Skip to content

Predicting the Demilitarization Degree Required for Military Items

Notifications You must be signed in to change notification settings

nancyhamdan/demil-degree-prediction

Repository files navigation

Predicting the Demilitarization Degree of Military Items - Multiclass Classification

This project presents findings from exploring the Military Equipment for Local Law Enforcement dataset that is provided from The Defense Logistics Agency (DLA) in the United States. Excess military equipment and devices from the Department of Defense in the US are handled by the DLA which transfers this excess equipment to federal, state, and local law enforcement agencies in the US as part of the LESO program (Law Enforcement Support Office program). The dataset contains 130958 records representing the transfers that happened between 01/01/1980 and 30/09/2021.

All military items that are transferred must be demilitarized to some degree; the degree of demilitarization required is represented by the DEMIL Code attribute in the dataset. Each DEMIL Code has a corresponding DEMIL IC representing the integrity of the DEMIL Code registered for the transfer. This project details the process that was followed in building a high performing deep learning predictor of the DEMIL Code and in deriving key insights about the data.

You can find the dataset that was used here along with its dictionary here

About

Predicting the Demilitarization Degree Required for Military Items

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published