Autoplay
Autocomplete
Previous Lesson
Complete and Continue
Improving Dialogflow ES accuracy
Basic Terminology
Don't set up your bot to fail (5:02)
Regular vs fallback intents (2:17)
Deflection rate definition (2:48)
Accuracy definition (5:30)
Deflection rate vs accuracy (3:29)
Step by step process to improve Dialogflow ES accuracy
Why chatbots are so brittle (5:35)
Auto Analyze chat logs (6:46)
Understand the shape of your bot (4:46)
Increase quantity and quality of training phrases (6:43)
Quickly debug your bot (9:47)
Fail gracefully (4:47)
Update your bot without breaking it (10:07)
Measure and improve accuracy (8:36)
Fix false negatives (1:01)
Fix false positives (5:22)
Tips
Don't delete the Default Fallback Intent (1:14)
Understand Candidate Intents (3:08)
Avoid the slot filling feature (3:20)
Simulate follow up intents (2:58)
Understand cohesion and separation (4:05)
Context Lifespan (5:31)
Training phrase quality (4:39)
Regression testing (1:49)
NLU (3:36)
Training module (3:45)
Wildcard entities (2:15)
Short utterances (2:08)
Knowledge Base (3:14)
Mega Agents (3:17)
Voice bot accuracy (6:36)
Managing large Dialogflow ES FAQ Bots
Three must know concepts (3:18)
Small talk (3:41)
Conversation Testing (1:22)
Avoid 1-click training (1:46)
False positives and false negatives (5:50)
Test coverage (2:28)
Quantifiable metrics (1:58)
Tools (5:24)
Automated Testing Basics
The two steps you need to do (1:59)
Test coverage (4:26)
Slot filling (2:38)
Context lifespan (6:43)
The problem with using exact training phrases (2:17)
Utilizing true positives and true negatives (1:58)
A neat trick: filler words (6:26)
Automated Testing Implementation using Python
Why pytest and PyCharm (1:50)
Project directory structure (1:15)
CSV Test Script (2:30)
Service account JSON file (1:09)
pytest setup (2:16)
Code walkthrough (5:54)
Need some help? (2:18)
Sample Code
Teach online with
Voice bot accuracy
Lesson content locked
If you're already enrolled,
you'll need to login
.
Enroll in Course to Unlock