“…The focus of Part I has been on defining graph properties through the mathematical formalism of linear algebra, while Part II introduced graph counterparts of several important standard data analytics algorithms, again for a given graph. However, in many modern applications, graph topology is not known a priori [1,2,3,4,5,6,7,8,9,10,11,12,13,14], and the focus of this part is therefore on simultaneous estimation of data on a graph and the underlying graph topology. Without loss of generality, it is convenient to assume that the vertices are given, while the edges and their associated weights are part of the solution to the problem considered and need to be estimated from the vertex geometry and/or the observed data [15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31].…”