Efficient and flexible crowdsourcing of specialized tasks with precedence constraints

Abstract

Many companies now use crowdsourcing to leverage external (as well as internal) crowds to perform specialized work, and so methods of improving efficiency are critical. Tasks in crowdsourcing systems with specialized work have multiple steps and each step requires multiple skills. Steps may have different flexibilities in terms of obtaining service from one or multiple agents, due to varying levels of dependency among parts of steps. Steps of a task may have precedence constraints among them. Moreover, there are variations in loads of different types of tasks requiring different skill-sets and availabilities of different types of agents with different skill-sets. Considering these constraints together necessitates the design of novel schemes to allocate steps to agents. In addition, large crowdsourcing systems require allocation schemes that are simple, fast, decentralized and offer customers (task requesters) the freedom to choose agents. In this work we study the performance limits of such crowdsourcing systems and propose efficient allocation schemes that provably meet the performance limits under these additional requirements. We demonstrate our algorithms on data from a crowdsourcing platform run by a non-profit company and show significant improvements over current practice.

A Survey on Mobile Crowdsensing Systems: Challenges, Solutions, and Opportunities

“…Considering user trajectories and recurrent spots with a good channel condition, the authors suggest that background communication traffic can be scheduled by the client according to application data priority and expected quality data. In [180], the authors consider precedence constraints in specialized tasks among multiple steps, which require flexibility in order to the varying level of their dependency. Furthermore, they consider variations of loads needed for accomplishing different tasks and require allocation schemes that are simple, fast, decentralized and provide the possibility to choose contributing users.…”

“…However, it is possible to detect road conditions via accelerometer and GPS readings without providing explicit notification to the user [293]. As opposed to the participatory approach, the MCS platform can distribute the tasks dynamically by communicating with the background thread on the mobile device and by considering various criteria [180], [209], [210]. As opportunistic sensing completely decouples the user and MCS platform, as a result of the lack of user pre-screening mechanism on the quality of certain types of data (e.g.…”

Equity Promotion in Online Resource Allocation

“…Other related works. There are a few works that have considered online resource allocation in different contexts, see, e.g., allocation of public housing for lower-income families (Benabbou et al 2018), distribution of emergence aid for natural disasters like wildfires (Wang, Bier, and Sun 2019), task assignment in crowdsourcing platform (Chatterjee et al 2018), online set selection with fairness and diver-sity constraints (Stoyanovich, Yang, and Jagadish 2018), online resource-allocation problems with limited choices in the long-chain design (Asadpour, Wang, and Zhang 2020), and income inequality among rideshare drivers, see, e.g., (Lesmana, Zhang, and Bei 2019;Sühr et al 2019).…”

A survey of task allocation techniques in MAS

“…Depending on the technique used, an optimal, or almost always, an approximate solution can be found and a different degree of scalability, complexity and adaptability of the problem will exist. The tasks or the applications of task allocation in MAS include search and rescue missions (SAR) [11]- [14], military op-erations such as attack or surveillance [15]- [18], physical disasters management [11], [12], [19]- [22] where unmanned systems are mainly used, but also crowd-sourcing platforms usage, cloud computing [23]- [28], smart grids, resources allocation in manufacturing [29]- [32] and others.…”