LinkedIn end-to-end data product to measure customer happiness
Who is this presentation for?Data scientists or analysts
Understanding how customers feel about your level of service and product has caused a revolution in several approaches, including customer satisfaction (CSAT), Net Promoter Score (NPS) surveys, etc. An overarching question that arises is how you can predict and scale customer happiness if customers don’t respond to the survey. The question is of pivotal importance to the LinkedIn global customer operation team in elevating members’ and customers’ experience to discover the full value of LinkedIn. To tackle this issue, the data science team at LinkedIn developed an end-to-end data product.
Kelly Wan, Jason Wang, and Lili Zhou walk you through the interesting problems the LinkedIn data science team faces and the innovative approaches they’ve adopted and milestones they reached. In particular, they focus on the problem of understanding, measuring, and predicting customer happiness toward customer support service and products—a problem that’s received a lot of attention in the last couple years. They describe the data they leverage: structured data, that is, operational efficiency key metrics, and unstructured data such as customer email. You’ll learn how the team synthesizes multidimensional signals and develops a comprehensive way to measure customer experience. This includes feature engineering and feature selection and how to predict customer sentiment through scalable text mining techniques on customer verbatim by leveraging neural net-based methodologies. They detail a few methods—FastText supervised model, attention- and transformer-based architecture, recurrent neural network (RNN) and convolutional neural network (CNN) with fine tuning, etc. You’ll dive into how the team designed and developed an innovative data product to provide actionable insights and strategies, which enlightens and recommends LinkedIn to perfect customer experience and further boost its product engagement and revenues.
- General knowledge about stats and machine learning
What you'll learn
- Understand how data scientists at LinkedIn developed an end-to-end approach to tackle how to measure customer happiness toward its services
- Learn to leverage neural net-based deep learning models for text analysis
- Discover how good customer services impacts product engagement metrics and revenues
Kelly Zhiling Wan
Kelly Wan is a senior data scientist at LinkedIn, Sunnyvale. She’s a technology and data science evangelist. Previously, Kelly worked in investment banking for five years in New York City and has undergone a career transformation into the data science area in Silicon Valley. Kelly obtained her master’s of computer science degree from Columbia University and her bachelor’s degree from Southeast University in China.
Chih-Hui “Jason” Wang is a data scientist on the global customer operations (GCO) data science team at LinkedIn. At LinkedIn, he uses data to advocate the voices of customers and members. Previously, he was a data scientist at LeanTaaS where he helped transform healthcare operations through data science. He holds a master’s degree in statistics from the University of California, Berkeley.
Lili Zhou is a manager of the data science team at LinkedIn. Lili has intensive experience in customer operations, billing and collection, risk management, fraud detection, revenue forecasting, and online gaming. She’s passionate about leveraging large-scale data analytics and modeling to drive insights and business value.
Leave a Comment or Question
Help us make this conference the best it can be for you. Have questions you'd like this speaker to address? Suggestions for issues that deserve extra attention? Feedback that you'd like to share with the speaker and other attendees?
Join the conversation here (requires login)
Premier Diamond Sponsors
Premier Exhibitor Plus
For conference registration information and customer service
For more information on community discounts and trade opportunities with O’Reilly conferences
For information on exhibiting or sponsoring a conference
For media/analyst press inquires