Skip to content

we took nearly 100000 teacher details for training and a small amount of them for testing where we used 1 for least recommended and 5 for most recommended as output for training data.

Notifications You must be signed in to change notification settings

rahul13289/teacher_recommend_gpuKNNclassifier

Repository files navigation

COMPUTER ARCHITECTURE AND ORGANIZATION PROJECT

Implementation of KNN and CUDA(GPU) in Teacher Recommendation System

The effectiveness of education heavily depends on the alignment between student needs and the teaching style of educators. In this paper, we present a teacher recommendation system utilizing the k-Nearest Neighbors (KNN) algorithm, implemented with CUDA (Compute Unified Device Architecture) for accelerated performance on GPU hardware. The recommendation system aims to assist educational institutions in matching teachers to students or courses based on various attributes such as expertise, student feedback, and course requirements. Traditional KNN algorithms, while effective, are computationally expensive when dealing with large datasets, which is a common challenge in modern educational systems. By leveraging the parallel processing capabilities of CUDA on GPUs, our implementation significantly reduces the time complexity of the KNN algorithm, enabling real-time recommendations even with extensive datasets. We evaluate the system on multiple datasets, demonstrating improved accuracy and efficiency. The experimental results reveal that GPU-accelerated KNN provides a scalable solution that outperforms conventional CPU-based implementations, offering a practical approach for large-scale teacher recommendation systems.

CONTRIBUTION: @Rahul @Ranjith @Sai Mahesh

About

we took nearly 100000 teacher details for training and a small amount of them for testing where we used 1 for least recommended and 5 for most recommended as output for training data.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published