Refer to the sample conversational stories provided above. One can also build an ‘action’ for sending emails from Python.Ĭreating more stories: Use train_online.py file to create more stories. Try using regex features and synonyms for extracting entities.īuild actions for the bot: Read through the Zomato API documentation to extract the features such as the average price for two people and restaurant’s user rating. NLU training: One can use rasa-nlu-trainer to create more training examples for entities and intents. 300 to 700 More than 700 User: in range of 300 to 700 For example:īot: What price range are you looking at? The bot should ask the user to select one of the three price categories. Following is an example for the same:īot: What kind of cuisine would you prefer?Ĭhinese Mexican Italian American South Indian North Indian User: I’ll prefer Italian!Īverage budget for two people: Segment the price range (average budget for two people) into three price categories: lesser than 300, 300 to 700 and more than 700. The bot should list out the following six cuisine categories (Chinese, Mexican, Italian, American, South Indian & North Indian) and the customer can select any one out of that. Your chatbot should provide results for tier-1 and tier-2 cities only, while for tier-3 cities, it should reply back with something like "We do not operate in that area yet".Ĭuisine Preference: Take the cuisine preference from the customer. The bot should be able to identify common synonyms of city names, such as Bangalore/Bengaluru, Mumbai/Bombay etc. Consider 'X ' cities as tier-1 and 'Y' as tier-2. Under the section 'current classification' on this page, the table categorizes cities as X, Y and Z. You can use the current HRA classification of the cities from here. For example:īot: In which city are you looking for restaurants?Īssume that Foodie works only in Tier-1 and Tier-2 cities. The bot takes the following inputs from the user:Ĭity: Take the input from the customer as a text field. The project brief provided to you is as follows. The main purpose of the bot is to help users discover restaurants quickly and efficiently and to provide a good restaurant discovery experience. You have been hired as the lead data scientist for creating this product. Problem Statement An Indian startup named 'Foodie' wants to build a conversational bot (chatbot) which can help users discover restaurants across several Indian cities. Integrates with Zomato : API to fetch restaurant information. Assignment : Foodie Restaurant Search Case Study: Restaurant Bot :Ī restaurant chatbot using open source chat framework RASA.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |