Mobile phones and sensors have become very useful to understand and analyze human lifestyle because of the huge amount of data they collect every second. This triggered the idea of taking advantage of such technologies to predict miscarriages and help pregnant women react in advance to a probable miscarriage. To achieve this, we propose to combine benefits and advantages of both machine learning and Big Data tools applied to smartphone/sensors real time generated data. Kmeans clustering algorithm is used for miscarriage prediction and predicted clusters (partitions) are transmitted to the pregnant woman in her front-end user interface in the mobile application, so that she can make quick decisions in case of miscarriage or probable miscarriage. We used real-world data to validate our system and assess its performance and effectiveness. All data management and processing tasks are conducted over Apache Spark Databricks. Peer-review under responsibility of the Conference Program Chairs.
Asri, H., Mousannif, H., & Moatassime, H. A. (2017). Real-time Miscarriage Prediction with SPARK. In Procedia Computer Science (Vol. 113, pp. 423–428). Elsevier B.V. https://doi.org/10.1016/j.procs.2017.08.272