veyepar: Search Logs + Machine Learning = Auto-Tagged Inventory

Hi user

Admin Login:

Name: Search Logs + Machine Learning = Auto-Tagged Inventory
--client pyohio --show pyohio_2019 --room cartoon2 14866 --force
Prev: 12 Learn How Computers Work Between Silicon and Assembly: Build a CPU with Python
Next: 12 Dynamic Data Pipelining with Luigi

show more...

Author(s):	John Berryman
Location	Cartoon 2
Date	jul Sun 28	Days Raw Files
Start	15:15	First Raw Start	15:08
Duration	0:45:0	Offset	0:06:46
End	16:00	Last Raw End	16:08
Chapters	00:00 0:17:06
Total cuts_time	42 min.

https://www.pyohio.org/2019/presentations/136
raw-playlist raw-mp4-playlist encoded-files-playlist host archive tweet mp4 svg png

final audio

assets release.pdf Search_Logs_Machine_Learning_AutoTagged_Inventory.json
logs
Admin: episode episode list cut list raw files day marks day marks day image_files

State:
Locked:	clear this to unlock
Locked by:	user/process that locked.
Start:	initially scheduled time from master, adjusted to match reality
Duration:	length in hh:mm:ss
Name:	Video Title (shows in video search results)
Emails:	email(s) of the presenter(s)
Released:	has someone authorised pubication
Normalise:
Channelcopy:	m=mono, 01=copy left to right, 10=right to left, 00=ignore.
Thumbnail:	filename.png
Description:	For e-commerce applications, matching users with the items they want is the name of the game. If they can't find what they want then how can they buy anything?! Typically this functionality is provided through search and browse experience. Search allows users to type in text and match against the text of the items in the inventory. Browse allows users to select filters and slice-and-dice the inventory down to the subset they are interested in. But with the shift toward mobile devices, no one wants to type anymore - thus browse is becoming dominant in the e-commerce experience. But there's a problem! What if your inventory is not categorized? Perhaps your inventory is user generated or generated by external providers who don't tag and categorize the inventory. No categories and no tags means no browse experience and missed sales. You could hire an army of taxonomists and curators to tag items - but training and curation will be expensive. You can demand that your providers tag their items and adhere to your taxonomy - but providers will buck this new requirement unless they see obvious and immediate benefit. Worse, providers might use tags to game the system - artificially placing themselves in the wrong category to drive more sales. Worst of all, creating the right taxonomy is hard. You have to structure a taxonomy to realistically represent how your customers think about the inventory. Eventbrite is investigating a tantalizing alternative: using a combination of customer interactions and machine learning to automatically tag and categorize our inventory. As customers interact with our platform - as they search for events and click on and purchase events that interest them - we implicitly gather information about how our users think about our inventory. Search text effectively acts like a tag and a click on an event card is a vote for that clicked event is representative of that tag. We are able to use this stream of information as training data for a machine learning classification model; and as we receive new inventory, we can automatically tag it with the text that customers will likely use when searching for it. This makes it possible to better understand our inventory, our supply and demand, and most importantly this allows us to build the browse experience that customers demand. In this talk I will explain in depth the problem space and Eventbrite's approach in solving the problem. I will describe how we gathered training data from our search and click logs, and how we built and refined the model. I will present the output of the model and discuss both the positive results of our work as well as the work left to be done. Those attending this talk will leave with some new ideas to take back to their own business. Eventbrite is exploring a new machine learning approach that allows us to harvest data from customer search logs and automatically tag events based upon their content. The results have allowed us to provide users with a better inventory browsing experience. markdown
Comment:	production notes

recording sheet

recording sheet

2019-07-28/15_08_14.ts Apply: 15:08:14 - 15:21:01 ( 00:12:47 ) S: 15:08:14 - E: 15:38:14 D: 00:30:00 (End: 767.0) show more...

vlc ~/Videos/veyepar/pyohio/pyohio_2019/dv/cartoon2/2019-07-28/15_08_14.ts :start-time=00.0 --audio-desync=0

	Raw File	Cut List
	15:08:14		seconds: 0.0 Wall: 15:08:14
Duration	00:30:00
	15:38:14		seconds: 767.0 Wall: 15:21:01
Comments:
mp4 mp4.m3u	dv.m3u	Split: Sequence:: delete

audio viz

2019-07-28/15_08_14.ts Apply: 15:21:01 - 15:21:08 ( 00:00:07 ) S: 15:08:14 - E: 15:38:14 D: 00:30:00 (Start: 767.0) (End: 774.0) show more...

vlc ~/Videos/veyepar/pyohio/pyohio_2019/dv/cartoon2/2019-07-28/15_08_14.ts :start-time=0767.0 --audio-desync=0

	Raw File	Cut List
	15:08:14		seconds: 767.0 Wall: 15:21:01
Duration	00:30:00
	15:38:14		seconds: 774.0 Wall: 15:21:08
Comments:
mp4 mp4.m3u	dv.m3u	Split: Sequence:: delete

audio viz

2019-07-28/15_08_14.ts Apply: 15:21:08 - 15:38:14 ( 00:17:06 ) S: 15:08:14 - E: 15:38:14 D: 00:30:00 (Start: 774.0) show more...

vlc ~/Videos/veyepar/pyohio/pyohio_2019/dv/cartoon2/2019-07-28/15_08_14.ts :start-time=0774.0 --audio-desync=0

	Raw File	Cut List
	15:08:14		seconds: 774.0 Wall: 15:21:08
Duration	00:30:00
	15:38:14		seconds: 0.0 Wall: 15:08:14
Comments:
mp4 mp4.m3u	dv.m3u	Split: Sequence:: delete

audio viz

2019-07-28/15_38_14.ts Apply: 15:38:14 - 16:03:36 ( 00:25:22 ) S: 15:38:14 - E: 16:08:13 D: 00:29:59 (End: 1522.0) show more...

vlc ~/Videos/veyepar/pyohio/pyohio_2019/dv/cartoon2/2019-07-28/15_38_14.ts :start-time=00.0 --audio-desync=0

	Raw File	Cut List
	15:38:14		seconds: 0.0 Wall: 15:38:14
Duration	00:29:59
	16:08:13		seconds: 1522.0 Wall: 16:03:36
Comments:
mp4 mp4.m3u	dv.m3u	Split: Sequence:: delete

audio viz

2019-07-28/15_38_14.ts Apply: 16:03:36 - 16:08:13 ( 00:04:37 ) S: 15:38:14 - E: 16:08:13 D: 00:29:59 (Start: 1522.0) show more...

vlc ~/Videos/veyepar/pyohio/pyohio_2019/dv/cartoon2/2019-07-28/15_38_14.ts :start-time=01522.0 --audio-desync=0

	Raw File	Cut List
	15:38:14		seconds: 1522.0 Wall: 16:03:36
Duration	00:29:59
	16:08:13		seconds: 0.0 Wall: 15:38:14
Comments:
mp4 mp4.m3u	dv.m3u	Split: Sequence:: delete

audio viz

Rf filename:	root is .../show/dv/location/, example: 2013-03-13/13:13:30.dv
Sequence:
get this:	check and save to add this

2019-07-28/15_08_14.ts
audio viz

audio viz

2019-07-28/15_38_14.ts
audio viz

audio viz

Veyepar Video Eyeball Processor and Review