IdeaCredIdeaCred

galafis/Recommendation-System-ML

53

Content-based movie recommendation using TF-IDF and cosine similarity with scikit-learn

What's novel

Content-based movie recommendation using TF-IDF and cosine similarity with scikit-learn

Code Analysis

2 files read · 2 rounds

A content-based movie recommendation system that uses TF-IDF vectorization on movie metadata (genres and plot keywords) to compute cosine similarity and return top-K similar titles.

Strengths

Clean class-based architecture with clear separation of concerns (load, preprocess, train, recommend). The code is well-documented in the README, supports Docker, and handles basic edge cases like missing files or empty datasets gracefully.

Weaknesses

Lacks unit tests entirely. Error handling is minimal (e.g., no case normalization for text features, potential issues with non-English keywords due to hardcoded English stop words). The main block logic overwrites data aggressively if the CSV is missing columns.

Score Breakdown

Innovation
3 (25%)
Craft
54 (35%)
Traction
6 (15%)
Scope
51 (25%)

Signal breakdown

Innovation

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

Craft

Ci-1
Tests-2
Polish+0
Releases+0
Has License+5
Code Quality+19
Readme Quality+15
Recent Activity+7
Structure Quality+4
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+3
Languages+5
Subsystems+0
Bloat Penalty+0
Completeness+7
Contributors+5
Authored Files+4
Readme Code Match+3
Architecture Depth+3
Implementation Depth+8

Evidence

Commits

7

Contributors

1

Files

9

Active weeks

4

TestsCI/CDREADMELicenseContributing

Repository

Language

Python

Stars

1

Forks

0

License

MIT