Ridesharing services are a very important part in the broader mobility market. Ridesharing services like Uber and Lyft are now mainstream in the major cities in the US and has challenged the traditional services like taxi and rental cars. It’s no surprise that increasing popularity of these ride-sharing services has increased its demand especially during major events like Halloween in the US. For my project, I have collected twitter data for 3 days during Halloween Period (October 29, 2019 to November 03, 2019) using bounding box of the USA. This project will explore the sentiment of the tweets related to Uber, Lyft and Taxi to find out the customer’s sentiment during Halloween. This exploration will provide a comparative understanding of the customer’s experience of the Uber, Lyft and Taxi during such major events. It can also help to take the decision which service to use during such event. Next, I will develop a user representation from the textual and other information from the tweets. This can help to understand the attributes/features of the ridesharing and Taxi customer. Finally, I will develop a model to classify or predict the social media users by 3 categories- Uber user, Lyft user and Taxi user. The prediction of the user’s category can be helpful to predict the demand of a region/city and hence customer can get better service given necessary actions are taken by the service providers. In addition, it can contribute to social media marketing of the respective service providers.

Ridesharing services are a very important part in the broader mobility market. Ridesharing services like Uber and Lyft are now mainstream in the major cities in the US and has challenged the traditional services like taxi and rental cars. It’s no surprise that increasing popularity of these ride-sharing services has increased its demand especially during major events like Halloween in the US. For my project, I have collected twitter data for 3 days during Halloween Period (October 29, 2019 to November 03, 2019) using bounding box of the USA. This project will explore the sentiment of the tweets related to Uber, Lyft and Taxi to find out the customer’s sentiment during Halloween. This exploration will provide a comparative understanding of the customer’s experience of the Uber, Lyft and Taxi during such major events. It can also help to take the decision which service to use during such event. Next, I will develop a user representation from the textual and other information from the tweets. This can help to understand the attributes/features of the ridesharing and Taxi customer. Finally, I will develop a model to classify or predict the social media users by 3 categories- Uber user, Lyft user and Taxi user. The prediction of the user’s category can be helpful to predict the demand of a region/city and hence customer can get better service given necessary actions are taken by the service providers. In addition, it can contribute to social media marketing of the respective service providers.