Identifying Uncollected Garbage in Urban Areas using Crowdsourcing and Machine Learning

被引：0

作者：

Singh, Shubhendu ^{[1
]}

Mehta, Kushal Samir ^{[2
]}

Bhattacharya, Nishant ^{[2
]}

Prasad, Jyotsna ^{[2
]}

Lakshmi, Kaala S. ^{[2
]}

Subramaniam, K. V. ^{[2
]}

Sitaram, Dinkar ^{[2
]}

机构：

[1] Janaagraha, Bangalore, Karnataka, India

[2] PES Univ, Comp Sci & Engn, Bangalore, Karnataka, India

来源：

2017 IEEE REGION 10 INTERNATIONAL SYMPOSIUM ON TECHNOLOGIES FOR SMART CITIES (IEEE TENSYMP 2017) | 2017年

关键词：

D O I：

暂无

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

Waste management in Urban Cities of India is a serious concern with growing amounts of uncollected garbage on the streets. In this paper, we present a mobile application based solution to empower the citizens to report instances of uncollected garbage and draw the attention of authorities. Our application has been successfully deployed and has seen more than a million complaints registered across many Indian cities. One of the challenges we have had to address on deployment of the application was the presence of a number of spurious complaints, such complaints result in unnecessary work by the municipal authorities. We tackled the challenge by using machine learning techniques to identify spurious complaints. We have been able to achieve an accuracy of over 85% in segregating the spurious complaints from actual ones by analyzing the image uploaded along with the complaint. The application can be used in real time once the model is trained.

引用

页数：5