Sentiment Analysis of Tweets & predicting Stock Prices

Sentiment Analysis of Tweets & predicting Stock Prices

Amartya Ranjan Saikia

Amartya Ranjan Saikia

Guwahati, Assam

Analyzing the fluctuations in the Stock market for #AMZN based on user sentiments on tweets for #Amazon

Artificial Intelligence

  • 0 Collaborators

  • 0 Followers

    Follow

Description

I worked on a project on Sentiment Analysis of tweets of #Amazon and predicting stock prices of Amazon based on user sentiments.Shaping the project in 24 hours was fun and exciting.The challenge was clustering of moods and intaking real-life reactions of people on events/environmental disasters/festivals etc.Also, accounting other factors like CEO activities, company brand tarnishing etc were other important challenging factors to be implemented.I started by scraping the tweets about #Amazon and collecting the stock Data from Yahoo Finance.Next, I cleaned up the data and found the tweet sentiments.Later found the correlation between the existing data of stock fluctuations and sentiments.Finally, I built a TensorFlow Deep Neural Network to train the data and on testing the model I got standard accuracy.It was great to learn about stocks and ML models while structuring the project.At the end of the day, I developed my motivation, confidence to complete and of course updated my tech-stack.Link of some initial phase of the repository - https://github.com/SKKSaikia/Stock_Prediction

Links

github repository

Medium 0 1jvrzphihzgyumsp8l6mtk2s ziamr scpfm8gud8dglmc8z3hf4xdlm3wal0z 0ah64ljrmtymljospntwngfkdlymajolsntwjbpuskq04jgh1n9u4a3exqa

Amartya R. created project Sentiment Analysis of Tweets & predicting Stock Prices

Medium 3a66fae0 a4f7 426d b88e 4d56311e1303

Sentiment Analysis of Tweets & predicting Stock Prices

I worked on a project on Sentiment Analysis of tweets of #Amazon and predicting stock prices of Amazon based on user sentiments.Shaping the project in 24 hours was fun and exciting.The challenge was clustering of moods and intaking real-life reactions of people on events/environmental disasters/festivals etc.Also, accounting other factors like CEO activities, company brand tarnishing etc were other important challenging factors to be implemented.I started by scraping the tweets about #Amazon and collecting the stock Data from Yahoo Finance.Next, I cleaned up the data and found the tweet sentiments.Later found the correlation between the existing data of stock fluctuations and sentiments.Finally, I built a TensorFlow Deep Neural Network to train the data and on testing the model I got standard accuracy.It was great to learn about stocks and ML models while structuring the project.At the end of the day, I developed my motivation, confidence to complete and of course updated my tech-stack.Link of some initial phase of the repository - https://github.com/SKKSaikia/Stock_Prediction

No users to show at the moment.

No users to show at the moment.