Multiple Imputation with LightGBM in Python

Three in one. capture audio with mobile mic and then play, play songs from internal download directly, displaying spectogram (frequency) of an audio in Flutter.

End-to-end Predictive Analytics ML Project

Assignment for Data & information quality course attended @ Polimi. Analyzed performances of different imputation techniques.

Different imputation technique with example

R project for comparing different Missing Value Imputation (MVI)* approaches across three datasets.

Two Movie recommenders deployed on Flask app. The recommender are build on NMF and Cosine Similarity model, trained on MovieLens 100k, and automatically updated every 12 hour!

Additional imputation function for the `R` package `mice`, according to the Heckman model.

A webapp for recommending movies based on two models: collaborative filtering with non-negative matrix factorisation and k-nearest neighbours algorithm.

Multiple imputation with chained equation implemented from scratch. This is a low performance implementation meant for pedagogical purposes only.

the multivariate analysis compares different rows and columns for beat accuracy eg:knn imputer in univariate analysis it only compares with the same columns eg mean or median for numbers