Another startup idea: social network filtering using predictive scores

[admin]

Purpose: spam detection, elimination of spammy posts, elimination of problem members (spammers / scammers) and automated rejection of member applications, in networks that require new members to be approved.

Methodology: when a new member signs up, use a web crawler to harvest info about the email address and the self-reported name of the person. The web crawler will look for spam posted elsewhere by the same name / email before membership approval. Other metrics to consider: name and email add…

More blog entries from AnalyticBridge...