1: \begin{abstract}

2: In this paper, we consider the symmetric multi-type non-negative matrix tri-factorization problem (SNMTF), which attempts to    factorize several symmetric non-negative matrices simultaneously. This can be considered as a  generalization  of the classical non-negative matrix tri-factorization problem and includes a non-convex objective function which is a multivariate sixth degree polynomial and a has convex feasibility set. It has a special importance in data science, since it serves as a mathematical model for the fusion of different data sources in data clustering.

3: 

4: We develop four methods to solve the SNMTF. They are based on four theoretical approaches known from the literature: the fixed point method (FPM), the block-coordinate descent with projected gradient (BCD), the gradient method with exact line search (GM-ELS) and the adaptive moment estimation method (ADAM). For each of these methods we offer  a software implementation: for the former two methods we use Matlab and for the latter Python with the TensorFlow library.

5: 

6: We test these methods on three data-sets: the synthetic data-set we  generated, while the others represent real-life similarities between different objects.

7: 

8: Extensive numerical results show that with sufficient  computing time  all four methods perform satisfactorily and ADAM  most often yields the best mean square error ($\MSE$). However, if the computation time is limited,  FPM gives the best $\MSE$ because it shows the fastest convergence at the beginning.

9: 

10: All data-sets and codes are publicly available on our GitLab profile. % \cite{git}.

11: 

12: \keywords{Non-negative matrix factorization \and Fixed point method \and Block coordinate descent \and Projected gradient method \and ADAM}

13: 

14: \end{abstract}

15: