Monotonicity Detection and Enforcement in Longitudinal Classification

Ovchinnik, Sergey; Otero, Fernando E. B.; Freitas, Alex A.

doi:10.1007/978-3-030-34885-4_5

Cited by 1 publication

(2 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A full description of attributes used and their meaning can be found in a related study that previously used the same data preparation techniques in the context of automatic feature selection [12]. The same dataset has also been previously used to evaluate other longitudinal data mining approaches [11].…”

Section: Experimental Methodologymentioning

confidence: 99%

“…In [11], the XGBoost algorithm was used to learn boosted decision tree models from the same longitudinal datasets used in this current paper. That study focused on improving model acceptability by using monotonicity constraints to produce monotonic classification models, instead of improving the longitudinal awareness of the models.…”

Section: The Proposed Longitudinal Classification Algorithmmentioning

confidence: 99%

See 1 more Smart Citation

Nested trees for longitudinal classification

Ovchinnik

Otero

Freitas

2022

Proceedings of the 37th ACM/SIGAPP Symposium on Applied Computing

Self Cite

View full text Add to dashboard Cite

Longitudinal datasets contain repeated measurements of the same variables at different points in time. Longitudinal data mining algorithms aim to utilize such datasets to extract interesting knowledge and produce useful models. Many existing longitudinal classification methods either dismiss the longitudinal aspect of the data during model construction or produce complex models that are scarcely interpretable. We propose a new longitudinal classification algorithm based on decision trees, named Nested Trees. It utilizes a unique longitudinal model construction method that is fully aware of the longitudinal aspect of the predictive attributes (variables) and constructs tree nodes that make decisions based on a longitudinal attribute as a whole, considering measurements of that attribute across multiple time points. The algorithm was evaluated using 10 classification tasks based on the English Longitudinal Study of Ageing (ELSA) data. CCS CONCEPTS• Computing methodologies → Classification and regression trees; • Information systems → Data mining.

show abstract

Section: Experimental Methodologymentioning

confidence: 99%

Section: The Proposed Longitudinal Classification Algorithmmentioning

confidence: 99%