La Kopi @ Developers Space: Data Science

• Automate Boring Stuff Data Science Way
  Florentin Anggraini Purnama, Data Analyst, Traveloka
  
  Have you ever used Slack as a work diary? Do you have lots of private channels where you store your codes and important scripts? Do you have anxiety that one day all your hard-built knowledge base will be lost when you lose access to the workspace?
  
  This session will show you how to solve them. After this talk, you'll know how to create automatic cronjob for backing up any Slack channel you want to BigQuery.
  

• Hey, Customer Support: Using Topic Modeling to Dig into Tweeted Requests
  

  Alexandra Khoo, Cultural Data Scientist, Synthesis
  
  Are you looking for a good way to extract "hidden" patterns from large volumes of short text? Through the lens of uncovering sources of discontent in tweets sent to an ecommerce platform for customer support, we will walk through text processing methods and discuss how best to use topic models.
  
  While the focus is on the use of the Non-Negative Matrix Factorisation model, this talk will briefly touch on two other common methods: Latent Dirichlet Allocation and Biterm topic model.you want to BigQuery.

  
  
• Simkit - Generative & Probabilistic Modelling Framework for Reinforcement Learning

  Jet New, Machine Learning Engineer Intern, Grab
  
  Many pricing and decision making problems at the core of Grab's ride-hailing and deliveries business can be formulated as reinforcement learning problems, with interactions of millions of passengers, drivers and merchants from over 65 cities across the Southeast Asia region.
  
  In Jet's internship as a Machine Learning Engineer, he worked with scientists and engineers to develop Simkit, a new framework for building generative and probabilistic models, in order to train reinforcement learning agents to be served in production services and make optimal, real-time decisions.

Sign up as a speaker for our future La Kopi sessions here.