SentimentAnalysisDarijaNZ 0.3.1

C'est une bibliothèque Python modulaire qui fournit des outils pour le calcul de similarité, le filtrage de données, et l'analyse de sentiments.

SentimentAnalysisDarijaNZ

SentimentAnalysisDarijaNZ is a modular Python library designed for text processing in Moroccan Darija (Arabic dialect). The library provides tools for similarity calculation, filtering, and sentiment analysis.

Key Features

1. Similarity Calculation

This module includes advanced similarity measures tailored for Moroccan Darija. It provides functions to compute similarity scores between words or phrases while handling linguistic nuances such as:

• Levenshtein Similarity: Accounts for minor variations in spelling.
• Phonetic Similarity: Matches words based on their phonetic resemblance.
• Sequence Similarity: Uses sequence matching to evaluate structural similarity.

Example:

from SentimentAnalysisDarijaNZ import similarity
similarity_score = similarity.levenshtein_similarity("makla", "maakla")
print(similarity_score)

2. Filtering

This module identifies and filters text based on specific brands, quality indicators, or prices. It also includes spam detection tailored for Moroccan Darija.

Spam Filtering Example:

from SentimentAnalysisDarijaNZ import filtering
positive_spam, neutral_phrases = filtering.spam_analysis(["Check this link www.fake.com", "Free money!"], spam_dict)
print(positive_spam)

Brand Filtering Example:

filtered, not_filtered = filtering.filterbrand([["laptop of good quality"]], "quality")
print(filtered)

3. Sentiment Analysis

Detects positive and negative sentiments in Darija text, accounting for nuances like negations and intensifiers.

Example:

from SentimentAnalysisDarijaNZ import sentiment
positive, negative, neutral = sentiment.sentiment_analysis_darija(["The product is amazing!", "Terrible quality"])
print(positive)

Installation

Install the library via pip:

pip install SentimentAnalysisDarijaNZ

Usage

After installation, import the necessary modules:

from SentimentAnalysisDarijaNZ import similarity, filtering, sentiment

License

This project is licensed under the Apache License 2.0.

Licence

Ce projet est sous licence MIT. Veuillez consulter le fichier LICENSE pour plus d'informations.

Auteur

Nawfal BENHAMDANE

• Elève-Ingénieur à l'Ecole Centrale Casablanca Zaynab RAOUNAK
• Elève-Ingénieur à l'Ecole Centrale Casablanca Hamza Laraisse
• Elève-Ingénieur à l'Ecole Centrale Casablanca