Spam ∅

A Spam Detector Extension

Spam ∅ - A Spam Detector Extension

Spam ∅ is a Chrome extension developed by students of Bicol University. It uses a machine learning model, trained with a Naive Bayes Classifier Algorithm, to detect spam comments on YouTube. The extension collects comments from YouTube using the YouTube Data API and classifies them as spam or ham based on predefined criteria. The model was evaluated using various metrics to ensure accuracy.

Download

Extension stats

By: geraldjoyb

Users: 62

-3

Rating: 5.00

(2)

Version: 1.1 (Last updated: 2023-05-13)

Creation date: 2023-04-26

Risk impact: Moderate risk impact

Risk likelihood:

Very low risk likelihood

Manifest version: 3

Permissions:

activeTab
tabs

Size: 5.19M

Email: bl*****@gmail.com

URLs: Privacy policy

Ranking

# 82,472 ▼ 992

Other rankings

#833 in Social Networking #71 in machine learning algorithms keyword 6 #103 in algorithm keyword 2 ▲ 3 #226 in detect keyword 15 ▼ 2 #280 in model keyword 10 ▼ 8

Other platforms

Spam Remover (v1.9)

3.91

(55) 8,400+

Not available on Firefox

Not available on Edge

Want to check extension ranking and stats more quickly for other Chrome extensions? Install Chrome-Stats extension to view Chrome-Stats data as you browse the Chrome Web Store.

Extension summary

Analyze keywords

A Spam Detection Tool for YouTube using Naive Bayes Classifier and Youtube data API.

To summarize, the developers developed a machine learning model using a Naive Bayes Classifier Algorithm to detect spam. This model was integrated into a browser extension that allows the user to collect comments from the YouTube Platform. This browser extension uses the YouTube Data API to collect or scrape comments from the website which will then be fed to the Naive Bayes algorithm for classification. The Naive Bayes Algorithm developed was given training data that was also scraped from the website using the YouTube Data API and was taken from the videos of Popular channels such as Linus Tech Tips and PewDiePie. All these were collected, cleaned, and manually labeled as spam or ham following a set of criteria that determines what makes a spam comment. This dataset was then evaluated using Evaluation Metrics such as Accuracy, Recall, Precision, and F1 Score. The dataset was made to gain a satisfying score from these metrics to get an accurate model.

Developed by Students of Bicol University