La Kopi @ Developers Space: Data Science

Developer Open Mic Session

• Re-Imagining Topic Modeling in NLP: A Break from Conventional Approach
  Shubhanshu Gupta, Machine Learning Engineer, Citibank
  
  When it comes to topic modeling, most of the NLP practitioners use LDA (Latent Dirichlet Allocation) or LSA (Latent Semantic Analysis). While there's nothing wrong with any of the approaches, it's certainly possible to challenge the convention and think outside the box.
  
  This talk presents a new way to approach and think about mining topics from your text data. Shubhanshu will walk you through an example using Named Entity Recognition and Collocations, which yields high quality, meaningful and intuitive topics and visualizations.
  
  
• Forecast Your Products’ Demand with Machine Learning
  Silvia Lara, Data Scientist, LinkedIn
  
  Do you have experience with Machine Learning but do not know how to get started with time-series data? Or maybe you want to scale up your demand forecasting efforts and increase your accuracy? We will explore how to build an effective Machine Learning forecasting model that reflects your business context.
  
  You will learn how to design features, choose the most relevant error metric, validate your model to future-proof your accuracy, and how to decrease your errors through ensembles. Most importantly, you walk out from this talk with a framework to build any forecasting model.
  
  
• Building A Data Pipeline for Conversation Logs
  Dewi Oktaviani, Data Engineer, Kata.ai
  
  We will explore how to ingest millions of conversational log data from various sources, transform the data to a columnar format and move the data to a storage for further analysis.
  
  You will learn how to build a scalable data pipeline, and we'll deep dive into tools such as RabbitMQ, Apache Kafka, and implementing both tools together.
  
  

  ---

  
  Other topics and speakers will be updated closer to the event date. Stay tuned!
  
  Sign up as a speaker for our future La Kopi sessions here.