In order to join virtual sessions, you must be registered and logged-in(Were you registered for the in-person meeting in Denver? if yes, just log in.) 
Note: All session times are in Mountain Daylight Time.

A Review on Deep Learning Applications in Geospatial Social Sciences

Authors: Sakib Hasan*, University of Wisconsin - Milwaukee, Zengwang Xu, University of Winconsin - Milwaukee
Topics: Spatial Analysis & Modeling, Social Geography, Cyberinfrastructure
Keywords: Deep learning, Geospatial research, Neural Network
Session Type: Paper
Presentation File: No File Uploaded


Deep learning, as one of the modern marvels of machine learning techniques, has been used in applications like image analysis, speech recognition and text understanding. It can process information of large volume and large variety and very often produces results with great accuracy. Deep learning is based on Artificial Neural Network of multiple layers and possibly convoluted structures, and it extracts information from unsupervised dataset through complex models unlike traditional probabilistic machine learning techniques. With the help of an unsupervised pre-training and a supervised fine-tuning strategy, the deep learning models can learn hierarchical features and representations of big data in deep architectures for the tasks of classification and recognition. We review the existing literature and techniques on the applications, frameworks, advantages and challenges of deep learning, especially when it is applied in geospatial research.

Abstract Information

This abstract is already part of a session. View the session here.

To access contact information login