Ernesto Dal Bó, Professor

Closed (1) Website Design for the Atlas of Archaeological Sites

Closed. This professor is continuing with Spring 2018 apprentices on this project; no new apprentices needed for Fall 2018.

In this project, economists and archaeologists are working together to build an atlas of past civilizations. We are compiling information from thousands of archaeological sites into a single database, which allows us to study why some societies flourished while others did not.

To make the database accessible to the research community and to the public, we are creating an online platform with the help of an undergraduate research assistant. Tasks include elements of website design, online database maintenance, logo design, and mapping geographic data. It is a great opportunity for a student to work on an interdisciplinary project with faculty and showcase a wide variety of web design skills. Students learn how to integrate various components into a professional product at the intersection of academia and science communication.

Day-to-day supervisor for this project: David Schönholzer, Ph.D. candidate

Qualifications: Required: at least one of the following - web design skills, writing and communication skills, knowledge of geo-coding software (e.g., arcGIS)

Weekly Hours: to be negotiated

Off-Campus Research Site: Haas School of Business

Related website: http://faculty.haas.berkeley.edu/dalbo/

Closed (2) Data Development for Atlas of Archaeological Sites

Applications for Fall 2018 are now closed for this project.

In this project, economists and archaeologists are working together to build an atlas of past civilizations. We are compiling information from thousands of archaeological sites into a single database, which allows us to study why some societies flourished while others did not. The approach is to apply machine learning tools of natural language recognition to machine-read thousands of archaeological journal articles and obtain the information that is needed. In the current phase, we need to verify details of archaeological sites that have been researched through machine learning already. This verification will involve two steps. (i) The first will rely crucially on human URAPs consulting archaeological journal articles to determine key features of sites. (ii) The second step will be to build algorithms to compare the coding done by humans vs that done by our machine learning programs. The Fall 2018 semester will cover all of the first step and, it is hoped, a substantial segment of the second.

Our diverse team of undergraduate research assistants includes students from various social science disciplines working under the guidance of economics graduate students and faculty. The main task in step (i) consists of investigating sources on archaeological sites by consulting specialized literature. In past semesters, students working in similar data development positions have learned about the archaeology and history of societies from across the globe, expanding into a wide set of social scientific skills under close supervision of graduate students, while working in a diverse team of young researchers. Students linked to step (ii) will get to learn about natural language processing and the many steps involved in machine-reading and interpreting large quantities of text.

Day-to-day supervisor for this project: Mehmet Seflek, Graduate Student

Qualifications: The data development aspect (step i) requires self-motivation, curiosity about human populations, and attention to detail. For the machine learning aspect (step ii), expertise with Python will be extremely valuable.

Weekly Hours: 6-8 hrs

Related website: http://faculty.haas.berkeley.edu/dalbo/