Python DeepLearning Project - Adavellisahaja/Pyhton-DeepLearning-Project GitHub Wiki

Python-Deep Learning Project Increment-2

Team Members:

Sahaja Adavelli(01)

Zeal Patel(18)

Sree Valli Tata(23)

Objective: The main objective of increment 2 is to clean the data and find the polarity values. If the polarity value is less than 0 then it is negative, if the value is greater than 0 then it is positive. Then train the model using logistic regression.

Implementation:

Using the twitter API we have collected 20000 tweets using the keywords related to stocks and convert to .csv file.

Import the required libraries and read the data using pandas.

Drop all the columns except tweets as they are not required and change the name of the column to tweet_text

Data Cleaning: We are cleaning the data by removing all other characters except alphabets, consider all small and capital alphabets.

Textblob: Using textblob we are analyzing the sentiment of by finding polarity values. If polarity value is less than 0 then it is negative if the polarity value is greater than zero then it is positive.

Using the WordCloud we are representing the negative and positive words

Split the data into train and test data using train_test_split

TfidVectorization: As it is difficult to train using words so we are Vectorize the words to get Unicode for each word and train the words using unicodes.

Train the model using Logistic regression

Find the words which contribute more for the analysis

Contribution:

Zeal: Working on Data collection from different sources and collect more tweets for product based companies

Sree Valli: Contributed for data cleaning and find the sentiment using polarity values.

Sahaja: Did the model training using machine learning logistic regression .