Project Description
Indicia is experienced in using data science and machine learning tools to understand patterns in consumer/user behavior. Indicia researcher Stephen Paff, a data scientist and ethnographer, specializes in developing custom-tailored data models to break down consumer behavior. For the California Cybernetic project, he developed decision trees and random forests machine learning algorithms to understand how people consume electricity. Machine learning algorithms provide a way to program custom-built data modeling for the needs of a specific ethnographic project.
Ethnographic research specializes in understanding specific contexts and the perspectives of those in a particular setting or settings. Because machine learning algorithms on some level adapt to specific data, we feel that they provide a potentially powerful tool for unique and contextual modeling fitted to the particularities of a context, uniquely useful for ethnographic research. Like techniques in traditional ethnographies, some machine learning approaches are buttom-down, local, and interpretive. Thus, we hope to use machine learning algorithms when applicable within and alongside our ethnographic projects as a quantitative complement to our qualitative research techniques.
In the sixth and final phase of the California Cybernetic project, Indicia affiliate, Stephen Paff, developed decision trees and random forests to model how individuals' emotional connections to technology influences their home energy consumption. These provided a set of criteria for classifying people emotional connections that is simultaneously visually insightful for humans, easily programmable computationally, and tested easily demographically. This project was an example of how to integrate data science and machine learning techniques into an ethnographic project.