![](https://db.cs.pitt.edu/group/wp-content/uploads/2024/02/Brian-T-Nixon-IPIN-2023.jpg)
- by ADMT
- September 28, 2023
Kudos to Anna Baskin and Brian T. Nixon for collaborating and presenting the paper, titled “RETSINA: Reproducibility and Experimentation Testbed for Signal-Strength Indoor Near Analysis” at the International Conference on Indoor Positioning and Indoor Navigation (IPIN2023) in Germany. For further details, please refer to the paper under the ADMT Lab’s Publications.
![The pipeline for RETSINA which consists of the following components (from left to right): the input data and input parameter which modify different modules of the pipeline as well as the number of processes to use, the preprocessing component which filters out dataset instances based on provided distance thresholds, the feature selection component which uses the mRMR algorithm to select the least redundant features, the data sampling component which sample each target class to help with data imbalance, the ensemble estimators which contain the collection of ML models for making predictions, and the output from the pipeline which includes the ensemble model, statistics such as accuracy, precision, etc., and the features that were selected.](https://db.cs.pitt.edu/group/wp-content/uploads/2024/02/Retsina-Pipeline-IPIN-2023-1024x180.png)