Summary
1,369
Dialogues
15
Average number of turns per dialogue
268
Hotels
109
Cities
4.58
Average user satisfaction (from 1-5)
19,986
Turns
Data Analysis
User and Machine Actions
Here, we provide some insight into the annotation scheme: each frame is given an id (here, 1 and 2). When values from one frame are used in another frame, we use a ref tag to point to the reference frame.
User | I’d like to book a trip to boston from London on Saturday, August 13, 2016 for 8 adults. I have a tight budget of 1700. | Frame #1 |
Act 1 | inform(intent=book) | |
Act 2 | inform(dst_city = boston, or_city = London, str_date = Saturday\, August 13\, 2016, n_adults = 8, budget = 1700)” | |
Wizard | “Hi…I checked a few options for you, and unfortunately, we do not currently have any trips that meet this criteria. Would you like to book an alternate travel option?” | Frame #1 |
Act 1 | no_result | |
Act 2 | suggest(dst_city) | |
User | “Yes, how about going to Detroit from London on August 13, 2016 for 5 adults. For this trip, my budget would be 1900.” | Frame #2 |
Act 1 | inform(dst_city = Detroit, n_adults = 5, budget = 1900, ref = [1{or_city = London, str = August 13\, 2016}]) | |
Wizard | “I checked the availability for those dates and there were no trips available. Would you like to select some alternate dates?” | Frame #2 |
Act 1 | no_result(str_date, end_date) | |
Act 2 | suggest(str_date, end_date) |
Frames
Frame Change Statistics
16
Maximum number of frame changes
3.58
Average number of frame changes
Frame Statistics
35
Maximum number of frames in a conversation
6.71
Average number of frames in a conversation
People
Hannes Schulz
Senior Researcher
Shikhar Sharma
Senior Research Software Development Engineer
Justin D. Harris
Principal Software Engineer
Rahul Mehrotra
Senior Program Manager
Jeremie Zumer
SDE II