This project contains the Jupyter Notebook, Python language and, NLTK. In this project, we are predicting the positive & negative sentiments from the Twitter data.
This project is based on Machine Learning, NLTK and, Python language.
In this project, we have used different types of Machine Learning like Pandas, NumPy, Matplotlib, Seaborn, NLTK. The Pandas library is used for implementing the different types of datasets. The NumPy library is used for different types of mathematical computations. The Matplotlib library is used for different types of plotting of graphs. The Seaborn library is used for different heat maps and to put the ROC curve. The NLTK library is used for Natural Language Processing and to extract the different text data from the twitter-like datasets.
At first, we extract the dataset of Twitter from the Kaggle site. Then we implement the different types of libraries of Python. Now, we preprocess the data and start the feature engineering and the feature selection. Then we implement different types of Machine Learning algorithms like Random Forest Classifier, Decision Tree Classifier, SVM, K-NN and, Logistic regression.
Now, we compare different ML algorithms, for that we just implement the Confusion Matrix, Pie Chart, ROC Curve, and then we also compare different accuracy of ML Algorithms.
NOTE : ( ALL OF IT IS DONE USING THE PYTHON LANGUAGE )
Submitted by Bhupendra Singh Rathore (BHUMAN1312)
Download packets of source code on Coders Packet