{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,18]],"date-time":"2026-03-18T17:55:08Z","timestamp":1773856508803,"version":"3.50.1"},"reference-count":0,"publisher":"Slovenian Association Informatika","issue":"10","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IJCAI"],"abstract":"<jats:p>The proliferation of IoT-based systems necessitates intelligent edge devices capable of autonomously adapting to dynamic environments while operating under highly constrained resource conditions. However, current TinyML solutions rely predominantly on static models that cannot evolve after deployment, which limits their effectiveness in real-world scenarios where environmental conditions and data distributions continuously change. To address this limitation, this paper presents S\u00b3OvA (SVM, SMO, Syed with OvA), a reformable TinyML solution that enables progressive model updating directly on microcontrollers (MCUs) through hybrid offline\/online methodologies. It integrates Support Vector Machines (SVM) with Sequential Minimal Optimisation (SMO) for efficient training, Syed et al.\u2019s incremental learning method for dynamic updating, and the One-vs-All (OvA) strategy for multiclass problems. In addition, it implements linear, polynomial, and RBF kernels while preserving computational efficiency on resource-constrained platforms. Consequently, experimental validation on the ESP32-S Module demonstrates its ability to achieve accuracy improvements of up to 20 points through incremental learning, while preserving memory stability throughout adaptive cycles. At the same time, the evaluation reveals operational limits with empirically determined constraints on the one hand, and validates performance within binary classification, multiclass extension, and adaptive scenarios on the other. In parallel, memory analysis confirms implementation efficiency through stable heap usage and controlled resource progression. Thus, S\u00b3OvA bridges the critical gap between static TinyML deployments and the dynamic requirements of IoT-based systems, enabling local intelligence that simultaneously reduces latency, preserves data privacy, and enhances energy autonomy. This reformable solution therefore establishes a new paradigm for self-adaptive edge computing, where distributed intelligence emerges through continuous sensor-level learning, fundamentally transforming IoT-based systems from passive data collectors into autonomous decision-making entities.<\/jats:p>","DOI":"10.31449\/inf.v50i10.11947","type":"journal-article","created":{"date-parts":[[2026,3,18]],"date-time":"2026-03-18T11:12:49Z","timestamp":1773832369000},"source":"Crossref","is-referenced-by-count":0,"title":["S\u00b3OvA: A Reformable TinyML Solution for Self-Adaptive IoT-based Systems"],"prefix":"10.31449","volume":"50","author":[{"given":"Mohamed","family":"Maoui","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Rohallah","family":"Benaboud","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"16141","published-online":{"date-parts":[[2026,3,18]]},"container-title":["Informatica"],"original-title":[],"link":[{"URL":"https:\/\/www.informatica.si\/index.php\/informatica\/article\/download\/11947\/6616","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.informatica.si\/index.php\/informatica\/article\/download\/11947\/6616","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,18]],"date-time":"2026-03-18T11:12:50Z","timestamp":1773832370000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.informatica.si\/index.php\/informatica\/article\/view\/11947"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,3,18]]},"references-count":0,"journal-issue":{"issue":"10","published-online":{"date-parts":[[2026,3,18]]}},"URL":"https:\/\/doi.org\/10.31449\/inf.v50i10.11947","relation":{},"ISSN":["1854-3871","0350-5596"],"issn-type":[{"value":"1854-3871","type":"electronic"},{"value":"0350-5596","type":"print"}],"subject":[],"published":{"date-parts":[[2026,3,18]]}}}