MarCol: A market-based recommender system

MarCol, a Web-based, market-based collaborative information-filtering (IF) system employs a pricing mechanism to motivate users to provides their own judgments based on their preferences. This system uses Google as the underlying search engine and store all user logs, including queries and judgments. The filtering engine of the MarCol, generates the documents most relevant to the user's query by activating the Google for nonevaluated documents and by obtaining documents from the local collaborative database. When the user wants to use recommendations or provide judgments, the filtering engine activates the pricing evaluation to compute the payment or reward. The pricing mechanism used in MarCol can also identify fraudulent judgments submitted by programs or users to earn points. The system can postpone a user's reward until it collects sufficient previous judgment for the document to validate the judgment's credibility.