Skip to content

kumar-abhishek/sentimentscale

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

56 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Latest personal project highlights:

 Food Recipe query bot (MIT License (MIT) Copyright © 2020 Amyylam)

  • solve a common daily problem with rules-based chatbot by querying a recipe dataframe

prototype code

New add-on! - speech recognition mode app source code

recipe bot speech recognition mode demo

  • update on Jul8: added Cantonese language support! now a Billingual app!

recipe_demo_chickenham_gif

bilingual_recipe_bot_gif

Bilingual version source_code

button-clicking recipe query Dash app

button app source code

 Business Sentiment Analyzer: a streamlit app for making sentiment analyzer predictions

  • Deployed a finetuned text classifier from pretrained BERT embeddings to a frontend web application. User can input a news headline or a CSV of headlines to make inferences.

Simple news sentiment analyzer app made with Streamlit

code file

  • To collect emoji comments as distant supervised sentiment label on news articles for downstream training of news sentiment analysis model
  • Analyze Startup Fundraising Deals from Crunchbase (Dataquest.com Course Mission 167)

Earlier machine learning model/visualization experiments:

Sentiment Analysis of US President Donald Trump’s tweet on Chinese Equities

  • Used stock market reaction as distant supervision of positive/negative sentiment to analyze A collection and manual annotations of >300 tweets from Mr. Trump about ‘China’. Trained machine learning models including Naïve Bayes and Logistic Regression, best test set prediction accuracy at 0.65.

CNN Greed and Fear Index 3-year Backtrack Analysis

  • Explored and visualized how the composite investor sentiment barometer correlates with stock market performance, and what ensued after extreme positive/negative readings.

Dow Jones Index Up/Down Ratio Sentiment Track

  • Visualized most frequent words in retail investors’ tweets when stocks rally/plummet

License

The scripts contained in this repository are under the MIT License (MIT) Copyright © 2020 Amyylam. Please attribute the original source if you adapt the scripts for your use and redistribute the adapted scripts.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Jupyter Notebook 95.4%
  • Python 4.0%
  • Other 0.6%