04-637   Mobile Big Data Analytics and Management

Course description

New sources of digital data from phones, mobile applications, social media companies, financial services providers, satellites, and other digital sensors are opening up new opportunities for societal-scale data analysis and inference. This course will define “mobile big data” broadly to introduce students to methods for storing and analyzing large digital traces, and to explore the potential and limitations of these data to learn about social and environmental phenomena. Much of the course will focus on equipping students with practical skills to work with mobile big data sources, including machine learning and data mining techniques specific to mobile big data’s unique characteristics (multi-dimensional, personalized, spatiotemporal, sparse, and real-time). The course will also cover applications of mobile big data, particularly in topics related to sustainable development in Africa, as well as privacy and security challenges to working with mobile big data.

Outcomes

At the end of this course, students will be able to:

• Distinguish the characteristics of different types of mobile big data, including mobile phone metadata, internet traces, social media data, satellite data, and in situ sensing
• Apply appropriate data management techniques for mobile big data, including SQL and NoSQL Prepare and clean mobile big data for analysis
• Draw on a set of data analysis techniques, including supervised and unsupervised machine learning, natural language processing, geospatial techniques, and time series methods to analyze mobile big data
• Discuss the limitations of mobile big data for societal-scale analysis, and identify use cases where mobile big data may be more or less suited to answering research questions
• Assess the risks of mobile big data analysis, including issues related to privacy, security, fairness, and interpretability, and apply safeguards to mitigate these risks

Content details

This course covers:

• Mobile big data types and sources: Mobile phone metadata, smartphone application data, social media data, Internet of Things (IoT) data, remote sensing data
• Application of mobile big data analysis: Poverty measurement, targeting humanitarian aid, epidemic prediction, digital gender gaps, political polarization, smart cities, flood prediction, precision agriculture Mobile big data management and processing tools: SQL, NoSQL, distributed processing
• Mobile big data analysis techniques: Data mining, machine learning, time series methods, NLP, network analysis, crowdsourcing
• Risks and limitations of working with mobile big data, including privacy, fairness, representativity, and transparency

Prerequisites

None

Syllabus

https://emilylaiken.github.io/Mobile%20Big%20Data%20Syllabus.pdf

Faculty