Sentiment Analysis on Text | Comparative Machine Learning Project
Sentiment Analysis Project — Machine Learning on English Texts
Using AI and Machine Learning algorithms (KNN, K-Means, Random Forest), we analyze over 21,000 English texts to classify emotions with high accuracy.
In this project, I performed sentiment analysis on a large dataset of English texts. The texts were converted into numerical data using TF-IDF, and three different algorithms were applied: KNN, K-Means, and Random Forest. The performances of the algorithms were compared using accuracy, F1-score, and confusion matrix metrics. Random Forest achieved the highest accuracy of 88%. The results were also visualized using PCA and graphs. This is a comprehensive work for those interested in NLP and machine learning.
Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut elit tellus, luctus nec ullamcorper mattis, pulvinar dapibus leo.
Text preprocessing and TF-IDF vectorization
Classification using K-Nearest Neighbors (KNN)
Clustering with K-Means and visualization with PCA
Achieving high accuracy with Random Forest classifier
Evaluating model performances using accuracy, F1-score, and confusion matrix
Detailed analysis of results with graphical visualizations
The Random Forest algorithm stood out with an 88% accuracy rate. This project serves as a comprehensive example for those who want practical applications in NLP and machine learning.