Information retrieval scoring
The implementation based on R language of the evaluation algorithm of information retrieval systems
The evaluation algorithm of information retrieval systems was implemented using R-language.
In particular, the method requires partial scores for the two information retrieval systems based on non-intersecting group requests. The requirement was to receive a summary rating of the quality of all the systems with a minimum of computational and experimental costs to perform additional queries.
The algorithm was tested on open data (‘text retrieval conference dataset’). Statistical parameters were calculated comparing the actual ratings of the system with the estimates obtained in the course of the algorithm.
The executed solution as well as its results were comparable to the data published by the author of the method.
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
API driven statistics dashboard
We’ve developed a web-based dashboard showing statistics data. The dashboard is being frequently updated to provide the most actual data on the stores’ inventory condition.