IdeaCredIdeaCred

kyawhtetyang/FakeNewsDetector_BiLSTM

51

BiLSTM-based fake news detection system supporting binary and multi-class classification, with data preprocessing, feature engineering, model training, and Dockerized workflow.

What's novel

BiLSTM-based fake news detection system supporting binary and multi-class classification, with data preprocessing, feature engineering, model training, and Dockerized workflow.

Code Analysis

19 files read · 5 rounds

A binary fake-news classifier using a BiLSTM neural network, served via a Flask REST API with a static HTML frontend for text input and prediction display.

Strengths

Clean separation of concerns across preprocessing, feature engineering, model, evaluation, and pipeline orchestration. The inference artifact caching pattern and CORS handling show practical deployment awareness.

Weaknesses

The model architecture is a textbook BiLSTM with no novel contribution; tests are minimal (only 3 test functions, one of which trains a tiny model on 2 samples) and lack edge-case coverage for the API layer. The Evaluator.evaluate method calls plt.show() which will hang in a headless server environment.

Score Breakdown

Innovation
3 (25%)
Craft
47 (35%)
Traction
5 (15%)
Scope
55 (25%)

Signal breakdown

Innovation

Not Fork+1
Code Novelty+0
Concept Novelty+0

Craft

Ci-1
Tests+5
Polish+0
Releases+0
Has License+5
Code Quality+17
Readme Quality+12
Recent Activity+7
Structure Quality+5
Commit Consistency+2
Has Dependency Mgmt+5

Traction

Forks+0
Stars+6
Hn Points+0
Watchers+0
Early Traction+0
Devto Reactions+0
Community Contribs+0

Scope

Commits+5
Languages+8
Subsystems+5
Bloat Penalty+0
Completeness+7
Contributors+5
Authored Files+12
Readme Code Match+3
Architecture Depth+7
Implementation Depth+8

Evidence

Commits

18

Contributors

1

Files

34

Active weeks

3

TestsCI/CDREADMELicenseContributing

Repository

Language

Python

Stars

1

Forks

0

License

MIT