Conversational bots in ruby

Transcript

1. Conversational Bots in ruby Aboobacker MK
3. My background • tachyons in internet • Ruby on rails developer at Red Panthers • Open source enthusiast and voulnteer and Swathanthra Malayalam Computing
4. This talk is not about • Tutorial for building a chat bot in ruby • Live coding a bot
5. Then ?
6. Let's talk about bots Image credit : The next web What We have ?
9. • Order Pizza • Book Tickets • Turnoff light • Make a phone call
10. Lita
11. Packed with lot of features • Can connect to any chat service • Simple installation and setup • Easily extendable with plugins • Data persistence with Redis • Built-in web server and routing • Event system for behavior triggered in response to arbitrary events • Support for outgoing HTTP requests • Group-based authorization • Internationalization • Configurable logging • Generators for creating new plugins
12. Adapters An adapter is a plugin that allows Lita to connect to a particular chat service
13. Handlers A handler is a plugin that adds new functionality to Lita at runtime Consist of route defnisions and call backs
14. Extensions An extension is a plugin that adds new functionality to Lita's core in order to extend the capabilities of adapters and handlers Eg: • Confiramation • Dot env support • Wizards
15. Great !
16. But is that everything we need ?
17. Natural Language Processing
18. Collection of natural language tools in ruby • https://github.com/arbox/nlp-with-ruby • https://github.com/diasks2/ruby-nlp
19. Classification aims to assign a document or piece of text to one or more classes or categories making it easier to manage or sort. require 'stuff-classifier' cls = StuffClassifier::TfIdf.new("Cats or Dogs") cls.ignore_words = [ 'the', 'my', 'i', 'dont' ] cls.train(:dog, "Dogs are awesome, cats too. I love my dog") cls.train(:cat, "Cats are more preferred by software developers. I never could stand cats. I have a dog") cls.train(:dog, "My dog's name is Willy. He likes to play with my wife's cat all day long. I love dogs") cls.train(:cat, "Cats are difficult animals, unlike dogs, really annoying, I hate them all") cls.train(:dog, "So which one should you choose? A dog, definitely.") cls.train(:cat, "The favorite food for cats is bird meat, although mice are good, but birds are a delicacy") cls.train(:dog, "A dog will eat anything, including birds or whatever meat") cls.train(:cat, "My cat's favorite place to purr is on my keyboard") cls.train(:dog, "My dog's favorite place to take a leak is the tree in front of our house") cls.classify("This test is about cats.") #=> :cat
20. Processing Date & Time require 'chronic' Chronic.parse("Ruby conf schedule for this saturday") => 2017-01-28 12:00:00 +0530
21. And lot of other stuffs like • Full text search • Keyword Ranking • Language Detection • Text similarity • Tokenizer
22. Language processing as service • Api.ai (aquired by Google) • Luis By microsoft • Wit.ai by Facebook
23. Luis
24. But what about follow up questions What is his twitter handle?
26. So build one ! • Wrote ruby wrapper for luis (https://github.com/tachyons/luis) • Ruby version of microsoft's bot framework (https://github.com/tachyons/botframework-ruby) • Built a simple prototype using sinatra https://github.com/tachyons/stock-bot-ruby/blob/master/stock.rb
27. But Code looks ugly !
28. DSL is the way to go • Using rack • Event driven ? BotFramework.configure do |connector| connector.app_id = ENV['MICROSOFT_APP_ID'] connector.app_secret = ENV['MICROSOFT_APP_SECRET'] End BotFramework::Bot.on :activity do |activity| # Activity.id , identifier of the activity # activity.timestamp # activity.channel_id # activity.from, sender # activity.conversation # and so on reply(activity,activity.text) end
29. https://github.com/tachyon s/botframework-ruby/blob/ dialog/examples/ruby_conf /bot.rb
30. Demo
31. def speaker_name(name) twitter_account(name) end def twitter_account(name) twitter_started_at(handle) end
32. Revamp • Bots are the new apps • Lita provides framework for building • Natural Language processing and tools available in ruby • Natural Language processing as service • Demo
33. Questions ?