Skip to content
SDGs Center UDINUS

Evidence record · UDINUS · SDGs

An Adaptive Hybrid Algorithm for Time Series Prediction in Healthcare

Rektorat UDINUS 2010 Published 2026-10-30

Evidence record · UDINUS · SDGs

Unit
Rektorat UDINUS
Period
2010
Goals
Record
a3c3ae7f-6c79-489b-abf7-3ea4c09b67ec
Published
2026-10-30

Universitas Dian Nuswantoro conducted an activity on developing an adaptive hybrid algorithm for time series prediction in healthcare, with Purwanto, Chikkannan Eswaran, and Rajasvaran Logeswaran participating in the activity. The activity addressed prediction models based on different concepts and considered the limitations of linear models, including exponential smoothing, linear regression, and autoregressive integrated moving average. The source states that these linear models do not achieve high accuracy rates because they handle nonlinear relationships among data poorly. Neural network models were considered more capable of handling nonlinear relationships. However, healthcare time series data, including Morbidity of Tuberculosis, contain complex linear and nonlinear patterns. Therefore, using only linear models or only neural network models may make it difficult to obtain high prediction accuracy rates.

To address this condition, UDINUS proposed an adaptive hybrid algorithm that combines linear and neural network models. The stated purpose of the algorithm was to achieve the best results for time series prediction in healthcare. The activity also included a comparison of the proposed model with other known models based on accuracy rates. The source does not provide the compared models, the values of the accuracy rates, the specific adaptive procedure, or the results of the comparison. In relation to RAW_GOALS, the activity concerns the development and comparison of a healthcare prediction approach. Institutionally, this activity demonstrates UDINUS involvement in healthcare-related research addressing prediction challenges associated with complex linear and nonlinear time series patterns.

Reference https://www.scopus.com/pages/publications/79952076171?origin=resultslist