You can deploy this model and start earning money today!

AzrilFahmiardi/
sdd-sentiment-kepolisian

safetensors

bert

Views

No views yet

Model Card: Sentiment Kepolisian

Model Overview

Model Name: sdd-sentiment-kepolisian Base Model: indobenchmark/indobert-base-p2 Task: Sentiment classification (4-class: negatif, netral, positif, irrelevant) Language: Indonesian Domain: Police-related news articles

Model Description

Fine-tuned IndoBERT for sentiment analysis of Indonesian police-related news articles. Trained on full article text (title + fulltext, untruncated) rather than a short extracted excerpt — full context proved essential for distinguishing sentiment classes in this domain (see optimization report for the ablation that established this).

Classes:

negatif (Negative)
netral (Neutral)
positif (Positive)
irrelevant (Not relevant to policing sentiment)

Performance Metrics

Metric	Value
Accuracy	0.8200
Macro F1	0.8158
Weighted F1	0.8158
Recall (negatif)	0.8900
Latency (mean)	2.00 ms
Latency (p95)	2.09 ms
Model Size	474.7 MB

Target: Macro F1 >= 0.80 and Recall (negatif) >= 0.85 — both PASS.

Per-Class Performance (test set, n=200/class)

Class	Precision	Recall	F1
negatif	0.8054	0.8900	0.8456
netral	0.7135	0.6100	0.6577
positif	0.9205	0.8100	0.8617
irrelevant	0.8362	0.9700	0.8981

netral is the weakest class across every experiment run for this model, and remains the main area for future improvement (see optimization report).

Usage

Load Model

python

from transformers import AutoTokenizer, AutoModelForSequenceClassification
import torch

model_name = "AzrilFahmiardi/sdd-sentiment-kepolisian"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForSequenceClassification.from_pretrained(model_name)

device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
model = model.to(device)

Inference

python

def analyze_sentiment(title: str, fulltext: str) -> dict:
    text = f"{title} {fulltext}"
    inputs = tokenizer(text, return_tensors="pt", truncation=True, max_length=512).to(device)

    with torch.no_grad():
        outputs = model(**inputs)
        logits = outputs.logits

    probabilities = torch.softmax(logits, dim=-1)[0].cpu()
    predicted_class = logits.argmax(-1).item()
    predicted_label = model.config.id2label[predicted_class]
    confidence = probabilities[predicted_class].item()

    return {
        "sentiment": predicted_label,
        "confidence": confidence
    }

# Example
title = "Polisi Tangkap Pelaku Curanmor di Jakarta Selatan"
fulltext = "Kepolisian Resor Jakarta Selatan berhasil menangkap seorang pelaku pencurian kendaraan bermotor yang telah beraksi di beberapa lokasi..."
result = analyze_sentiment(title, fulltext)
print(f"Sentiment: {result['sentiment']} ({result['confidence']:.2%})")

Output Format

json

{
  "sentiment": "negatif",
  "confidence": 0.8734
}

Input/Output

Parameter	Type	Example
Input	str	title + fulltext berita kepolisian Indonesia (jangan dipangkas manual), max 512 tokens
Output	dict	`{"sentiment": "negatif", "confidence": 0.87}`

Preprocessing Note

This model expects title + full article text, untruncated, as input. Do not manually extract or truncate the article to a short excerpt before tokenizing — that preprocessing was tested and found to significantly degrade performance (Macro F1 dropped from ~0.86 to ~0.78, mainly on the netral class). The tokenizer's own 512-token truncation already handles arbitrarily long articles safely; no crash risk regardless of input length.