Nowadays, environmental problems are becoming more and more serious and urgent to deal with. We cannot deny the impact of food and agriculture in general, on the environment. If we consider what we eat and how we grow it, we would find extensive damage to the environment (green gas emissions, soil depletion etc.) and also the wildlife (due to pesticides, fertilizers etc.). We are looking to use the data provided by FAOSTAT giving access to over 3 million time-series and cross sectional data relating to food and agriculture all over the world and try to generate insights and stories on the evolution of different socio-environmental factors such as the correlation between greenhouse gas emissions and agricultural growth for example. Through this work, we hope to gain a deeper insight on the evolution and the environmental impact of agriculture and food.
- Could we find and support a correlation between greenhouse gas emissions and agricultural growth? Could we also assess the influence of the type of culture on the emissions?
- Based on this, is it possible to make predictions of greenhouse gas emissions by extrapolating the agricultural growth and land usage?
- Is it possible to rank some patterns of land usage with social and environmental factors such as employment, life satisfaction (if data are presents), income inequalities (if data are presents), emissions and finally soil quality and sustainability?
- Could we assign a kind of “eco” score to each recipe from the Stanford cooking recipes based on indicators derived from the main dataset?
- Is the temporal evolution of agricultural land always correlated with the temporal evolution of the quantity of fertilizers used (and waste emission) or there is some point in time where this trend slows or reverses for some countries?
- What is the ratio of country support/cost vs crop production in developed countries? Could we find other metrics to explain this support? Should we stop supporting farmers in Switzerland?
- Global Food & Agriculture Statistics (https://www.kaggle.com/unitednations/global-food-agriculture-statistics/)
- Cooking recipes (http://infolab.stanford.edu/~west1/from-cookies-to-cooks/recipePages.zip/)
1.11.2019
- Define the required datasets
- Subdivide the tasks of cleaning these datasets among the members of the group.
7.11.2019
- Refine the research questions based on first exploration of the data.
- Generate first meaningful general visualization of the data (general descriptive statistics).
- Rebalance the workload among the members of the group.
14.11.2019
- For each research question, develop a strategy to answer it and prepare a first iteration of the code required to extract the information and to present it.
- Meeting to discuss first results, exchange feedback and ideas.
- Define a complete data story to present our research and our results in a clean and ordered manner.
- Update the plan and the balance of the work according to the expected difficulties and the schedule of the group.
21.11.2019
- Second iteration of the code and the results.
- Meeting to discuss the results, exchange feedback and ideas.
- Update of the data story based on the last conclusions.
- First draft of the final notebook.
25.11.2019
- Last iteration of the code and the results.
- Update of the notebook based on the individual and collective work.
- Review of the results and conclusions.
28.11.2019
- Last review of the notebook.
- Final commit for milestone 2.
- Can we decide to reduce our analysis to a few countries if we lack of data ?
- Can we update the dataset in the process if we find another source of informations ?
- Could we explore a new question (not previously mentioned in the research questions) if we stumble upon a new interesting subject in the data-exploration phase ?