In August 2015, the IAPA ACT Chapter had the opportunity to hear from Yongzheng (Tiger) Zhang, Manager, Business Analytics at LinkedIn.
Yongzheng provided an outline on two new LinkedIn platforms:
- Modoop - a machine learning platform for propensity modeling; and
- Voices - a text analytics platform for understanding the voice of the member
The Modoop platform allows internal users to easily build prediction models using customizable signals such as sign ups, job views, ads clicks that are extracted from big data on Linkedin's Hadoop infrastructure. It takes only a few clicks and minutes to build, deploy and monitor such a prediction model.
The voices platform derives insights and creates value for Linkedin members from massive amount of member feedback in text within Linkedin's ecosystem. To achieve this, it uses natural language processing and machine learning techniques to build text mining functions such as topic modeling, text categorization, sentiment analysis, to name a few.
With these functions, it can derive insights and detect emerging trends from a variety of unstructured data sources so different organizations can listen to the members and take quick actions to improve products and user experiences.