Crowdsourcing for Urban Laborers and Time Optimization

Abstract

The job opportunities in metro cities are immense but finding people with required skillset becomes difficult. Similarly, the worker’s migrating to these cities finds it difficult to get jobs and spend a considerable amount from their wages to reach the job location. Our system helps to ease these difficulties by providing jobs to the workers using recommendation systems and finding cost-efficient routes from the worker’s present location to the job site. The recommender system exploits the worker’s job history to provide recommendations by using the concept of real Boltzmann machine along with matrix factorization, and the location services take into consideration of various local transport services like shared-autos and taxis apart from trains and buses to suggest a cost-efficient path to ease the worker’s travel in a new city. The deep learning approach further enhances the recommender system efficiency by approximating rank for each job before recommending them to the worker.