Back

Brain Activation-based Filtering: a Scholarly Journal Recommendation Model for Human Neuroimaging Studies

Kim, J.

2020-10-06 neuroscience
10.1101/2020.10.06.327684 bioRxiv
Show abstract

The development of noninvasive neuroimaging techniques such as functional magnetic resonance imaging was followed by a large volume of human neuroimaging studies of mental processes, mechanisms, and diseases. Due to the high volume of studies and the large number of journals, it is increasingly challenging for neuroscientists to review existing scholarly journals and find the most suitable journal to publish their studies. Therefore, this paper proposes a scholarly journal recommendation model for human neuroimaging studies called brain activation-based filtering (BAF). Based on the collective matrix factorization technique, BAF recommends journals relevant to the activated brain regions that are described in a given neuroimaging study. For instance, if social brain regions such as the dorsomedial prefrontal cortex, precuneus, and temporoparietal junction are activated in a study, BAF recommends relevant social neuroscience journals (e.g., Social Cognitive and Affective Neuroscience). Five-fold cross-validation shows that BAF predicts journals with a reliable area under the curve score of 0.855. Furthermore, an interactive Google Colab notebook is offered to recommend relevant journals for a novel human neuroimaging study (https://github.com/JunsolKim/brain-activation-based-filtering).

Matching journals

The top 9 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.