Sentiment Analysis

• 

  polarity and subjectivity table

• 

  graphical display of sentiments

Inspiration

This inspiration comes from MLH Global Hack Week challenge.

What it does

• First, it imports amazon product review datasets.
• It then does basic exploratory data analysis (EDA)
• It then carries out quite an extensive data cleaning.
• It imports spacy library and it's text blob component.
• This is followed by calculation of polarities and subjectivities of the reviews.
• Subsequently, it groups the reviews into positive, negative and neutral sentiments.
• Lastly, it saves and visualizes the updated dataset showing the percentage of positive, negative and neutral sentiments.

How we built it

Imports the necessary python libraries, does data processing, especially data cleaning, then imports spacy text blob library for sentient analysis. It also carried out other auxiliary works such as it's n-gram display.

Challenges we ran into

Detailed Data cleaning and, in particular the regular expression part, was quite challenging.

Accomplishments that we're proud of

Sentiment Analysis and python lambda skills.

What we learned

• Python Lambda
• Sentiment Analysis

What's next for Sentiment Analysis

Automate and build a web interface for it.

Built With

• natural-language-processing
• nltk
• python
• spacy