Inspiration
We were inspired by the discrepancies between the East and West coasts of the US in terms of railroad networks, and we wanted to apply this self organizing map.
What it does
We leveraged the self-organized map algorithm to engineer an optimized network system
We leveraged the self-organized map algorithm to engineer an optimized network system for Uganda. The self-organizing map, proposed for artificial neural networks, creates spatially organized internal representations of input signal features. This capability enables the extraction of semantic relationships in sentences during the self-organization process. The algorithm's spatially ordered responses, optimal matching cell selection, and weight vector adaptation were crucial in our approach. Our work included practical demonstrations, recommendations for applying the self-organizing map, and an illustrative example of hierarchical data clustering. Additionally, fine-tuning through learning vector quantization was addressed, enriching the algorithm's utility. Our project showcases the real-world impact of the self-organized map in crafting optimized solutions for complex systems like transportation networks.
How we built it
We utilized a self organizing map algorithm and modified it to begin with a square map and a topographical component.
Challenges we ran into
Some challenges we ran into included the pruning algorithm which was very inefficient. HTML had many issues. Additionally we deleted everything we wrote on devpost which caused great tension within the group
What we learned
we learned how to use self organizing maps, coding in python and html
What's next for Optimized Transportaiton Network using Self Organizing Maps
Traversiing over bodies of water, because the algorithm does not know right now what is a water body and what is not
Log in or sign up for Devpost to join the conversation.