LTRRS: A Learning to Rank Based Algorithm for Resource Selection in Distributed Information Retrieval

2019 ◽  
pp. 52-63
Author(s):  
Tianfeng Wu ◽  
Xiaofeng Liu ◽  
Shoubin Dong
Keyword(s):  
Information Retrieval ◽  
Resource Selection ◽  
Learning To Rank ◽  
Distributed Information ◽  
Distributed Information Retrieval

Distributed Information Retrieval: Developments and Strategies

2015 ◽  
Vol 16 ◽  
pp. 110-144 ◽  
Author(s):  
Benjamin Ghansah ◽  
Sheng Li Wu
Keyword(s):  
Information Retrieval ◽  
Real Time ◽  
Network Traffic ◽  
Resource Selection ◽  
Federated Search ◽  
Comprehensive Review ◽  
Distributed Information ◽  
Distributed Information Retrieval ◽  
Result Merging ◽  
Resource Description

Opposed to centralized search where Websites are crawled and indexed, Distributed Information Retrieval (DIR), also known as Federated Search, is a powerful way to comprehensively search multiple databases in real-time simultaneously. DIR is preferred to centralized search environments in a number of ways, characteristically among them are: 1. the diversity of resources that are made available; 2. improving scalability and reducing server load and network traffic; 3. the leverage of accessing the hidden or deep Web.There are three major phases/tasks of a DIR (i) resource description or collection representation (ii) resource selection and (iii) result merging. This paper aims at providing a comprehensive review on the various phases of DIR and also some current strategies being recommended in enhancing and improving the smooth implementation of a DIR system.

DISTRIBUTED INFORMATION RETRIEVAL: A MULTI-OBJECTIVE RESOURCE SELECTION APPROACH

2003 ◽  
Vol 11 (supp01) ◽  
pp. 83-99 ◽  
Author(s):  
SHENGLI WU ◽  
FABIO CRESTANI
Keyword(s):  
Information Retrieval ◽  
Data Fusion ◽  
Resource Selection ◽  
Distributed Information ◽  
Monetary Cost ◽  
Distributed Information Retrieval ◽  
Multi Objective ◽  
Selection Approach ◽  
Objective Model ◽  
The Given

Information retrieval is becoming increasingly concerned with resource selection and data fusion for distributed archives. In distributed information retrieval, a user submits a query to a broker, which determines a solution for how to yield a given number of documents from all available resources. In this paper, we present a multi-objective model for resource selection, in which four aspects: a document's relevance to the given query, time, monetary cost, and the chance of getting document duplicates from resources, are considered simultaneously. Some variants of this multi-objective model, aimed at achieving better implementation efficiency, are also proposed.

Improving Results Aggregation Strategies in Distributed Information Retrieval

2015 ◽  
Vol 17 ◽  
pp. 94-104
Author(s):  
Benjamin Ghansah ◽  
Sheng Li Wu ◽  
Nathaniel Ekow Ghansah
Keyword(s):  
Information Retrieval ◽  
User Satisfaction ◽  
Information Needs ◽  
Relevant Information ◽  
General Purpose ◽  
Distributed Information ◽  
Distributed Information Retrieval ◽  
Result Diversification ◽  
Result Merging ◽  
Ranked List

The top-ranked documents from various information sources that are merged together into a unified ranked list may cover the same piece of relevant information, and cannot satisfy different user needs. Result diversification(RD) solves this problem by diversifying results to cover more information needs. In recent times, RD has attracted much attention as a means of increasing user satisfaction in general purpose search engines. A myriad of approaches have been proposed in the related works for the diversification problem. However, no concrete study of search result diversification has been done in a Distributed Information Retrieval(DIR) setting. In this paper, we survey, classify and propose a theoretical framework that aims at improving diversification at the result merging phase of a DIR environment.

Sign in / Sign up

Export Citation Format

Share Document