2026-05-31T02:20:01.602Z http://ezaposleni.singidunum.ac.rs/rest/sciNaucniRezultati/oai

ezaposleni.singidunum.ac.rs/rest/sciNaucniRezultati/oai:3:11935 2026-05-27T11:54:26Z 3

Introduction to Intricate Artificial Psychology with Python 2026 http://ezaposleni.singidunum.ac.rs/rest/sciNaucniRezultati/oai/record/3/11935 https://www.sciencedirect.com/science/chapter/edited-volume/abs/pii/B9780443302480000152?via%3Dihub N. Kovač T. Bezdan H. Farahani M. Simeunović P. Watson In this chapter, we will explore different prediction methods, starting with traditional approaches that use hand-crafted features and statistical models and progressing to modern graph-based techniques and advanced machine learning algorithms. These methods enable predictions at the individual level and extend to complicated relationships such as those in social networks or brain connectivity studies. Emphasis will be placed on understanding the mathematical foundations behind these methods, while also providing practical applications and insights into their use in psychological research. We focus particularly on the use of nodes in a network predicting links between them, mapping nodes to lower-dimensional spaces via embedding, and identifying and interpreting nodal characteristics such as betweenness, degree, centrality, and clustering coefficients. The connectivity between nodes is further illustrated in graphical neural networks (GNNs) and graphical attention networks. Further enhancements to improve the predictive performance of GNNs are also presented. Large-scale node prediction and temporal graph networks are also considered. We conclude by describing practical aspects to node prediction in small datasets. bookPart Academic Press 255 287 https://doi.org/10.1016/B978-0-443-30248-0.00015-2 Academic Press