SUPR
Influence of Encoding Schemes on Machine Learning for Human Behavioral Data
Dnr:

NAISS 2025/22-970

Type:

NAISS Small Compute

Principal Investigator:

Sameen Mansha

Affiliation:

Kungliga Tekniska högskolan

Start Date:

2025-09-01

End Date:

2026-09-01

Primary Classification:

10201: Computer Sciences

Webpage:

Allocation

Abstract

Behavioral sciences process large amounts of images, textual, tabular, and graph-based datasets for classification, regression, or clustering-based tasks. Which input data representations and encoding schemes should be preferred over others for increasing accuracy of Machine Learning (ML) based models? Studying how different encoding schemes may be useful in enhancing the accuracy of ML-based methods. The project findings can be deployed in a wide range of real life applications.