Ruj Akavipat

Small world peer networks in distributed web search

In ongoing research, a collaborative peer network application is being proposed to address the scalability limitations of centralized search engines. Here we introduce a local adaptive routing algorithm used to dynamically change the topology of the peer network based on a simple learning scheme driven by query response interactions among neighbors. We test the algorithm via simulations with 70 model users based on actual Web crawls. We find that the network topology rapidly converges from a random network to a small world network, with emerging clusters that match the user communities with shared interests.

ReDS: A Framework for Reputation-Enhanced DHTs

Distributed hash tables (DHTs), such as Chord and Kademlia, offer an efficient means to locate resources in peer-to-peer networks. Unfortunately, malicious nodes on a lookup path can easily subvert such queries. Several systems, including Halo (based on Chord) and Kad (based on Kademlia), mitigate such attacks by using redundant lookup queries. Much greater assurance can be provided; we present Reputation for Directory Services (ReDS), a framework for enhancing lookups in redundant DHTs by tracking how well other nodes service lookup requests. We describe how the ReDS technique can be applied to virtually any redundant DHT including Halo and Kad. We also study the collaborative identification and removal of bad lookup paths in a way that does not rely on the sharing of reputation scores, and we show that such sharing is vulnerable to attacks that make it unsuitable for most applications of ReDS. Through extensive simulations, we demonstrate that ReDS improves lookup success rates for Halo and Kad by 80 percent or more over a wide range of conditions, even against strategic attackers attempting to game their reputation scores and in the presence of node churn.

Sixearch.org 2.0 peer application for collaborative web search

Sixearch.org is a peer application for social, distributed, adaptive Web search, which integrates the Sixearch.org protocol, a topical crawler, a document indexing system, a retrieval engine, a P2P network communication system, and a contextual learning system. With a single click, the Sixearch.org application will build your personal Web collection. You can search not only your collection, but also other Sixearch peers. When you submit a query, your Sixearch agent will determine which peers are best suited to answer it based on previous interactions. Your agent will also learn from the results it receives, so that it can continuously improve.

ReDS: reputation for directory services in P2P systems

Peer-to-peer (P2P) architectures are gaining popularity and importance for applications ranging from massive-scale Internet content delivery to mobile social networks. Such P2P systems must provide directory services for locating peers with the desired content and services. These directory services are themselves decentralized, such as with distributed hash tables (DHTs), which allow for efficient locating of objects without any centralized directory. Being a distributed system over a diverse set of untrusted nodes, however, such directory services must be resilient to adversarial behavior. Otherwise, the entire P2P system can be crippled by manipulating or simply denying access to resources. We propose Reputation for Directory Services (ReDS), a framework for using reputation management to enhance the security of finding information in distributed systems. While previous reputation systems have addressed several specific applications of P2P networks (e.g., by identifying peers who share bad files), directory services form the backbone of P2P systems and have unique properties with respect to reputation that make them worth investigating. In this extended abstract, we motivate our investigation of ReDS and describe preliminary results that show its effectiveness in the Salsa P2P system.

ExpertEyes: Open-source, high-definition eyetracking

ExpertEyes is a low-cost, open-source package of hardware and software that is designed to provide portable high-definition eyetracking. The project involves several technological innovations, including portability, high-definition video recording, and multiplatform software support. It was designed for challenging recording environments, and all processing is done offline to allow for optimization of parameter estimation. The pupil and corneal reflection are estimated using a novel forward eye model that simultaneously fits both the pupil and the corneal reflection with full ellipses, addressing a common situation in which the corneal reflection sits at the edge of the pupil and therefore breaks the contour of the ellipse. The accuracy and precision of the system are comparable to or better than what is available in commercial eyetracking systems, with a typical accuracy of less than 0.4° and best accuracy below 0.3°, and with a typical precision (SD method) around 0.3° and best precision below 0.2°. Part of the success of the system comes from a high-resolution eye image. The high image quality results from uncasing common digital camcorders and recording directly to SD cards, which avoids the limitations of the analog NTSC format. The software is freely downloadable, and complete hardware plans are available, along with sources for custom parts.

Intelligent Peer Networks for Collaborative Web Search

Collaborative query routing is a new paradigm for Web search that treats both established search engines and other publicly available indices as intelligent peer agents in a search network. The approach makes it transparent for anyone to build their own (micro) search engine, by integrating established Web search services, desktop search, and topical crawling techniques. The challenge in this model is that each of these agents must learn about its environment— the existence, knowledge, diversity, reliability, and trustworthiness of other agents — by analyzing the queries received from and results exchanged with these other agents. We present the 6S peer network, which uses machine learning techniques to learn about the changing query environment. We show that simple reinforcement learning algorithms are sufficient to detect and exploit semantic locality in the network, resulting in efficient routing and high-quality search results. A prototype of 6S is available for public use and is intended to assist in the evaluation of different AI techniques employed by the networked agents.

Adaptive query routing in peer web search

An unstructured peer network application was proposed to address the query forwarding problem of distributed search engines and scalability limitations of centralized search engines. Here we present novel techniques to improve local adaptive routing, showing they perform significantly better than a simple learning scheme driven by query response interactions among neighbors. We validate prototypes of our peer network application via simulations with 500 model users based on actual Web crawls. We finally compare the quality of the results with those obtained by centralized search engines, suggesting that our application can draw advantages from the context and coverage of the peer collective.

MANORAA (Mapping Analogous Nuclei Onto Residue And Affinity) for identifying protein–ligand fragment interaction, pathways and SNPs

Protein–ligand interaction analysis is an important step of drug design and protein engineering in order to predict the binding affinity and selectivity between ligands to the target proteins. To date, there are more than 100 000 structures available in the Protein Data Bank (PDB), of which ∼30% are protein–ligand (MW below 1000 Da) complexes. We have developed the integrative web server MANORAA (Mapping Analogous Nuclei Onto Residue And Affinity) with the aim of providing a user-friendly web interface to assist structural study and design of protein–ligand interactions. In brief, the server allows the users to input the chemical fragments and present all the unique molecular interactions to the target proteins with available three-dimensional structures in the PDB. The users can also link the ligands of interest to assess possible off-target proteins, human variants and pathway information using our all-in-one integrated tools. Taken together, we envisage that the server will facilitate and improve the study of protein–ligand interactions by allowing observation and comparison of ligand interactions with multiple proteins at the same time. (http://manoraa.org).

Analyzing the efficiency of strategies for MAS-based sensor interpretation and diagnosis

One of the factors holding back the application of multi-agent, distributed approaches to large-scale sensor interpretation and diagnosis problems is the lack of good techniques for predicting the performance of potential systems. In this paper we use a consideration of Bayesian network inference algorithms to construct formulas that describe the computational and communication resources required by several strategies for MAS-based distributed SI/diagnosis.