FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4307261738> ?p ?o ?g. }

• W4307261738 endingPage "8060" @default.
• W4307261738 startingPage "8060" @default.
• W4307261738 abstract "Time series classification is an active research topic due to its wide range of applications and the proliferation of sensory data. Convolutional neural networks (CNNs) are ubiquitous in modern machine learning (ML) models. In this work, we present a matched filter (MF) interpretation of CNN classifiers accompanied by an experimental proof of concept using a carefully developed synthetic dataset. We exploit this interpretation to develop an MF CNN model for time series classification comprising a stack of a Conv1D layer followed by a GlobalMaxPooling layer acting as a typical MF for automated feature extraction and a fully connected layer with softmax activation for computing class probabilities. The presented interpretation enables developing superlight highly accurate classifier models that meet the tight requirements of edge inference. Edge inference is emerging research that addresses the latency, availability, privacy, and connectivity concerns of the commonly deployed cloud inference. The MF-based CNN model has been applied to the sensor-based human activity recognition (HAR) problem due to its significant importance in a broad range of applications. The UCI-HAR, WISDM-AR, and MotionSense datasets are used for model training and testing. The proposed classifier is tested and benchmarked on an android smartphone with average accuracy and F1 scores of 98% and 97%, respectively, which outperforms state-of-the-art HAR methods in terms of classification accuracy and run-time performance. The proposed model size is less than 150 KB, and the average inference time is less than 1 ms. The presented interpretation helps develop a better understanding of CNN operation and decision mechanisms. The proposed model is distinguished from related work by jointly featuring interpretability, high accuracy, and low computational cost, enabling its ready deployment on a wide set of mobile devices for a broad range of applications." @default.
• W4307261738 created "2022-10-31" @default.
• W4307261738 creator A5022893014 @default.
• W4307261738 date "2022-10-21" @default.
• W4307261738 modified "2023-10-14" @default.
• W4307261738 title "Matched Filter Interpretation of CNN Classifiers with Application to HAR" @default.
• W4307261738 cites W2017634428 @default.
• W4307261738 cites W2551393996 @default.
• W4307261738 cites W2657631929 @default.
• W4307261738 cites W2759690896 @default.
• W4307261738 cites W2898186212 @default.
• W4307261738 cites W2962949934 @default.
• W4307261738 cites W2964350391 @default.
• W4307261738 cites W3003728340 @default.
• W4307261738 cites W3011785450 @default.
• W4307261738 cites W3100777112 @default.
• W4307261738 cites W3128981305 @default.
• W4307261738 cites W3136077883 @default.
• W4307261738 cites W3136388241 @default.
• W4307261738 cites W3154435685 @default.
• W4307261738 cites W3162538709 @default.
• W4307261738 cites W3196399896 @default.
• W4307261738 cites W4205137648 @default.
• W4307261738 cites W4280587517 @default.
• W4307261738 cites W4289172806 @default.
• W4307261738 cites W4294691681 @default.
• W4307261738 doi "https://doi.org/10.3390/s22208060" @default.
• W4307261738 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/36298408" @default.
• W4307261738 hasPublicationYear "2022" @default.
• W4307261738 type Work @default.
• W4307261738 citedByCount "3" @default.
• W4307261738 countsByYear W43072617382023 @default.
• W4307261738 crossrefType "journal-article" @default.
• W4307261738 hasAuthorship W4307261738A5022893014 @default.
• W4307261738 hasBestOaLocation W43072617381 @default.
• W4307261738 hasConcept C119857082 @default.
• W4307261738 hasConcept C153180895 @default.
• W4307261738 hasConcept C154945302 @default.
• W4307261738 hasConcept C188441871 @default.
• W4307261738 hasConcept C2776214188 @default.
• W4307261738 hasConcept C41008148 @default.
• W4307261738 hasConcept C81363708 @default.
• W4307261738 hasConcept C95623464 @default.
• W4307261738 hasConceptScore W4307261738C119857082 @default.
• W4307261738 hasConceptScore W4307261738C153180895 @default.
• W4307261738 hasConceptScore W4307261738C154945302 @default.
• W4307261738 hasConceptScore W4307261738C188441871 @default.
• W4307261738 hasConceptScore W4307261738C2776214188 @default.
• W4307261738 hasConceptScore W4307261738C41008148 @default.
• W4307261738 hasConceptScore W4307261738C81363708 @default.
• W4307261738 hasConceptScore W4307261738C95623464 @default.
• W4307261738 hasFunder F4320322804 @default.
• W4307261738 hasIssue "20" @default.
• W4307261738 hasLocation W43072617381 @default.
• W4307261738 hasLocation W43072617382 @default.
• W4307261738 hasLocation W43072617383 @default.
• W4307261738 hasOpenAccess W4307261738 @default.
• W4307261738 hasPrimaryLocation W43072617381 @default.
• W4307261738 hasRelatedWork W2610906757 @default.
• W4307261738 hasRelatedWork W2743258233 @default.
• W4307261738 hasRelatedWork W2758063741 @default.
• W4307261738 hasRelatedWork W2908861653 @default.
• W4307261738 hasRelatedWork W2977314777 @default.
• W4307261738 hasRelatedWork W3034884618 @default.
• W4307261738 hasRelatedWork W3120400911 @default.
• W4307261738 hasRelatedWork W3208883981 @default.
• W4307261738 hasRelatedWork W4307834408 @default.
• W4307261738 hasRelatedWork W4320925816 @default.
• W4307261738 hasVolume "22" @default.
• W4307261738 isParatext "false" @default.
• W4307261738 isRetracted "false" @default.
• W4307261738 workType "article" @default.