The project involves engineering an algorithm which can analyze a dataset and classify the sentiments of the sentences by training the algorithm with previous dataset and then classifying the data as positive, negative or neutral.
The project takes into account a dataset which has sentences representing different sentiments. The code mentioned hereby cleans the dataset from unwanted characters, and analyzes individual words to decide whether the sentence is a positive, negative or neutral emotion. This solution is then represented in a graphical form which can give the user an idea whether the dataset has primary positive, negative or neutral emotions.
Furthermore graphs representing the taining and validation accuracy are represented as well.
It should be thoroughly noted that all the libraries which are imported in the first block of the code are important for the smooth execution of the tool and should be imported or installed if not present on the users enviornment.