Performing Sentiment Analysis to Prediction of Stock Market using Twitter Database
Budget: $30 – $250 USD
Initially we have to download the tweets related to geolocation from Multinational companies i.e., Microsoft, Tesla.
After getting the tweets we need to perform data pre-processing.
Capture only key words from the tweets
Once the key words are captured, we have to perform sentimental analysis
We have to identify the words related to stock market to see if the stock value is increasing or depreciating
After identifying the set of acronyms related to market value increasing or decreasing. We have to assign weighted value to the terms. Based on this weighted value we need to perform sentimental analysis and classify tweets into positive, neutral or negative.
The accuracy of the weighted values for the words listed should be more than 80 percentage and the prediction of the stocks finally should be higher than 80%.
The output format should be a data visualization chart which should also include comparison chart of the previous model predicting less than 80%.
After getting the tweets we need to perform data pre-processing.
Capture only key words from the tweets
Once the key words are captured, we have to perform sentimental analysis
We have to identify the words related to stock market to see if the stock value is increasing or depreciating
After identifying the set of acronyms related to market value increasing or decreasing. We have to assign weighted value to the terms. Based on this weighted value we need to perform sentimental analysis and classify tweets into positive, neutral or negative.
The accuracy of the weighted values for the words listed should be more than 80 percentage and the prediction of the stocks finally should be higher than 80%.
The output format should be a data visualization chart which should also include comparison chart of the previous model predicting less than 80%.