Skip to content
SDGs Center UDINUS

Evidence record · UDINUS · SDGs

Performance Evaluation of Transformer and Informer Based Models for Indoor Absolute Humidity Prediction using Outdoor Absolute Humidity

Rektorat UDINUS 2024 Published 2026-10-30

Evidence record · UDINUS · SDGs

Unit
Rektorat UDINUS
Period
2024
Goals
Record
030b8baf-55d3-4a32-96d6-82b7966d40ab
Published
2026-10-30

Universitas Dian Nuswantoro conducted an activity evaluating Transformer and Informer models for predicting indoor absolute humidity from outdoor humidity measurements. The activity addressed the importance of indoor absolute humidity prediction for effective building management, energy efficiency, and occupant comfort. The participants named in the source were Rahim, Muhamad Sharifuddin Abd; Yakub, Fitri; Shaikh Salim, Sheikh Ahmad Zaki; Dhamanti, Inge; Muljono; and Niranjana, G.

The activity used a dataset consisting of outdoor humidity measurements to forecast indoor conditions. UDINUS developed Transformer and Informer models focused on attention mechanisms and sequence generation. Their performance was evaluated using MAPE, MSE, MAE, RMSE, and R-squared. The Transformer model slightly outperformed the Informer model. The reported Transformer results included a MAPE of 3.42% and an R-squared of 0.911, while the Informer results included a MAPE of 4.55% and an R-squared of 0.867.

The source attributes the superior Transformer performance to its enhanced attention mechanism and efficient sequence handling. The activity developed advanced models for accurate indoor humidity prediction, with implications for building management and energy savings. The source also states that future research could explore real-time implementation and application to other environmental parameters. This evidence draft records UDINUS activity in model development and performance evaluation while retaining the scope and findings explicitly reported in the source.

Reference https://www.scopus.com/pages/publications/85213392064?origin=resultslist