Markov Decision Theory-Based Crowdsourcing Software Process Model

Markov Decision Theory-Based Crowdsourcing Software Process Model

Kamalendu Pal (City, University of London, UK)
DOI: 10.4018/978-1-5225-9659-2.ch001

Abstract

The word crowdsourcing, a compound contraction of crowd and outsourcing, was introduced by Jeff Howe in order to define outsourcing to the crowd. It is a sourcing model in which individuals or organizations obtain goods and services. These services include ideas and development of software or hardware, or any other business-task from a large, relatively open and often rapidly-evolving group of internet users; it divides work between participants to achieve a cumulative result. It has been used for completing various human intelligence tasks in the past, and this is an emerging form of outsourcing software development as it has the potential to significantly reduce the implementation cost. This chapter analyses the process of software development at a crowdsourced platform. The work analyses and identifies the phase wise deliverables in a competitive software development problem. It also proposes the use of Markov decision theory to model the dynamics of the development processes of a software by using a simulated example.
Chapter Preview
Top

Introduction

Crowdsourcing is the Information Technology (IT) mediated engagement of crowds for the purposes of problem-solving, task completion, idea generation and production (Howe, 2006; Howe, 2008; Brabham, 2008). The latest breakthroughs in Information and Communication Technologies (ICT) have ushered a new dawn for researchers to design innovative crowdsourcing systems that can harness Human Intelligence Tasks (HITs) of online communities. The prime aim of crowdsourcing is to facilitate the wisdom of crowds. The theory suggests that the average response of many people, even amateurs, to a question is frequently more accurate than the view of a few experts. In this respect, a community of individuals with common interests and facing the same tasks can deliver better products and solutions than experts alone in the field. Information systems scholars Jean-Fabrice Lebraty and Katia Lobre-Lebraty confirmed that the “diversity and impudence of the members of a crowd” is a value addition to crowdsourcing operations (Lebraty & Lobre-Lebraty, 2013).

Therefore, the advantages of crowdsourcing lie mainly in the innovative ideas and problem-solving capacity that the diverse contributors – which may consist of experts and interested amateurs – can provide. The crowd can provide expert and faster solution to an existing problem. Depending on the challenge at hand, the solution provided may also prove innovative. In this way, crowdsourcing has emerged as a new labour pool for a variety of tasks, ranging from micro-tasks on Amazon Mechanical Turk (mTurk) to big innovation contests conducted by Netflix and Innocentive. Amazon mTurk today dominates the market for crowdsourcing small task that would be too repetitive and too tedious for an individual to accomplish. Amazon mTurk established a marketplace where requesters can post tasks and workers complete them for relatively small amount of money. Image tagging, document labeling, characterizing data, transcribing spoken languages, or creating data visualizations, are all tasks that are now routinely being completed online using the Amazon mTurk marketplace, providing higher speed of completion and lower price than in-house solutions.

Competitive crowdsourcing is reward based and has been used for variety of tasks from design of T-Shirts to research and development of pharmaceuticals and very recently for developing software (Howe, 2008; Lakhani & Lonstein, 2011; Stol & Fitzgerald, 2014).The mTurk is one of the best-known crowdsourcing platforms where HITs or microtasks are performed by thousands of workers (Ipeirotis, 2009).

There are different types of crowdsourcing platforms, such as virtual labour markets (VLMs), tournament crowdsourcing (TC) and open collaboration (OC), which each have different roles and characteristics (Estelles-Arolas & Gonzalez-Ladron-de-Guevara, 2012; Prpic, Taeihagh & Melton, 2014). Along with the growth of crowdsourcing, crowdsourcing platforms are very important to mediate the transactions. At the same time, IT-mediated platforms improve efficiency and decrease transaction costs and information asymmetry. However, these platforms are domain specific.

Key Terms in this Chapter

Human Intelligence Tasks: In crowdsourcing business model, employers post jobs known as Human Intelligence Tasks (HITs), such as identifying specific content in an image or video.

Crowdsourcing Software Engineering: Crowdsourcing software engineering derives from crowdsourcing. Using an open call, it recruits global online labour to work on different types of software engineering works, such as requirement elicitation, design, coding and testing.

Markov Decision Theory: In practice, decision is often made without a precise knowledge of their impact on future behaviour of systems under consideration. The field of Markov Decision Theory has developed a versatile approach to study and optimize the behaviour of random processes by taking appropriate actions that influence future evolution.

Software Process Model: In software engineering, a software process model is the mechanism of dividing software development work into distinct phases to improve design, product management, and project management. It is also known as a software development life cycle. The methodology may include the pre-definition of specific deliverables and artifacts that are created and completed by a project team to develop or maintain an application.

Crowdsourcing: Crowdsourcing is the Information Technology mediated engagement of crowds for the purposes of problem-solving, task completion, idea generation, and production.

Complete Chapter List

Search this Book:
Reset