|
|
--- |
|
|
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 |
