{"id":25664462,"url":"https://github.com/petridhsg/firewall-data-classification","last_synced_at":"2026-05-01T21:03:03.693Z","repository":{"id":277602771,"uuid":"932945546","full_name":"PetridhsG/Firewall-Data-Classification","owner":"PetridhsG","description":"A single implementation of a machine learning algorithm for a firewall data classification task","archived":false,"fork":false,"pushed_at":"2025-02-14T20:38:56.000Z","size":0,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-02-14T21:31:12.367Z","etag":null,"topics":["machine-learning","matplotlib","numpy","python","seaborn"],"latest_commit_sha":null,"homepage":"","language":"Jupyter Notebook","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/PetridhsG.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2025-02-14T20:16:56.000Z","updated_at":"2025-02-14T20:41:43.000Z","dependencies_parsed_at":"2025-02-14T21:31:16.481Z","dependency_job_id":"e1ccb3e2-aa21-4e77-89d7-7b749b4f6398","html_url":"https://github.com/PetridhsG/Firewall-Data-Classification","commit_stats":null,"previous_names":["petridhsg/firewall-data-classification"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/PetridhsG%2FFirewall-Data-Classification","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/PetridhsG%2FFirewall-Data-Classification/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/PetridhsG%2FFirewall-Data-Classification/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/PetridhsG%2FFirewall-Data-Classification/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/PetridhsG","download_url":"https://codeload.github.com/PetridhsG/Firewall-Data-Classification/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":240427315,"owners_count":19799471,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["machine-learning","matplotlib","numpy","python","seaborn"],"created_at":"2025-02-24T06:18:45.494Z","updated_at":"2025-11-16T21:03:05.772Z","avatar_url":"https://github.com/PetridhsG.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Internet Firewall Data Classification\n\n## Overview\nThis project applies several machine learning algorithms to classify internet firewall data into different action categories. The dataset used for this classification task comes from the [Internet Firewall Data](https://archive.ics.uci.edu/dataset/542/internet+firewall+data) repository.\n\n### Objective\nThe goal of this project is to implement and evaluate commonly used machine learning algorithms on a multi-class classification problem. By analyzing network traffic attributes, we aim to distinguish between different firewall actions, enhancing network security decision-making.\n\n## Machine Learning Algorithms Implemented\nThis project explores and implements the following machine learning techniques:\n\n1. **Principal Component Analysis (PCA)** - Used for dimensionality reduction.\n2. **Least Squares Classification** - A simple linear classification approach.\n3. **Logistic Regression** - A probabilistic model for binary and multi-class classification.\n4. **K-Nearest Neighbors (KNN)** - A distance-based classification method.\n5. **Naïve Bayes** - A probabilistic classifier based on Bayes' theorem.\n6. **Multilayer Perceptron (MLP)** - A feedforward neural network model.\n7. **Support Vector Machines (SVM)** - A powerful classification method using hyperplanes.\n8. **K-Means** - A clustering algorithm to identify patterns in the data.\n\nEach algorithm is tested on the firewall dataset to evaluate its performance in classifying network traffic behavior.\n\n## Dataset Description\nThe dataset consists of 12 features, with the **'Action'** feature representing the target variable. Below is the description of each feature:\n\n| Variable Name          | Description                            |\n|------------------------|----------------------------------------|\n| Source Port           | Sender's initiating port.             |\n| Destination Port      | Receiver's target port.               |\n| NAT Source Port       | Sender's port after NAT.              |\n| NAT Destination Port  | Receiver's port after NAT.            |\n| Bytes                | Packet size in bytes.                 |\n| Bytes Sent           | Bytes sent by the sender.             |\n| Bytes Received       | Bytes received by the receiver.       |\n| Packets              | Total packets transmitted.            |\n| Elapsed Time (sec)   | Duration of communication.            |\n| pkts_sent           | Packets sent by the sender.           |\n| pkts_received       | Packets received by the receiver.     |\n| Action              | Class label (e.g., allow, block, etc.).|\n\n### Classification Task\nThe goal is to classify each network traffic observation into one of the following four classes:\n\n- **allow**\n- **deny**\n- **drop**\n- **reset-both**\n\nEach record belongs to only one of these classes. The classification models are evaluated based on their accuracy and ability to generalize to unseen data.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fpetridhsg%2Ffirewall-data-classification","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fpetridhsg%2Ffirewall-data-classification","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fpetridhsg%2Ffirewall-data-classification/lists"}