Machine Learning is great for:

  • Problems for which existing solutions require a lot of fine-tuning or long lists of rules: one Machine Learning algorithm can often simplify code and perform better than the traditional approach.
  • Complex problems for which using a traditional approach yields no good solution: the best Machine Learning techniques can perhaps find a solution.
  • Fluctuating environments: a Machine Learning system can adapt to new data. Getting insights about complex problems and large amounts of data.

Examples of applications

  • Analyzing images of products in a production line to classify them (CNN)
  • Detecting tumors in brain scans (CNN)
  • News articles classification (NLP, RNN, CNN or Transformers)
  • Flagging unwanted content (NLP)
  • Summarizing long documents (Text Summarization)
  • Chatbot (NLP, NLU)
  • Forecasting revenue based on many performance metrics (Linear Regression, SVM, NN)
  • Segmenting clients based on their purchases (Clustering)
  • Recommending a product based on past purchases (Artificial NN)
  • AI bot (RL)

Types of ML

  • Trained with human supervision (supervised, unsupervised, semisupervised, RL)
  • Whether or not they can learn incrementally on the fly (online vs batch learning)
  • Whether they work by comparing new data points to known data points or by detecting patterns in the training data and building a model (instance-based vs model-based learning)

References

  • Aurélien Géron. “Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow.”