000			03476nas a2200337 a 4500
005			20260818145444.0
008			260219s20259999th u ms t 000 eng d
035			_a.b12478039
099		9	_aAIT Thesis no.CS-25-01
100	1		_aNayeem, Jannatun
245	1	0	_aLeveraging phonological clustering for word-level Bangla Sign language recognition
260			_aPathum Thani, Thailand : _bAsian Institute of Technology, _c2025
300			_a63 leaves : _bill.+ _e1 online resource
490	1		_aThesis ; _vno.CS-25-01
500			_aA thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in Computer Science
502			_aThesis (M. Sc.) - Asian Institute of Technology, 2025
520			_aBangla Sign Language (BdSL) recognition presents multifaceted challenges due to signer diversity and spatiotemporal variability. While flat classification pipelines are widely used, they often overlook the underlying phonological relationships among signs that can inform more structured and accurate recognition. To address this gap, we propose a hierarchical recognition framework that integrates phonological clustering into a Bidirectional Long Short-Term Memory (Bi-LSTM)-based sequence modeling pipeline. First, baseline classification{u2014}referring to a flat, non-clustered recognition approach{u2014}is performed us ing five Bi-LSTM configurations of increasing complexity to assess the trade-off between accuracy and model size. The resulting confusion matrices are analyzed to identify sign pairs with high misclassification rates, revealing underlying phonological similarities. Based on this analysis, a confusion-matrix driven clustering strategy is employed to group visually and phonologically similar signs. Cluster-specific feature engineering is then applied, and the same Bi-LSTM architecture is retrained separately within each cluster. Experiments are conducted on a curated 50-class subset of the SignBD-Word dataset. In the baseline setting, the most complex model (Bi-LSTM-1) achieves 91.10% accuracy with 2.6 million parameters. With the proposed confusion-matrix driven clustering architecture, four out of six clusters employing the lightweight Bi-LSTM classifiers outperform the baseline model, reaching up to 94.58% accuracy. Remarkably, the lightweight Bi-LSTM-4 model{u2014}with only 373K parameters (14.4% of Bi LSTM-1){u2014}achieves a weighted average accuracy of 92.60% on cluster-level classification, surpassing the baseline by1.5percentagepoints. Thisreflectsan85.6%reduction in a number of parameters, demonstrating that efficient cluster-specific feature engineering, which allows each model to capture nuanced patterns within each cluster can improve overall predictive accuracy.
650		0	_aSign language _xData processing
700	0		_aChantri Polprasert, _eChairperson
700	0		_aMongkol Ekpanyapong, _eExamination Committee
710	2		_aADB-Japan Scholarship Program (ADB-JSP), _eScholarship Donor
810	2		_aAsian Institute of Technology. _tThesis : _vno.CS-25-01
856	4	0	_3Full-Text _uhttp://203.159.5.9/ait-thesis/Viewer/viewer.php?id=B23610
907			_a.b12478039 _bmnarc _ca
902			_a260227
998			_b0 _c260226 _dm _eh _fa _g0
945			_lmnarc
942			_c67
909			_aBarcode : - _bCREATED : 2026-02-19 _cRECORD # : i13574966 _dLPATRON : 0 _eLCHKIN : - _f# RENEWALS : 0 _g# OVERDUE : 0 _hIUSE3 : 0 _iTOT CHKOUT : 0 _jTOT RENEW : 0
999			_c69324 _d69324