Skip to content

Latest commit

 

History

History
11 lines (6 loc) · 742 Bytes

README.md

File metadata and controls

11 lines (6 loc) · 742 Bytes

Under-Armour-Review-Mining

  • Conducted Python web-crawler; scraped google reviews of Under Armour and Nike offline stores.[Google Crawler]

  • Built keyword extraction tool using TF-IDF, Topic Model (LDA, LSA) and TextRank to identify customer concerns and find operational insights.[Keyword Extraction]

  • Develop topic and sentiment classifiers using word embedding and machine learning models (Logistic Regression, Random Forest, Boosting) to assess customer perception to sports brand.[Topic Sentiment Classifier]

  • Produce data visualization report using Tableau to interrupt data and convey results to all stakeholders.[Ongoing]

image