Distinguishing between Humans and Robots on the Web

Abstract

We present a new detection technique for distinguishing humans from bots on the web, comprised of one existing active and one existing passive technique. We arrived at the proposed new detection technique by evaluating each of 10 of the most prevalent existing techniques. Each technique was evaluated in three categories of criteria: administrator experience, user experience, and effectiveness. The administrator experience was evaluated based on the technique’s level of automation, and its library availability. The user experience was evaluated based on whether or not the user’s input was correct, and the difficulty rating they assigned. The effectiveness was evaluated based on a literature survey to determine the quantitative efficacy of current techniques. We compiled our data to determine a numerical score for each technique based on a rubric we created. Our proposed new detection technique combines the top scoring user-active and top scoring user-passive techniques.