A Geocoding Framework Powered by Delivery Data

Abstract

Over the last decade, India has witnessed an explosion in the ecommerce industry. There is increasing adoption of e-commerce in smaller towns and cities over and above the densely populated urban centers. In this paper, we discuss the practical challenges involved with developing high-precision geocoding engines for these geographical regions in India. These challenges motivate the next iteration of our geocoding framework. In particular, we focus on addressing three core areas of improvement: 1) leveraging customer delivery data for geocoding, 2) understanding and solving for the diversity and variations in addresses for these new regions, and 3) overcoming the limited coverage of our reference corpus. To this end, we present GeoCloud. Key contributions of GeoCloud are 1) a training algorithm for learning reference-representations from delivery coordinates and 2) a retrieval algorithm for geocoding new addresses. We perform extensive testing of GeoCloud across India to capture the regional, socio-economical and linguistic diversity of our country. Our evaluation data is sampled from 72 cities and 21 states from the delivery addresses of a large e-commerce platform in India. The results show a significant improvement in precision and recall over the state-of-the-art geocoding system for India, and demonstrate the effectiveness of our intuitive, robust and generic approach. While we have shown the effectiveness of the framework for Indian addresses, we believe the framework can be applied to other countries as well, particularly where addresses are unstructured. To the best of our knowledge, this is the first instance of geocoding by learning reference-representations from large-scale delivery data.

FastAddr

“…Unlike addresses with clear and consistent structures in developed countries such as the U.S. and Japan, in developing counties such as India and China, addresses entered by users are complex and of diverse qualities. For example, low-quality addresses cannot be geocoded by commercial Geocoding services, e.g., Google Maps and Baidu Maps [35,39,41,44]. Moreover, due to rapid development, new addresses and spatial entities are emerging every day.…”

“…As important components in spatio-temporal data, addresses and POIs have received lots of research interests [9,16,35,39,41,44,51]. One of the most popular research areas is next POI recommendation [9,51].…”

“…Admittedly, some works have studied problems related to abnormal address detection. Instead of directly addressing the abnormal address issue, some studies focus on applications related to address and Point-of-Interest (POI), such as coordinates generating for POIs and addresses [39,41], and POI alias discovery [16]. Meanwhile, there are many anomaly detection works in different modalities, e.g., image [15], time series [4,10], graph data [52], and text data [26,32], which have applications such as pickpocket suspects detection [14] and malicious behavior detection on social media [34,50].…”