My Machine Learning Projects

This repo with some of my machine learning Projects , Enjoy it

Contact : [email protected]

Projects description :

Customer Martial statuts predection

In this project, I tried to develop a machine learning model that tries to detect if a customer is single, married or divorced by understanding their behavior in the market. Go ahead and enjoy the project, the accuracy is not more than 70% because the data is small and unbalanced.

• Data Preparation :

  • Cleaning
  • Clustring
  • Feature engineering

• Models :

  • KNN
  • Naive Bayes
  • Random Forest

Financial sentiment analysis

In this project, I developed two models of natural language analysis. The first one is clustring, where I instruct the machine to differentiate between macroeconomic and microeconomic information.The second model is the financial sentiment understanding of information such as good or bad news in the financial market

• Data Preparation :

  • Cleanning : regex , tokonizer
  • Transform : Word to vec

• Models :

  • Clustering
  • Random forset poor accuracy
  • Neural Network with 72% accuracy