Collaborative Ranking and Profiling: Exploiting the Wisdom of Crowds in Tailored Web Search

Abstract

Abstract. Popular search engines essentially rely on information about the structure of the graph of linked elements to find the most relevant results for a given query. While this approach is satisfactory for popular interest domains or when the user expectations follow the main trend, it is very sensitive to the case of ambiguous queries, where queries can have answers over several different domains. Elements pertaining to an implicitly targeted interest domain with low popularity are usually ranked lower than expected by the user. This is a consequence of the poor usage of user-centric information in search engines. Leveraging semantic information can help avoid such situations by proposing complementary results that are carefully tailored to match user interests. This paper proposes a collaborative search companion system, CoFeed, that collects user search queries and accesses feedback to build user-and document-centric profiling information. Over time, the system constructs ranked collections of elements that maintain the required information diversity and enhance the user search experience by presenting additional results tailored to the user interest space. This collaborative search companion requires a supporting architecture adapted to large user populations generating high request loads. To that end, it integrates mechanisms for ensuring scalability and load balancing of the service under varying loads and user interest distributions. Experiments with a deployed prototype highlight the efficiency of the system by analyzing improvement in search relevance, computational cost, scalability and load balance.

CADA: Collaborative Auditing for Distributed Aggregation

“…In this case, the servers under the control of the attacker may generate fake insertions, e.g., to favor a given value over the others regardless of the inputs of the clients. As an example, in an aggregation system collecting feedback on website accesses and search queries [11], an Figure 3: Servers in aggregation middleware can bias the results in two ways: by performing fake insertions (insertion bias) or by returning incorrect aggregation results for the keys they are in charge of (aggregation bias).…”

“…Examples of such applications include monitoring, feedback aggregation [11], search mechanisms [4], [21], trust management [25], or popularity tracking and monitoring [5], [18]. More specifically, these systems rely on the aggregation of discrete distributions of data over a set of possible values.…”

On Task Assignment for Real-Time Reliable Crowdsourcing

“…Such automated tools can offer a common disaster view and help organizations ascertain the current situation, complementing the information available through fixed, static infrastructures. Other examples include automating stakeholder analysis [5], improving the search engines results [6], detecting service-level network events [7], and identifying transportation events of interest [8].…”

Gossip-Based Networking for Internet-Scale Distributed Systems

“…Each node acts both as a client of the provided service and at the same time as a server, collaborating with other peers to provide this same service. Examples of this first class of massive-scale systems include file sharing networks [1][2][3], collaborative search engines [4], multicast systems [5,6], publish/subscribe [7,8], etc.…”