About QuaQua : QuaQua is a digital experience platform that integrates 360-degree virtual reality content for the global travel and tourism industry.
Interviewing for Role : Consultant:Data Science & Analytics
Key Skills Required : Python, Regression and Classification algorithms, Unsupervised algorithms, OOPs
Round 1 | Data Science Assignment
The programming task was to fetch hotel reviews and ratings data using Google API for any location and use that data for predicting hotel rating based on review.
(1) Fetched data using Python and Google Cloud APIs to CSV file
(2) Performed EDA and data cleaning & presence of class imbalance
(3) Set baseline performance using CountVectorizer and decision tree
(4) Used Word2Vec (pre-trained) and Random Forest as final algorithm with ROC-AUC as evaluation metric
Round 2 | Telephonic Interview
This round focussed on case studies mostly:
(1) How to identify which customers are going to convert given data? What assumptions needed to be made?
(2) How to stitch videos in an intelligent manner given pictures?
Round 3 | On-site Interview
Again, this was mostly case study based. Was asked a single question: “What are the things/questions one needs to capture the persona of a traveller and how to go about predicting persona given the attributes?”
Final Outcome : SELECT
What I think worked for me
Liked my Word2Vec approach (pretty new at that time) and also approach towards solving case studies