Emotion Analysis + Spotify Track Hashtags

A Python project that analyzes text to detect emotions, cleans the data, and recommends Spotify tracks related to the dominant emotion, complete with hashtags generated from audio features and artist genres.

Features

• Clean Chat Data: Remove noise, unnecessary characters, and prepare text for analysis.
• Text Emotion Detection: Uses j-hartmann/emotion-english-distilroberta-base for classifying emotions in text.
• Dominant Emotion Extraction: Finds the most frequent emotion in a dataset.
• Spotify Track Search: Fetches tracks related to the detected emotion.
• Hashtag Generation: Creates hashtags from artist genres and audio features like #happy, #energetic, #danceable.

Requirements

• Python 3.13.7
• Spotify Developer Account
• .env file containing:

TASK=text-classification
MODEL_NAME=j-hartmann/emotion-english-distilroberta-base

SPOTIPY_CLIENT_ID=your_spotify_client_id
SPOTIPY_CLIENT_SECRET=your_spotify_client_secret

• Python packages:

pip install -r requirements.txt

Project Structure

SENTIMENTANALYSIS/
│
├─ .env
├─ requirements.txt
├─ .gitignore
│
├─ src/
│   ├─ Data/
│   │   ├─ chat.txt
│   │   └─ typesEmotion.txt
│   │
│   ├─ Scripts/
│       ├─ DataExtraction/
│       │   ├─ chat.ipynb          # Cleaning & preparing chat data
│       │   ├─ cleaned_chat.csv
│       │   └─ emotions.csv
│       │
│       ├─ EmotionAnalysis/
│       │   └─ analysis.py         # Fetch dominant emotion
│       │
│       ├─ Model/
│       │   └─ model.py
│       │
│       └─ Spotify/
│           ├─ __pycache__/
│           ├─ emotion_to_genre.py
│           ├─ getEmotion.py
│           └─ getSongs.py

Workflow

1. Clean Chat Data

• Load raw chat data (chat.txt)
• Remove unnecessary characters, empty lines, or irrelevant text
• Save cleaned data as cleaned_chat.csv

2. Fetch Emotion

• Use the cleaned CSV to predict emotions for each line
• Save results in emotions.csv
• Extract the dominant emotion for the dataset

3. Get Spotify Tracks

• Use the dominant emotion to search Spotify tracks
• Fetch artist genres and audio features (valence, energy, danceability)
• Generate hashtags for each track
• Print track names with hashtags

Example Usage

# Step 1: Clean Data
python src/Scripts/DataExtraction/chat.ipynb

# Step 2: Fetch Emotion
python src/Scripts/EmotionAnalysis/analysis.py

# Step 3: Get Songs
python src/Scripts/Spotify/getSongs.py

Example Output:

Dominant Emotion: neutral

Top Spotify tracks with hashtags:
Blinding Lights - The Weeknd
Hashtags: #pop #rnb #energetic #danceable
Happy - Pharrell Williams
Hashtags: #pop #funk #hiphop #happy #energetic #danceable