

Gaussian mixture model

 

Cluster loyalty-card customers into progressively more microsegmented groups

Hierarchical clustering

 

Inform product usage/development by grouping customers mentioning keywords in social-media data

Hierarchical clustering

 

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

Recommender system

 

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

 

Recommender system

 

Optimize the trading strategy for an options-trading portfolio

Reinforcement learning

 

Balance the load of electricity grids in varying demand cycles

Reinforcement learning

 

Stock and pick inventory using robots

Reinforcement learning

 

Optimize the driving behavior of self-driving cars

Reinforcement learning

 

Optimize pricing in real time for an online auction of a product with limited supply

Reinforcement learning

 

Diagnose health diseases from medical scans

Convolutional neural network

 

Detect a company logo in social media to better understand joint marketing opportunities (eg, pairing of brands in one product)

Convolutional neural network

 

Understand customer brand perception and usage through images

Convolutional neural network

 

Detect defective products on a production line through images

Convolutional neural network

 

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

Generate analyst reports for securities traders

Recurrent neural network

 

Provide language translation

Recurrent neural network

 

Track visual changes to an area after a disaster to assess potential damage claims (in conjunction with CNNs)

Recurrent neural network

 

Assess the likelihood that a credit-card transaction is fraudulent

Recurrent neural network

 

Generate captions for images

Recurrent neural network

 

Power chatbots that can address more nuanced customer needs and inquiries

Recurrent neural network

 

 

Adapted from mckinsey.com – An executive’s guide to AI

 Image source adapted from Oracle

 

 

Views: 1746

Comment

You need to be a member of Data Science Central to add comments!

Join Data Science Central

Videos

  • Add Videos
  • View All

© 2020   TechTarget, Inc.   Powered by

Badges  |  Report an Issue  |  Privacy Policy  |  Terms of Service

console.log("HostName");