Zur Kurzanzeige

dc.rights.licenseAll rights reserveden_US
dc.contributor.advisorDuffany, Jeffrey
dc.contributor.authorVarela Rosa, Carla Andrea
dc.date.accessioned2022-03-29T18:50:36Z
dc.date.available2022-03-29T18:50:36Z
dc.date.issued2021
dc.identifier.citationVarela Rosa, C. A. (2021). he Impact of Segmentation and Overlapping inFeature Extraction for Biometric Human Authentication Systems [Unpublished manuscript]. Graduate School, Polytechnic University of Puerto Rico.en_US
dc.identifier.urihttp://hdl.handle.net/20.500.12475/1418
dc.descriptionDesign Project Article of the Graduate Programs at Polytechnic University of Puerto Ricoen_US
dc.description.abstractOne of the arduous challenges in Machine Learning is how to extract features withenough information that will simplifythe learning process of classificationmodels;therefore,leading to better predictionsand human interpretations. We investigated the impact of segmentationand overlapping techniques used to extract features from accelerometer data to optimize the performance of Machine Learning models designed for Biometric User Authentication via walking patterns. Results showed that bigger segmentations were beneficial to the individual performance of the features and detrimental for systems fed with a set of features. Also, there was no evidence found supporting the increase in the overall performance of the system by usingthe method of overlapping. Finally, via a brute-force feature selection algorithm, we achieved a 71% classification accuracy (with 10/34 features) vs. 64% (with 34 features), regardless of the system’sconfiguration meaning that key features hold more weight than mere segmentationand overlapping methods. KeyTerms⎯Acceleration, Biometric Human Authentication, Feature Extraction, Supervised Learning.en_US
dc.language.isoenen_US
dc.publisherPolytechnic University of Puerto Ricoen_US
dc.relation.ispartofComputer Science Program;
dc.relation.ispartofseriesWinter-2021;
dc.relation.haspartSan Juanen_US
dc.subject.lcshPolytechnic University of Puerto Rico--Graduate students--Researchen_US
dc.subject.lcshPolytechnic University of Puerto Rico--Graduate students--Postersen_US
dc.subject.lcshArtificial intelligence
dc.subject.lcshBiometric identification
dc.titleThe Impact of Segmentation and Overlapping in Feature Extraction for Biometric Human Authentication Systemsen_US
dc.typeArticleen_US
dc.rights.holderPolytechnic University of Puerto Rico, Graduate Schoolen_US


Dateien zu dieser Ressource

Thumbnail
Thumbnail

Das Dokument erscheint in:

Zur Kurzanzeige