Cityclic: Predicting User Actions and Matching Procedures
This project, developed for the Cityclic challenge at HackEPS, utilizes a Machine Learning algorithm to predict the next action a user is likely to take. Additionally, the system can identify and match similar procedures to streamline user experience and efficiency.
The project name, Cityclic, refers to the 5 million parameters provided by the enterprise, which are integral to the performance of our model. Additionally, we leverage advanced language preprocessing techniques to create embeddings that help improve the accuracy of predictions and recommendations.
We have built an interactive web portal where users can engage with the algorithm. The frontend is developed using Angular, while the backend is powered by Python with Flask, allowing seamless communication with the ML model. After testing multiple algorithms, we selected the most suitable one for accurate predictions and matching.
Log in or sign up for Devpost to join the conversation.