This project aims to classify time series data into categories based on numerical patterns by employing a range of AI models, such as KNN, RNN, Decision Tree, Computer Vision, and Support Vector Learning.
Sponsored by: Volvo
Team Members
Jake Connolly Web Simmara Zak Young Chae Kim Tim Green
Instructor: Steven Shaffer
Project Poster
Click on any image to enlarge.
Project Video
Project Summary
Overview
Our goal of the project was to use AI to classify time series data into categories based on numerical patterns by employing a range of AI models. This list includes KNN, decision tree, ensemble, computer vision, and RNN. The time series data serves as input for these models which then will output predictions that maximize classification accuracy
Objectives
– Get the accuracy of the model to an acceptable level to show that there is a correlation between the time series data.
– Test each model and note the accuracy that it can produce on the data
Approach
– Understand objectives and how to go about solving the problem
– Researched different AI algorithms to figure out how they work
– Test different AI algorithms to find which one results in the best outcome
– Interpret and manipulate the data to increase the accuracy
– Change parameters through testing to increase accuracy
– Repeat manipulations of data and testing to try to achieve an acceptable output
Outcomes
These algorithms that we created will be able to be to categorize the time series data which will be able to be used for various applications. We were after all able to get an acceptable accuracy that does show the possibility of accurately getting an AI to classify objects