Giter Site home page Giter Site logo

haohe1113 / beer-rmcd-system Goto Github PK

View Code? Open in Web Editor NEW
0.0 1.0 0.0 2.67 MB

In this project, we scraped beer reviews from Beeradvocate.com and used them to build a beer recommendation system with various text mining concepts.

Jupyter Notebook 100.00%
top-rates-beers recommended-beers sentiment-scores reviews

beer-rmcd-system's Introduction

We Got Your Beers!

Crowdsourced Beer Recommendation System

intro

Quick Links

Introduction

Beeradvocate.com is an online beer rating webstie. Alongside the ratings, people also post their reviews about the experience with various beers. In this project, we scraped beer reviews from Beeradvocate.com and used them to build a beer recommendation system with various text mining concepts. The recommendation system was required to accept user inputs about desired attributes of a product and come up with 3 recommendations.

Pakages Used

  • Selenium
  • NLTK
  • spaCy (Cosine Similarity)
  • VADER (Word Level Sentiment Analysis)

Approach

  1. Scraped Beeradvocate.com for ~6k reviews on thousands of beer products.
  2. Identified 3 beer attributes assuming that a user of this recommendation system would input in order to search for desired beers.
  3. Performed a similarity analysis between the 3 attributes and the reviews using spaCy; Extracted 300 reviews that have the highest similarity scores.
  4. Performed sentiment analysis using VADER on these 300 reviews and sort them by the sentiment scores.
  5. Recommended 3 beers to the user based on a general ranking system which combines similarity scores and attribute sentiment scores.

You could find our code here

Analysis & Insights

1. Attributes Chosen
We picked the top 3 mostly mentioned beer attributes over all reviews scraped:

Sweet: Malty, grainy, caramel-like;
Fruity: Flavors reminiscent of various fruits;
Robust: Rich and full-bodied;

2. Top Beers With Highest Similarity Scores

Alt Text

3. Top Beers With Highest Sentiment Scores

Alt Text

4. Top Rated Beers From Dataset

Alt Text

5. Insights

Our Recommendation:

  1. Mornin' Delight
  2. Black Gold
  3. Marshmallow Handjee

By reading through the user reviews for the three products, we found that people love the maple syrup taste and the long-lasting aftertaste in Mornin' Delight which backs up the sweet preference of our user. The same thing goes to Black Gold, which is highly regarded for its chocolate malt, cake, vanilla and roasty flavor. Marshmallow Handjee gets high scores from users who love fruity flavor and the hit of citrus and raspberry. All of our three recommendations have a rich mix of fruity and honey flavors which fit perfectly with the user preference of sweet, robust and fruity drinks.

Other than that, the three recommended beers also have good reputations for the thick and silky monthfeel. Mornin Delight has been rated highly for its pillowy carbonation and encroaching sticky texture while Black Gold was frequently praised by the lingering and chewy taste. Marshmallow Handjee however, was commented by multiple reviewers that it has a sharp carbonation which cuts through its massive body.

Top Rates Beers From Dataset:

  1. Pliny The Younger
  2. Marshmallow Handjee
  3. Barrel Aged Imperial German Chocolate Cupcake Stout

We could see there is an overlap between the top rated beers and the recommended beers by our evaluation system: Marshmallow Handjee. This demonstrates the massive acception for Marshmallow Handjee from sweet and fruity beer lovers. However, for the rest two top rated beers, we didn't find a good match with the three desired attributes. Pliny The Younger is a triple IPA well known by its hoppy aroma and balanced taste with crisp and bitterness. As for Barrel Aged Imperial German Chocolate Cupcake Stout, most of its comments regarded its light, spicy, warming barrel appearance in taste as the shining point while barely mentioned sweet, robust or fruity.

We concluded from there that while looking for beers by user ratings could provide us nice beers with high general quality and unique tastes, this methodology lacks variety and reliable matching ability with features we want.

beer-rmcd-system's People

Contributors

haohe1113 avatar

Watchers

 avatar

beer-rmcd-system's Issues

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.