Azure Machine Learning Studio
Analysis of the loan returns probability using Azure ML
In this project, Azure machine learning platform was used to analyze the probabilities of the client to return a loan based on his personal information and specific parameters of the loan.
First, the dataset with information on the credit status was statistically and visually analyzed in order to select an optimal evaluation strategy. The following procedures were performed:
- a number of steps for data cleaning was evaluated,
- detection of the erroneous fields and typos along with recovery of the missing parameters based on the available distribution.
On the next stage, after appropriate parameters representation and application of the outlier detection techniques based on one-class SVM model non-linear classification model were trained.
Finally, machine learning experiment was converted into a web service, and prediction accuracy was estimated on different datasets with similar distribution parameters.
Virtual try-on tool for makeup products
The system consists of a face detection and segmentation model and an algorithm that allows recoloring objects without losing their original texture.
Online sign language interpreter
AI algorithm that converts video of a person using sign language into a text transcript
Workout helper app
Mobile app for the estimation of proper body positions during the workout.