October 2018
Intermediate to advanced
472 pages
10h 57m
English
If we think about building a chatbot with the TF-IDF approach, we first need to form a data structure that supports training data with labels. Now, let's take an example of a chatbot built to answer questions from users.
In this case, using historical data, we can form a dataset where we have two columns, one of which is the question, and the second of which is the answer to that question, as shown in the following table:
|
Question |
Answer |
|
When does your shop open? |
Our shop timings are 9:00 A.M-9:00 P.M on weekdays and 11:00 A.M-12:00 midnight on weekends. |
|
What is today's special? |
Today, we have a variety of Italian pasta, with special sauce and a lot more other options in the bakery. |
|
What is the ... |
Read now
Unlock full access