---
language:
- en
license: mit
tags:
- task-classification
- transformers
---
# Game Issue Review Detection

**This model is a fine-tuned version of RoBERTa on the Game Issue Review dataset**.

## What is Game Issue Review?

**Game Issue Review** refers to player feedback that highlights significant problems affecting the gaming experience.

## Model Capabilities

This model can detect:
- ✅ Technical issues (e.g., "Game crashes on startup")
- ✅ Design complaints (e.g., "This boss fight is poorly designed")
- ✅ Monetization criticism (e.g., "The pay-to-win mechanics ruin the game")
- ✅ Other significant gameplay problems

## Quick Start

```python
from transformers import pipeline
import torch

# Load the model
classifier = pipeline("text-classification", 
                     model="FutureMa/game-issue-review-detection",
                     device=0 if torch.cuda.is_available() else -1)

# Define review examples
reviews = [
    "Great game ruined by the worst final boss in history. Such a slog that has to be cheesed to win.",
    "Great game, epic story, best gameplay and banger music. Overall very good jrpg games for me also i hope gallica is real"
]

# Label explanations
LABEL_MAP = {
    "LABEL_0": "Non Game Issue Review",
    "LABEL_1": "Game Issue Review"
}

# Classify and display results
print("🔍 Game Issue Review Analysis Results:\n")
print("-" * 80)
for i, review in enumerate(reviews, 1):
    pred = classifier(review)
    label_explanation = LABEL_MAP[pred[0]['label']]
    print(f"Review {i}:")
    print(f"Text: {review}")
    print(f"Classification: {label_explanation}")
    print(f"Confidence: {pred[0]['score']:.4f}")
    print("-" * 80)
```

## Supported Languages
🌐 English

The model is particularly useful for:
- Game developers monitoring player feedback
- Community managers identifying trending issues
- QA teams prioritizing bug fixes
- Researchers analyzing game review patterns