|
Website Mirror at the University of São Paulo |
Full Paper AbstractsApproximate Object Location and Spam Filtering on Peer-to-peer SystemsFeng Zhou, Li Zhuang, Ben Y. Zhao, Ling Huang, Anthony D. Joseph and John Kubiatowics (U.C.Berkeley) Recent work in P2P overlay networks allow for decentralized object routing and location (DOLR) across networks based on unique IDs. We observe that many network applications involve the management of similar, but not identical replicas, creating a dilemma for content-hashed naming schemes in existing P2P systems. In this paper, we propose an extension to DOLR systems to publish objects using generic Feature Vectors instead of content-hashed GUIDs, which enables the systems to locate similar objects. We discuss the design of a distributed text similarity engine, named Approximate Text Addressing, built on top of this extension that locates objects by their text descriptions. Later we outline the design and implementation of a killer-app on ATA, a decentralized spam-filtering service. We achieve spam identification ratio of over 97% with zero false positive in our evaluation done with 30,000 real spam emails and 10,000 normal emails. |
Latest update: 13 June 2003 - Questions and Comments about the Site: fmc@inf.ufg.br |