FireShare has do do with online learning and is a subject of academic research being performed at Utah State University. Here are some papers and resources associated with the ongoing research.
Authors: Dennis Muhlestein and SeungJin Lim
On-line communities on the Internet are highly selforganizing, dynamic and ubiquitous. The prime interest of peers in this community is often sharing common interest, even when compromising privacy. This paper presents a peer coordination strategy and a data sharing process for peers on the Internet which allows them to discover their common interest in terms of sets of frequently visited URLs. To this end, sample data was collected by randomly following links on popular websites to simulate the algorithm in operation. Experiments were then performed to compare the number of discovered frequently visited URL sets and association rules with the overhead induced by our network.
Computer Science Masters Thesis
Author: Dennis Muhlestein
Communities on the Internet are highly self-organizing, dynamic and ubiquitous. The prime interest of peers in this community is often sharing common interests, even when compromising privacy. This thesis presents a model for peers on the Internet which allows them to discover their common interests in terms of sets of frequently visited URLs. This allows users to automatically be presented with URLs related to what they are currently browsing, thus saving them time searching for new information and helping to educate them on the current topic. FireShare was developed as a plugin for the popular web browser FireFox to implement the model and collect test data. An analysis of the data collected was then performed to compare the number of discovered frequently visited URL sets and association rules with the overhead induced by the network. While the proposed model was moderately validated with FireShare, analysis of the test data submitted shows high potential for success with future versions.
There are two data sets associated with the FireShare Experiment. One data set contains records of clients making requests to URLs on a web server. The other is the test results submitted by FireShare clients.
This data set contains one client access record on each line. Clients and URLs are numerically numbered in the order they were received by the FireShare server. There is also a timestamp with each record. The data is in the following format:
client_id,url_id,timestamp\n # timesamp is denoted by number of seconds since January 1st, 1970
Each time FireShare performs its data mining algorithm, it submits test data back to the server. (This functionality can be opted out of on the preferences dialog.) This data contains an integer representation for local visits, peers, and the urls that peers have visited. It also contains any association rules that the client discovered. The data is in the following format:
<test_data> = local_visits: <urls>\n<peers>\n<rules>\n #local_visits, peers and rules are not necessarily in order
<peers> = <peer>|<peer>\n<peers>
<peer> = peer_<id>: <urls>
<rules> = <rule>|<rule>\n<rules>
<rule> = rule_<id>: <urls> => <url> "|" <confidence> (<support>)
<confidence> = float
<support> = integer #(actual number of peers instead of percentage of total peers)
<id> = integer
<urls> = <url>|<url>,<urls>
<url> = integer
Note that there will be two newlines between each clients test submission.