Python in Plain English

New Python content every day. Follow to join our 3.5M+ monthly readers.

Follow publication

Collaborative Data Science

Digital Platform Innovations for Development Impacts

Capability-Complexity-Cost assessment of three Cloud-based Jupyter Notebook deployment options

Parvathy Krishnan
Python in Plain English
13 min readDec 6, 2021

--

Photo by Shahadat Rahman on Unsplash

This work was co-authored with Kai Kaiser. Also, special thanks to Pedro Camargo for his valuable feedback. All errors and omissions are those of the author(s).

A rapidly expanding range of digital data and methods now provide significant opportunities to strengthen descriptive and prescriptive insights to better address the challenges and impacts of public policy in developing countries. For example, in some of our recent work we — a combined team of public sector governance specialists, sector specialists, and data scientists — have been able to assess the level of basic health care access in Timor-Leste at a high level of spatial granularity.

The effort contributed to a Public Expenditure Review (PER), a core World Bank diagnostic report designed to help a wide range of developing countries assess how public resources are used relative to key Sustainable Development Goals (SDGs). Meeting this objective meant integrating key fit-for-purpose data, together with cloud-based analytics and optimization capabilities. By validating baseline geospatial data, and applying optimization algorithms, we were able to show where clinic and hospital infrastructure facility network upgrades could be best targeted to improve access with different budget constraints.

Figure 1. The 4As Journey to Data-Driven Decision Support for the Public Sector (Source: Adapted from Vietnam Big Data Observatory)

Across any number of problem statements for better data-driven decision support, impacts depend on successfully traversing the journey of data access, analysis, application, and adoption. But in practice, the analytic workflows and data sources to deliver insight value in a timely, transparent, and progressive fashion for developing country settings typically remain fragmented. Policymakers in government, topic experts, and data scientists are often on very different pages in terms of core skills, distractions, and how to best collaborate for impact and sustainability.

The right choice of

--

--

Published in Python in Plain English

New Python content every day. Follow to join our 3.5M+ monthly readers.

Written by Parvathy Krishnan

Lead Data Scientist | CTO at Analytics for a Better World | Public Sector Consultant

Responses (1)

Write a response