Finding Fraudulent Websites Using Twitter Streams

Social media offers new opportunities for Internet fraudsters to direct traffic to websites that make fraudulent offers. In an interesting twist, however, social media also enables consumer protection organizations and government agencies to monitor traffic for suspicious links to these websites. I developed an automated monitoring algorithm that searches Twitter in real time for suspicious links appearing in tweets. Results summary: Searching for tweets containing a URL and at least one of the keywords muscle, weight, diet, acai, cambogia, lose fast, and miracle pill, my program downloaded more than 70,000 tweets during a 24-hour period. The most-tweeted URL was tweeted over Rothchild D. Finding Fraudulent Websites Using Twitter Streams. Technology Science. 2015092905. September 29, 2015.http://techscience.org/a/2015092905 2 12,000 times. I visited the 50 most tweeted URLs and classified 28 of 50 (or 56 percent) of them as suspicious and 20 of 50 (or 40 percent) as unsuspicious. The 28 suspicious URLs redirected to only 8 distinct URLs. Of the 10 most frequently tweeted URLs, all 10 were suspicious.