sape344/URL-Classification sape344/URL-Classification

  • Stars
    star
    6
  • Rank 2,539,965 (Top 51 %)
  • Language
    Jupyter Notebook
  • Created almost 5 years ago
  • Updated almost 4 years ago
Twitter logo Share LinkedIn logo Share Facebook logo Share
sape344/URL-Classification
github

sape344

Reviews

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

Repository Details

In spite of all the advantages it provides, the internet has become a platform used for online crimes today. These offenses are committed through URLs. My goal in this project is to identify and classify malicious URLs and develop an ML algorithm that can alert users to potential threats in advance. I used URL dataset (ISCX-URL-2016) from Canadian Institute for Cybersecurity. You can reach dataset using this url: https://www.unb.ca/cic/datasets/url-2016.html

More Repositories

1

Log-Gammatone-Spectrogram

Log-Gammatone Spectrogram is used to classified sounds on Deep Learning Models. I wrote a python function to extract Log-Gammatone Spectrogram for a sound.
Python
3
star
2

Parallel-Sparse-Matrix-Vector-Multiplication-Symmetric-seven-banded-Matrix-using-Compress-Sparse-Row

The code is about The Parallel Processing and Symmetric seven-banded matrix sparse matrix multiplication. The Parallel Processing increase performance if used correctly. Sparse matrix keep too memory usage however Sparse matrix process speed can be increased and memory usage can be decreased thanks to compress method. I selected Symmetric seven-banded matrix using Compress Sparse Row Format (CSR) storage scheme.
C++
1
star
3

ENVIRONMENTAL-SOUND-CLASSIFICATION

Environmental sound classification is the preprocessing and the classification of the sounds obtained from the everyday environment. It is a crucial method to interpret the sound signals and use them in various applications. Our dataset includes a total of 2000 samples labeled under 50 specific types. To overcome this challenge, we tried various machine learning and deep learning methods such as CNN and CRNN. I also tried to augment the amount to data to provide more data to our model obtain higher performance from our models.
Jupyter Notebook
1
star