Prof. Ming Zhang from Peking University, Beijing, participated as keynote speaker in the 4th TYAN International Thematic Workshop and 1st African Symposium on Big Data, Analytics and Machine Intelligence for Financial, Health and Environmental Inclusion in Developing Countries, held from 10th – 12th June 2018, in Akure, Nigeria. The participation was realized through COMSATS sponsorship of Prof. Zhang under COMSATS’ sponsorship and fellowship activities to help build capacities of and facilitate exchanges between member countries, in this case, Nigeria and China.
The thematic workshop was jointly organized by TWAS Young Affiliates Network (TYAN) and Federal University of Technology Akure (FUTA), Nigeria, and had representation from 17 developing countries particularly of African region.
Prof. Zhang’s talk focused on ‘Learning the Joint Representation of Heterogeneous Temporal Events for Clinical Endpoint Prediction’ and ‘Knowledge Graph / Representative Learning’. In her talk, Prof. Zhang presented her model for learning the joint representation of heterogeneous temporal events in the records of patients’ history so as to predict the occurrence of a particular disease in the future. She delineated that the model adds a new gate to control the visiting rates of different events which effectively models the irregular patterns of different events and their nonlinear correlations. She stated that experiment results with real-world clinical data on the tasks of predicting death and abnormal lab tests prove the effectiveness of the proposed approach over competitive baselines. Further, in her talk, Prof. Zhang also introduced some joint research work on learning representations of graphs such as LINE, a node representation learning model, as well as LargeVis, a graph visualization model.