veyepar: Personalisation with Recombee

Author(s):	Murray Woodman
Location	Sassafras
Date	nov Thu 28	Days Raw Files
Start	16:15	First Raw Start	15:49
Duration	00:45:00	Offset	0:25:45
End	17:00	Last Raw End	17:12
Chapters	00:00 0:00:28 0:30:28
Total cuts_time	48 min.

https://drupalsouth.org/events/drupalsouth-hobart-2019/sessions/personalisation-recombee
raw-playlist raw-mp4-playlist encoded-files-playlist host tweet mp4 svg png

assets release.pdf Personalisation_with_Recombee.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:

markdown

Drupal provides a solid foundation for managing content. That is a solved problem. The next phase in content management is around personalising the user experience and ensuring that the experience is consistent across platforms. This presentation will provide a practical demonstration of how user behaviour and content data (from federated sites) can be used to personalise the content recommended to users.
The personalisation of content, customised to the user, is a powerful way to improve the user experience. Personalised content can deliver more relevant information to your users, keeping them engaged. It is especially useful on commerce sites where similar products can be recommended, or on content-driven sites where relevant information can be uncovered. In practical terms, it is able to increase the time users spend on a site and improve the chances of conversion.
Recombee is a leading SaaS-based recommendation engine which is able to combine user behaviour and content properties to provide real-time user and item recommendations. It exposes an API which provides tracking and recommendation functionality which can be accessed from the Drupal frontend. It offers an affordable and simple pathway to web personalisation.
The developers at Morpht have implemented a SearchAPI backend implementation for Recombee, allowing content to be synced across into the Recombee engine, augmenting the data collected through user interactions. Recommendation quality is improved, especially when relatively little data has been collected on the current user. The presentation will demonstrate the module publicly for the first time.
This presentation will cover the following areas:
- content modelling techniques for improving personalisation
- tracking user behaviour to drive recommendations
- a decoupled approach for presenting the recommendations,
- the advantages of using an external system such as GTM for orchestrating personalisation.
The session should be of interest to the following people:
- marketers who wish to find out more about practical personalisation,
- commerce and site owners who want to get more from their site,
- content modellers who are interested in modelling the enterprise,
- backend developers who are interested in SearchAPI plugins,
- front-enders who may be interested in a pragmatic implementation of snippets with handlebars templates.
The presentation is suitable for beginners and intermediates as it will mostly be conceptual in nature. There should be a practical demo for the audience to participate in as well.

Comment:

production notes