During my first supervisor meeting, we began to discuss the different planning steps of starting my research process. Some of the questions I prepared for her included: which tools to use to identify factors and variables that cause conflicts, If its better to focus on a particular region while researching. Some of the tips that Morgan supplied me with include: analyze the history of researchers using big data sets to model for violence and conflicts, how long this process takes, what the state of the art for the tools used is and if its feasible to copy these practices as a masters student. I was also informed to make sure that if I wish to use the same data as the researchers that I will be reviewing, that I am able to gain the same access to the data that they are using.

She provided me with an article to read ‘Machine Learning and Conflict Prediction – 2013.’ Through reading this article I discovered that the data sets that the researchers were using were all open to the public. The Armed Conflict Location and Event Dataset (ACLED), is already quite familiar to me as we have previously used it during my intensive ‘Data Science for Society’ to discover possible correlations between indicators and different violent event types. The data set was designed for dissaggregated conflict analysis and crisis mapping. It includes political armed conflicts in over 50 developing countries. Another data set that was mentioned in the article is the Global Data on Events, Location and Tone (GDELT).  This data set could be useful for developing better target classes, especially target classes at a global scope.

It was stated in the article that the forecasting and prediction models that were and will be used should not be the subject of focus. Instead the area of focus should be refining current indicators and discovering new and better ones.

Morgan also wanted me to consider looking into the sociological study of indicators and how over time researchers have tried to understand them. If I were to take this route in my dissertation it would look like this: Doing a sociological study on the different researchers, their techniques and tools used in trying to create a successful conflict prediction model, and then scaling down the different studies and testing out the seperate methodolgies used.

Notes and Useful Material from Supervisor Meeting / Jantz Andersson / Knowledge Integration and Project Planning: Future Governance (2023-24) by is licensed under a