===============
I wrote the post first in 'The Zen Pub' section, but I haven't realized that very few people goes there. I hope that this little off topic won't anger You nor admins, and Your support will be very helpful for me.

Dear Zenners,

Some of You might know me as author of Query Cache and few other modules. In my spare time (very limited though) I tried to support Zen Cart community with performance modules, database optimization tips and few bug reports.

Now I would like to ask You, Zen Cart community for help.

I’m Ph.D. student at West Pomerian University of Technology and I want to focus my research on shopping cart recommendation systems. I feel that current data mining methods are inconvenient and inadequate for ecommerce sites. They are slow, make poor predictions and usually does not work on dynamic data. I want to improve that.

What I need for my research

Data. Plain and simple. To work on efficient algorithms focused on e-commerce I need real world data from ecommerce sites like Yours.

What I’m asking for is:

• anonymized list of product ids Your customers bought,
• anonymized list of order ids,
• anonymized list of customer ids,
• nothing else.

What ‘anonymized data’ means? Basicly:

• the data can’t be linked back to You,
• Your customers data is not included,
• I don't know which products Your customers purchased.

If You wonder how that kind of data looks like? Here’s example (after anonymization, example from real e-commerce site):

1967921937,1633775438,1489186838
1582785331,117776134,2142463074
1472986697,92839866,721477612

What I’m NOT asking for

Basicly I don’t need any data that would identify You or Your customer or put them on any risk. In particular I don’t ask for:

• credit card numbers,
• any transaction data,
• email addresses,
• names,
• physical addresses (billing address nor delivery address),
• product prices,
• gender,
• date of birth of Your customers,
• PayPal (nor any other payment system) transaction ids, user ids etc.

How can You help

You can support my research by sharing with me anonymized data from Your store. I’ve created script that automates the process and does data anonymization on the fly, so You don’t have to do it Yourself. Code of the script is included so that You or the community may review it.

I will use the data only for my research. The data will not be transferred, shared or sold to any third party. I will not name Your company in my papers unless You explicitly state that You want that.

If You decide to help me please follow these instructions:

• download package with the script http://www.data-diggers.com/research/zc-get-data.zip (I also attached the file to the post)
• extract it and upload “dd_get_data.php” file to the root folder of Your store (there where index.php exists). Now go to http://www.YourStore.com/dd_get_data.php
• You will be asked to identify Yourself with admin password and login (so no one besides You could download the data).
• After successful authentication the script will generate file data_1234567.csv (or with similar name).
• Please mail the file (“data_124567.csv” or with similar name) to me (zip it if You like): [email protected] or [email protected]
• After that You can safely delete the file “dd_get_data.php”

Want to stay as anonymous as possible, yet You want to help me? Mail me the file from one time hotmail, gmail or yahoo account. There’s no way anyone (even me) could link the file back to.

I want to state again that no crucial data about Your store will be downloaded. No payment, customer, credit card or Your personal/company data is gathered. The data that is fetched is anonymized with one way function, meaning that e.g. although I may infer that some product is a bestseller there’s no way for me (or anybody else) to know which one it is.

If You have any questions please don’t hesitate to ask them - I’ll try to answer them all.

Thank You for Your support,
Maciej Klepaczewski aka. data_digger