Thick Profile

Home

Author

Jason Brownlee

Description

Thick Profile is an experiment into the preparation and exploitation of a deep digital profile of an online user. The project was conceived and developed over a two week period in late April 2008 by Jason Brownlee.

The approach involved constructing a model of a user based on the users specification of written contributions such as websites, technical reports, research papers, and blog posts. The development effort involved the preparation of code to parse such data data sources, construct a model, and visualizing that model as a graph or tag cloud.

Specific experiments were performed to assess the general approach, the results of which were captured in a series of three technical reports.

Running

1. Acquire the source code for the thickprofile project (either download the source or clone the git repository)
   1. Project home
2. Compile the source code using ant (assumes you have apache ant installed and configured)
   1. cd thickprofile/source
   2. ant

The codebase is predominately a framework for loading and parsing data sources and preparing and visualizing a text-based user model. There is no main GUI or program. Instead, a series of experiments and tests were prepared, each of which is its own independent program. These experiments and tests are located in the source/src directory in the packages: com.thickprofile.application.* and com.thickprofile.application.experiments.*

Reports

Three technical reports were written to capture and communicate the project premise and experimental results. The reports are provided in the project as PDFs in the ‘reports’ directory of the project. The following summarizes their content:

• Project: Thick Profile, Jason Brownlee, Technical Report 20080417A
  • This report provides an high level overview of the premise of the project including the motivation as to why one would be interested in preparing so-called ‘thick profiles’ of users. This report also presents a vision for three orders of thick user profiles that increase in sophistication from a minimal profile page to an embodied agent that acts on behalf of the user.
• Thick Profile Project: Users, Sources, And Tag Models, Jason Brownlee, Technical Report 20080419A
  • This report describes the general approach of building a tag-based thick model of a user from a series of specified online resources, and then visualizing the prepared models. The reports captures the state of the framework codebase and highlights its capabilities. Example visualizations are provided and the method for constructing a tag model of user source data is summarized along with the suggested application of the model: relative model affinity.
• Thick Profile Project: Interestingness Experiments, Jason Brownlee, Technical Report 20080426A
  • Summarizes a series of experiments on jasonb (Jason Brownlee) drawing upon his sources of technical reports, webpages and blog posts. Experiments include holistic tag cloud visualization, visualization of the user model affinity by year, holistic artifact interestingness, piecewise artifact interestingness, and a summary of possible extensions.

Technology

The Thick Profile codebase was written in Java version 1.6. The reports were written in MS Word and compiled to PDF. Java code will compile under Java 1.5.

Development

The code is partitioned into sources (such as HTML, txt, PDF, google blogs) and models (such as user source and user tag). Sources are utilities where as models are data which may be serialized and visualized. A simple command line interface is provided to capture data from the single user, and experiments are hard coded to capture specific data sources, prepare specific pre-defined models, and output specific visualizations.

Screenshots

The following are two screenshots taken from during some experimentation recorded in the prepared technical reports.

License

(The MIT License)

Copyright © 2008-2009, Jason Brownlee

Permission is hereby granted, free of charge, to any person
obtaining a copy of this software and associated documentation
files (the “Software”), to deal in the Software without
restriction, including without limitation the rights to use,
copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the
Software is furnished to do so, subject to the following
conditions:

The above copyright notice and this permission notice shall be
included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED “AS IS”, WITHOUT WARRANTY OF ANY KIND,
EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES
OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
OTHER DEALINGS IN THE SOFTWARE.