An Overview of Query Processing on Crowdsourced Databases

TL;DR

This paper surveys the current state and future trends of query processing in crowdsourced databases, highlighting methods, strengths, weaknesses, and the integration of human computation with traditional database systems.

Contribution

It provides a comprehensive overview of existing approaches to query processing in CSDBs and discusses future research directions in the field.

Findings

Analysis of traditional and preference query processing methods

Identification of strengths and weaknesses of current approaches

Discussion of future research trends in CSDB query processing

Abstract

Crowd-sourcing is a powerful solution for finding correct answers to expensive and unanswered queries in databases, including those with uncertain and incomplete data. Attempts to use crowd-sourcing to exploit human abilities to process these expensive queries using human workers have helped to provide accurate results by utilising the available data in the crowd. Crowd-sourcing database systems (CSDBs) combine the knowledge of the crowd with a relational database by using some variant of a relational database with minor changes. This paper surveys the leading studies conducted in the area of query processing with regard to both traditional and preference queries in CSDBs. The focus of this work is on highlighting the strengths and the weakness of each approach. A detailed discussion of current and future trends research associated with query processing in the area of CSDBs is also presented.