Climate-Sentiment-AraBERT: AraBERT for Climate Change Sentiment Analysis

Overview

Climate-Sentiment-AraBERT is a specialized Arabic PLM designed for analyzing social media content and classifying climate-related posts into the climate-related sentiment classes (positive, negative, or neutral). It works on multiple dialects (Egyptian, Gulf, Laventine, Maghrebi, Mesopotamian, etc..) and MSA.

This model can be used for additional fine-tuning and also for testing.

Model Details:

• Base Model: aubmindlab/bert-base-arabertv02-twitter
• Language: Arabic
• License: Apache License 2.0

Model Inference

You can use Climate-Sentiment-AraBERT directly on any dataset to classify sentiments in a climate-related data. To use it, follow the following steps:

1. Install the required libraries Ensure that you have installed the libraries before using the model using pip:

pip install arabert transformers torch

2. Load the Model and Tokenizer

# Import required Modules
from transformers import AutoModelForSequenceClassification, AutoTokenizer
import torch

# Load model and Tokenizer
model_name = 'hugsanaa/Climate-Sentiment-AraBERT'
model = AutoModelForSequenceClassification.from_pretrained(model_name, return_dict=False, num_labels=3)
tokenizer = AutoTokenizer.from_pretrained(model_name)

3. Predict

# Example of neutral sentiment
text = "سيول_جدة امطار جده كسرت رقم عام 2009م، وتصنف من الظواهر الطبيعية الشديدة"

# Tokenize input
inputs = tokenizer(text, return_tensors="pt", truncation = True, padding = True)

# Make Predictions
with torch.no_grad():
  logits = model(**inputs)[0]  # first element of the tuple
  predicted_class = torch.argmax(logits, dim=1).item()

# Interpret results
labels = ["Negative", "Neutral", "Positive"]
print(f"Prediction: {labels[predicted_class]}")

Inference using pipeline

import pandas as pd
from transformers import pipeline
import more_itertools
from tqdm import tqdm_notebook as tqdm

model = 'hugsanaa/Climate-Sentiment-AraBERT'

# load the dataset (the data must include text column)
data = pd.read_csv(your_climate_sentiment_data)

# Define maximum sequence length
max_len = 128

# generate prediction pipeline
pipe = pipeline("sentiment-analysis", model=model, device=0, return_all_scores =True, max_length=max_len, truncation=True)
preds = []
for s in tqdm(more_itertools.chunked(list(data['text']), 32)): # batching for faster inference
    preds.extend(pipe(s))

# Generate final predictions
data[f'preds'] = preds
final_pred = []
for prediction in data['preds']:
  final_pred.append(max(prediction, key=lambda x: x['score'])['label'])

data[f'Final Prediction'] = final_pred

Results

Below are the results obtained from testing Climate-Sentiment-AraBERT on testing samples from climate-related data

Class	Precision	Recall	F1-Score	Support
Negative	0.7721	0.7740	0.7730	1221
Neutral	0.7267	0.6966	0.7114	1546
Positivr	0.7489	0.8041	0.7755	827
Overall / Avg.	0.7492	0.7582	0.7533	3594