Luxxify: Ulta Makeup Reviews<\/a><\/h1>\nHey everyone,<\/p>\n
I recently released an open source dataset containing Ulta makeup products and its corresponding reviews!<\/p>\n
Custom Created Kaggle Dataset via Webscraping: Luxxify: Ulta Makeup Reviews<\/a><\/p>\nFeel free to use the dataset I created for your own projects!<\/p>\n
Webscraping Process<\/h1>\n
Web Scraping:<\/strong> Product and review data are scraped from Ulta<\/a>, which is a popular e-commerce site for cosmetics. This raw data serves as the foundation for a robust recommendation engine, with a custom scraper built using requests, Selenium, and BeautifulSoup4. Selenium was used to perform button click and scroll interactions on the Ulta site to dynamically load data. I then used requests to access specific URLs from XHR GET requests. Finally, I used BeautifulSoup4 for scraping static text data. Leveraging PostgreSQL UDFs For Feature Extraction<\/strong>: For data management, I chose PostgreSQL<\/a> so that I could clean the scraped data from Ulta. This data was originally stored in a complex JSON which needed to be unrolled in Postgres. <\/p>\nAs an example, I made a recommender model using this dataset which benefited greatly from its richness and diversity.<\/p>\n
To use the Luxxify Makeup Recommender click on this link: https:\/\/luxxify.streamlit.app\/<\/a><\/p>\nI’d greatly appreciate any suggestions and feedback \ud83d\ude42<\/p>\n
Link to GitHub Repo<\/a><\/p>\n<\/div>\n submitted by \/u\/pansali <\/a>
[link]<\/a><\/span> [comments]<\/a><\/span><\/p>
Feel free to use the dataset I created for your own projects!<\/p>\n
Webscraping Process<\/h1>\n
Web Scraping:<\/strong> Product and review data are scraped from Ulta<\/a>, which is a popular e-commerce site for cosmetics. This raw data serves as the foundation for a robust recommendation engine, with a custom scraper built using requests, Selenium, and BeautifulSoup4. Selenium was used to perform button click and scroll interactions on the Ulta site to dynamically load data. I then used requests to access specific URLs from XHR GET requests. Finally, I used BeautifulSoup4 for scraping static text data. Leveraging PostgreSQL UDFs For Feature Extraction<\/strong>: For data management, I chose PostgreSQL<\/a> so that I could clean the scraped data from Ulta. This data was originally stored in a complex JSON which needed to be unrolled in Postgres. <\/p>\n As an example, I made a recommender model using this dataset which benefited greatly from its richness and diversity.<\/p>\n To use the Luxxify Makeup Recommender click on this link: https:\/\/luxxify.streamlit.app\/<\/a><\/p>\n I’d greatly appreciate any suggestions and feedback \ud83d\ude42<\/p>\n Link to GitHub Repo<\/a><\/p>\n<\/div>\n submitted by \/u\/pansali <\/a>
[link]<\/a><\/span> [comments]<\/a><\/span><\/p>
