ODSC 2017

Here's the video of ODSC 2017 tutorial.

Thank you Kevin and the rest of the ODSC staff for releasing videos and making data science open!!

Quick Start

$ pip install nlpia
$ git clone https://github.com/totalgood/prosocial-chatbot.git
$ cd prosocial-chatbot/notebooks
$ jupyter notebook "04 Annoying Data.ipynb"

Slides

• [Training a Prosocial Chatbot](https://docs.google.com/presentation/d/1IND6PXOxgYb2IVmXnaSoNBcIOaiUVU-iZ6vnYz6J6-E/edit?usp=sharing
• Exploring Hyperspace

Ubuntu Dialog Corpus

• Ubuntu IRC channel (for developers and contributors)
• ~1.5 M Dialogs (interactions)
• ~10 M Utterances (statements)

Application

References

• Aira, A Visual Interpreter for the Blind
• WildML
• 2018, Lane, Howard and Hapke: Natural Language Processing in Action
• 2016, Lowe, et al: "The Ubuntu Dialogue Corpus: A Large Dataset for Research in Unstructured Multi-Turn Dialogue Systems"
• training set generator v1
• Lowe, training set generator v2
• Shmalko, [lexica and slang normalizers](https://github.com/rasendubi/noisy-text/tree/master/data
• retrieval-based vs rule-based and new modular approach
• retrieval-based bot
• cursing is an indication of literacy
• cursing can give you power and happiness