Logo Lanfrica

Iammteo/Sentiment-analysis-model-for-Nigerian-pidgin-Low-resource-language-

Domaine:

natural language processing

Type de record:

model
Créateur:
Iam
Hôte:
This is a machine learning project of a Nigerian Pidgin sentiment analysis model specifically designed for business reviews ## 📝 Sentiment Analysis on Pidgin-English Reviews This project performs **Sentiment Analysis** on Pidgin-English customer reviews using machine learning. It processes raw text reviews, extracts features using TF-IDF, and trains an SVM classifier to classify sentiments. ### 🚀 Features - Supports **Pidgin-English** reviews (low-resource language) - End-to-end ML pipeline with: - Text cleaning - TF-IDF vectorization - SVM classification - Hyperparameter tuning - Evaluation metrics: Accuracy & Classification Report - Runs smoothly in **Google Colab** ### 📁 Dataset You can download the dataset directly using this link: 🔗 Pidgin Reviews Dataset (.xlsx) --- ### 🧰 Requirements You can run this in **Google Colab** without any setup. For local use, install the dependencies: ```bash pip install pandas scikit-learn tabulate openpyxl ``` --- ### 🧪 How to Use #### In Google Colab: 1. **Import libraries and load the dataset:** ```python import pandas as pd import re from sklearn.model_selection import train_test_split, GridSearchCV, RandomizedSearchCV from sklearn.feature_extraction.text import TfidfVectorizer from sklearn.svm import SVC from sklearn.metrics import accuracy_score, classification_report from sklearn.preprocessing import LabelEncoder from tabulate import tabulate from scipy.stats import uniform # Load dataset from URL url = "docs.google.com" df = pd.read_excel(url) print("Dataset loaded successfully!") ``` 2. **Preprocess, train, and evaluate** the model using SVM and `GridSearchCV` or `RandomizedSearchCV`. ### 📊 Output The script will output: - Confirmation of dataset load - Accuracy score - Classification report (precision, recall, F1)