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

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

Downloads last month: 1

Safetensors

Model size

0.1B params

Tensor type

F32