Introduction

Source code and data for open source data science for social good. This is a data science portfolio.

List of projects

1) university_sexcrimes

Analysis of data on sex crimes in US university campuses.

2) heartdiseaserisk_prediction

Predicting heart disease risk from open data.

3) cancermortalityprediction

Predicting cancer survival using logistic regression from open data.

4) predictingnewspopularity

Predicting popularity of news articles from open data.

5) opensourcemappingproject

Open source mapping project. 

6) astroinformatics

Analysis of astronomy data using machine learning techniques.

7) scientific_collaboration

Project to analyze planetary scale scientific collaboration data.

8) accident_prediction

Road accident forecasting and data exploration project.

Interactive website using shiny at:

https://neelsoumya.shinyapps.io/accident_prediction/

9) patternsincrime

Predicting patterns of crime using data science. Larger cities have disproportionately more crime per capita compared to smaller cities (super-linear scaling of crime). We used techniques from dynamical systems and complex systems to explain the super-linear scaling of crime in cities and other socio-technological systems

10) spam_classification

Building an SVM based spam classifier trained on data from the UCI repository

11) breastcancerprediction

Downloads data from the UCI machine learning repository to make predictions
for breast cancer. A few features turn out to be really important for prediction like epithelial cell size. This uses a random forest.

12) fundingtrendsscience

Project to analyze data on funding trends in biomedical science.

13) infectiousdiseaseprediction

Project to analyze data on emerging infectious diseases.

14) forecasting_imports

Project to forecast imports and model supply chains.  

15) deeplearningbasic

Basic deep learning model using keras for prediction.

16) ai_healthcare

Machine learning and AI applied to healthcare.

17) aisocialgood

Machine learning, data science and AI for social good. 

18) aibigdatabiology

Machine learning and bioinformatics for big data in biology. 

19) browserbaseddata_science

Browser based data science for democratic access to data science tools. 

20) clinical_informatics

Open source privacy-preserving clinical informatics.

21) policypapergeneral_public

Policy paper for general public on Ethical Artificial Intelligence (EAI) for social good.

22) nlp

Resources, code and data for natural language processing.

23) selforganisingmapwinedataset

A self organising map (SOM) on the UCI wine dataset using the Orange data science tool. 

24) LLMs

Hackathons and resources for large-language models (LLMs).

25) outreach

Outreach for machine learning and AI for general public

26) teaching_resources

Teaching resources for machine learning, data science and AI for a general audience

What is this repository for?

• Quick summary

  • Open source code and data for open source data science.

Citation

• If you use this code, please cite the paper and code

  • Citizen Data Science for Social Good: Case Studies and Vignettes from Recent Projects https://doi.org/10.13140/RG.2.1.1846.6002
  • Citizen Data Science for Social Good in Complex Systems, Interdisciplinary Description of Complex Systems, 16(1):88-91, 2018 http://indecs.eu/index.php?s=x&y=2018&p=88-91
    
  • Banerjee, Soumya. (2017, September 3). Citizen Data Science for Social Good: Case Studies and Vignettes from Recent Projects (Supplementary Resources). Zenodo. http://doi.org/10.5281/zenodo.883783

  ](https://doi.org/10.5281/zenodo.883783)

• These projects are an example of my approach to data science for good. I work very closely with domain experts and stakeholders and use computational tools for good. I outline my design and work philosophy below.

  • 

Installation

Install R, R Studio, MATLAB and Python

Install R

https://www.r-project.org/

and R Studio

https://www.rstudio.com/products/rstudio/download/preview/

r source("https://raw.githubusercontent.com/neelsoumya/rlib/master/INSTALL_MANY_MODULES.R")

Install Python dependencies as follows:

r pip3 install -r requirements.txt

Contact

 Soumya Banerjee

 https://sites.google.com/site/neelsoumya/

 sb2333@cam.ac.uk