Note : View Notebook preview if not loading click here
Implementing logistic regression for sentiment analysis on tweets. Given a tweet, we will decide if it has a positive sentiment or a negative one. We will be using a data set of tweets.The twitter_samples contains subsets of 5,000 positive tweets, 5,000 negative tweets.we split dataset in Train test split , 20% will be in the test set, and 80% in the training set. Hopefully we will get more than 99% accuracy.
We can devide this Project in below parts:
- Part 0 : Load and Preprocess Data
- Part 1 : Logistic regression
- Part 2 : Extracting the features
- Part 3 : Training Model
- Part 4 : Test logistic regression
- Part 5 : Error Analysis
- Part 6 : Predict with custom tweet
We implement Logestic Regression From Scretch in this project, and get accurecy 99.55% in the used dataset.