Crowd computing
January 19th, 2012 § Leave a Comment
Crowdsourcing is a distributed problem-solving model in which a job is outsourced to a large group of unknown people called the crowd, through an open call. The crowd returns the solutions to the job provider called crowdsourcer, who usually rewards them through a micropayment method. Achieving large-scale distributed computation in this manner is called crowd computing. Tasks, which are very easy and natural for humans to perform but computationally expensive and hard for computers, are especially suited for crowd computing. Crowd computing provides elastic, on-demand, low cost human workforce in the same way as cloud computing provides computation resources.
Thesis
January 17th, 2012 § Leave a Comment
A crowdsourcing framework for software localization
I have started working on this project for my master thesis at Universitat Politecnica de Catalunya in collaboration with CA Labs, Barcelona. Stay tuned for updates
Project description
Software industry is growing rapidly day-by-day and there is a strong competition between the multinational software companies. In order to make their software globally available and to explore new markets, they need to translate their software (user interface, websites, user manuals etc) into several languages. Software localization is not an easy and cheap process. A fast and cheap way of solving this issue is to use automatic Machine Translation (MT) systems. However, the quality of MT is far from satisfactory, especially for less popular languages. Mostly the industrial localization process involves a hybrid approach by combining MT and human translators.
Our goal is to replace the current localization process with a new framework, which solves the software localization process in a cost-efficient and fast way, and provides excellent quality. The benefit of developing such a framework is twofold. Firstly, it will deal with the heterogeneous workload. Secondly, it will enable companies to explore new markets without spending a fortune. We investigate the use to crowdsourcing for software localization process.
