Main Charateristics of MF
 MF based models consider preferences of learners that can be defined by latent factors.

Matrix decomposition becomes an optimization problem with loss functions and constraints.

Latent factors are considered as low dimensional hidden factors for contents and learners.
Figure 2 shows an example of MF where latent factors are low dimensional hidden factors for learners and contents.
Latent factors
Latent factors are considered as low dimensional hidden factors for contents and learners. Figure 2 shows an example of MF where latent factors are low dimensional hidden factors for learners and contents.

Assume there are 4 dimensional latent factors (D or n_factors = 4) for both contents and learners.
ContentC: number 4 indicates for example 4 different features about the content such as: (a) how technicaloriented is it; b) how engaging is it; c) how easy are the examples; d) how recent is it?

LeanerL: number 4 in Learner Latent Factor matrix indicates for example: a) how much do you like the content; b) how much do you like recent contents; c) how technically strong are you; d) how knowledgeable are you about the content.
For more details, please contact me here.
Date of last modification: 2021