Suggests clothing shops to you based on Active Learning strategies. Just a research project!
The conversation-based active learning critiquing algorithm used is heavily based on
Tweaking Critiquing, Lorraine McGinty and Barry Smyth, 2003.
PDF (as of June 2013)
It differs by allowing negative critiques and uses these as an indicator of positive or negative progress. Why? Imagine a clothing item selection: you may not always say 'Hey, I like this color!', but also 'OMG, this color looks bad!'. It also helps limit the options for use on a smaller screen.
- Create a new class in shopr.algorithm.model extending GenericAttribute.
- Define end-user name in CritiqueActivity's ItemFeatureAdapter.getView().
The code in this repository is licensed under Apache License, Version 2.0 except where noted otherwise.
Other content is licensed under a Creative Commons Attribution 3.0 Germany License. Be aware that contents of the data set may be subject to intellectual property rights.