دانلود مقاله ترجمه شده فعال سازی شهرهای هوشمند شناختی با استفاده از کلان داده ها و یادگیری ماشین: رویکردها و چالش ها

Abstract

The development of smart cities and their fast-paced deployment  is  resulting  in  the  generation  of  large  quantities  of data at unprecedented rates. Unfortunately, most of the generated data  is  wasted  without  extracting  potentially  useful  information and  knowledge  because  of  the  lack  of  established  mechanism sand  standards  that  benefit  from  the  availability  of  such  data. Moreover,  the  high  dynamical  nature  of  smart  cities  calls  for new generation of machine learning approaches that are flexible and  adaptable  to  cope  with  the  dynamicity  of  data  to  perform analytics and learn from real-time data. In this article, we shed the light on the challenge of under utilizing the big data generated by smart cities from a machine learning perspective. Especially, we present the phenomenon of wasting unlabeled data. We argue that  semi-supervision  is  a must for  smart  city  to  address  this challenge. We also propose a three-level learning framework for smart  cities  that  matches  the  hierarchical  nature  of  big  data generated  by  smart  cities  with  a  goal  of  providing  different levels  of  knowledge  abstractions.  The  proposed  framework  is scalable to meet the needs of smart city services. Fundamentally, the framework benefits from semi-supervised deep reinforcement learning where a small amount of data that has users’ feedback serves  as  labeled  data  while  a  larger  amount  is  without  such users’ feedback serves as unlabeled data. The framework utilizes a  mix  of  labeled  and  unlabeled  data  to  converge  toward  better control   policies   instead   of   wasting   the   unlabeled   data.   This paper  also  explores  how  deep  reinforcement  learning  and  its shift  toward  semi-supervision  can  handle  the  cognitive  side  of smart city services and improve their performance by providing several use cases spanning the different domains of smart cities. We also highlight several challenges as well as promising future research directions for incorporating machine learning and high-level  intelligence  into  smart  city  services.