Donation Dashboard 1.0
A Berkeley Center for New Media Project

Project Team:

Tavi Nathanson, Ephrat Bitton, Ke Xu, Connie Wang, Prof. Ken Goldberg
Automation Sciences Lab, UC Berkeley

Based on discussions with Jim Buckmaster, CEO,

Visual design by Gil Gershoni, Gershoni Creative Agency

Charity Information Sources:

Related Information:

Related Links:

What is Donation Dashboard?

Press Release

The Generosity of Crowds: New Website Matches Non-Profits to Donors

"It's more difficult to give money away intelligently than it is to earn it in the first place." -- Andrew Carnegie

The US has over a million registered non-profit institutions, ranging from the Red Cross to the National Rifle Association, including tens of thousands you've never heard of. You'd like to contribute, but effectively allocating your available funds among all these good causes seems like a hopeless task.

Giving is getting easier with an experimental website called "Donation Dashboard," which uses machine learning techniques to recommend a customized portfolio of good causes based on your personal ratings of sample non-profit organizations.

Here's how it works: you are presented with brief descriptions of non- profit institutions and asked to rate each in terms of how interested you are in donating to it. The system analyzes your ratings in light of others' ratings and does its best to allocate your available funds in proportion to your interests. Your customized "donation portfolio" is presented in an easy-to-understand pie chart that you can save at the site for future reference.

Donation Dashboard, which is being developed by the Berkeley Center for New Media, extends machine learning techniques used by commercial websites to recommend movies, music, and books. Donation Dashboard goes beyond existing charity ranking sites by statistically combining your ratings with the ratings entered by your fellow good samaritans to compute a porfolio customized to your interests.

The Donation Dashboard website is a pilot system that includes information on 70 non-profit institutions. If the system is successful, the developers hope to expand it with other features and partner with a third party that can streamline collecting and distributing funds.

"There's strength in numbers; the system should improve over time as the number of ratings increases, in this sense each person who visits the site contributes to the collective wisdom about good causes," notes UC Berkeley Professor Ken Goldberg, who is developing the system with graduate students Tavi Nathanson and Ephrat Bitton at UC Berkeley, with conceptual input from Jim Buckmaster at craigslist.

Please Note:


Related Articles:

Eigentaste Introduction

Eigentaste, the algorithm that powers Donation Dashboard, was developed at UC Berkeley by Prof. Ken Goldberg et. al. It uses universal queries to elicit real-valued user ratings on a common set of items and applies principal component analysis (PCA) to the resulting dense subset of the ratings matrix. PCA facilitates dimensionality reduction for offline clustering of users and rapid computation of recommendations.

For more information about Eigentaste, please see the following paper:

Eigentaste: A Constant Time Collaborative Filtering Algorithm,
Ken Goldberg, Theresa Roeder, Dhruv Gupta, and Chris Perkins, Information Retrieval Journal, 4(2), pp. 133-151. July 2001.

Eigentaste Applications

Eigentaste was patented by UC Berkeley in 2003. It has many possible applications, such as the recommendation of books, movies, toys, stocks, and music.

It was originally used in an online joke recommendation system called Jester, which recommends new jokes to users based on their ratings of an initial set.

The Jester dataset is freely available for research use, when referenced. It is downloadable at:

New Research

Eigentaste 5.0 is the latest version of Eigentaste, which improves upon the original algorithm by dynamically adapting the order that items are recommended. It does this by integrating user clustering with item clustering and monitoring item portfolio effects.

For more information about Eigentaste 5.0, please see the following paper:

Eigentaste 5.0: Constant-Time Adaptability in a Recommender System Using Item Clustering,
Tavi Nathanson, Ephrat Bitton, and Ken Goldberg. Working Paper Track. ACM Conference on Recommender Systems (RecSys), Minneapolis, MN, Oct 2007.

Jester currently assigns users to either Eigentaste or Eigentaste 5.0, randomly (in order to compare the two algorithms). Note that Donation Dashboard currently uses the original version of Eigentaste.

More Information

For more information, see the following paper:

Algorithms, Models and Systems for Eigentaste-Based Collaborative Filtering and Visualization
Tavi Nathanson. Master's thesis, EECS Department, University of California, Berkeley, May 2009.