Data Science – List of Common Machine Learning Problems with Examples

• Regression
• Classification
• Clustering
• Association Rules
• Artificial Neural Networks

Examples – Regression Models

• Real Estate – Housing price estimation
• Financial – Stock price estimation
• Insurance – Estimate medical care expenses
• Sales & Marketing – Sales vs Ad spend
• Company growth estimation

Examples – Classification Models

Following are four different algorithms whose examples have been listed below:

• Naive Bayes – Following are some of the examples:
  • Text classifications problems such as Spam filter, topic categorization
  • Diagnose medical conditions
  • Intrusion detection
• K-Nearest Neighbors – Following are some of the examples:
  • Facial recognition (Computer vision)
  • Patterns identification in gene data
  • Cancer disease diagnosis such as Breast cancer
• Classification trees – Following are some of the examples:
  • Credit scoring models
  • Marketing analysis of customer satisfaction
  • Diagnosis of medical conditions
• Support Vector Machine (SVM) – Following are some of the examples:
  • Text categorization (organize document by topics)
  • Fraud analysis, security breaches identification
  • Cancer and other genetic disease identification based on classification of microarray gene expression
  • OCR

Examples – Clustering Models

• Customer Segmentation: Segmenting customers into groups with similar demographics or buying patterns for targeted marketing campaigns
• Anomalous Behavior Detection: Detecting anomalous behavior, such as unauthorized intrusions into computer networks, by identifying patterns of usage
• Simplifying extremely large datasets by grouping a large number of features with similar values

Examples – Association Rule Models

• Analysis of frequently occurring patterns: Used in various ecommerce sites. One such example is to identify items that are purchased together.
• Patterns of purchases or medical claims in relation with fraudulent credit card or insurance use

Examples – Artificial Neural Networks (ANN) Models

• Speech & handwriting recognition
• Automated devices such as self-driven cars, self-driven drones
• Weather and climate patterns
• Social & economic phenomena
• Tensile strength, fluid dynamics etc.