2nd edition of the Hackathon for Peace, Justice and Security
Problem: Displacement Tracking Matrix (DTM) collects and produces data and data products for humanitarian response and programming in a variety of crises that drive human displacement across the globe, including generalized violence, conflict, and sudden and slow onset disasters. DTM operations are conducted throughout all crisis stages: emergency, preparedness, transition and recovery. Strong data and analysis means better response. Nearly 4,000 public DTM products (including but not limited to: reports, spread sheets, figures, maps) have been produced to date. To generate relevant and useful products, DTM utilizes multiple different platforms, databases, documents and languages. To maximize the use of versatile resources essential to data collection, analysis, and presentation, DTM is working to expand its ability to detect relevant data and sources across internal and external documents using a combined NLP/Image-recognition tool that can extract relevant data from both DTM products and outside sources, such as media, which make reference to DTM data. Additionally, DTM plans for AI innovatoin to support quality control of DTM data products by reviewing for gaps or inconsistencies and enabling DTM to cross-check internal and external materials.
Outcome: DTM is commited to implementing an effective and predicatable tool for sustaining and enhancing quality data collection and analysis while preventing and mitigating errors in data products. Going forward, we seek a tool and process that utilizes AI to:
-Detect relevant information, including references, figures, populations, and geographical information (among other elements) to recognize material in external documents and platforms that refer to DTM’s data or products.
-Bolster DTM’s capacity to cross-check external citations regarding DTM data and products to information and data in DTM’s publications.
-Identify possible inconsistencies, typos, and grammatical issues that went overlooked in DTM’s drafted documents prior to publication.
DTM datasets can be viewed here.
If you are developing a deep neural network based solution, you can use the free GPUs provided by either Google Colab or Kaggle. Here are the instructions for both:
Go to Google Colab and start a new notebook. You can use the instructions in this notebook to upload your datasets and work on them in Colab. When you need to use GPUs to train your models, switch GPU on under Edit --> Notebook settings. This notebook shows you an example of training a model built in TensorFlow using a GPU.
Go to Kaggle and create a user profile. Then, go to Kernels --> New Kernel. A blank notebook opens up, and you can write your code in here. You will see on the right side of the notebook an option to turn on the GPU. On the top right side of your notebook, you will see a symbol that looks like a cloud with an arrow on it. You can click on this to upload your data to Kaggle.
If you have a Google Cloud, or Amazon Web Services, or Paperspace account, etc, you can use one of these to train your models.