1

I am desperately looking for a way to either create or find an existing data set for products that lists the most common features / components or attributes of that product.

An example could be:

DSLR camera

  • Camera
  • Megapixels
  • Lens
  • Digital
  • Glass
  • Pro
  • Shutter
  • LCD
  • Optical
  • Zoom
  • Resolution
  • Megapixels
  • CCD
  • CMOS
  • Stabilisation
  • Macro
  • Telephoto
  • Wide angle
  • Filters
  • Batteries
  • Flash
  • Lighting
  • Memory card

As can be seen this example, I just listed as many attributes I could, related to a specific product. Such that you might do in a brainstorming session on a product.

I need a way in order to do this at scale. I am looking for all help possible on this, it would be greatly appreciated!

migrated from datascience.stackexchange.com May 18 '18 at 1:46

This question came from our site for Data science professionals, Machine Learning specialists, and those interested in learning more about the field.

2

A quick search led me to this set of product details from Amazon in 2014.

The description mentions:

The dataset covers products from 6 main categories, Automotive, Books, Electronics, Movies, Phones and Home including 1529 sub-categories. All products are listed over 334 independent attributes & the size 2000K of value space.

Some of the attributes might be a little useless for a machine learning model, but other may help you do things quite nicely, like clustering.

Here is an actual example from the dataset for a DSLR camera:

ITEM 1604
Binding=Electronics
Brand=Case Logic
Color=Black
EAN=0085854223799
EANList=0085854223799
Feature=Compatible with DSLR camera with attached lens
Feature=Fitted Day Holster provides bump and scratch protection
Feature=Holster tether attaches to the camera's strap so you never set the case down to take a shot
Feature=Internal dedicated lens cap slip pocket, so your lens cap is never misplaced
Feature=Holds Cameras up to 5.9 x 4 x 6 x 7.5"
ItemDimensions=46059024750
Label=Case Logic
ListPrice=2999USD$29.99
Manufacturer=Case Logic
Model=SHC-101-BK
MPN=SHC-101-BK
PackageDimensions=27694526787
PackageQuantity=4
PartNumber=SHC-101-BK
ProductGroup=Photography
ProductTypeName=CAMERA_BAGS_AND_CASES
Publisher=Case Logic
SKU=12360967Pixmania14388
Studio=Case Logic
Title=Case Logic SHC-101-BK DSLR Day Holster, Small, Black
UPC=085854223799
UPCList=085854223799
  • +1! Definitely a great find. I’ll find a use for this dataset. =) – Christopher Klaus May 18 '18 at 4:58
0

You may also want to consider the GS1 Global Product Classification (GPC).

GPC classifies products by grouping them into categories based on their essential properties as well as their relationships to other products. GPC offers a universal set of standards for everything from a car to a litre of milk. 

The GS1 standards are the basis for global supply chains.

GPC Hierarchy:

GPC Example Hierarchy

Brick attributes / values.

Example of brick attributes / values

After that there are taxonomies / dictionaries for attribute values.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy