Turker, IlkerAksu, Serkan2024-09-292024-09-2920220003-682X1872-910Xhttps://doi.org/10.1016/j.apacoust.2022.108660https://hdl.handle.net/20.500.14619/4373The proposed method contributes the time-series classification literature with a novel time-convexity based representation, which extends the current graph conversion approaches by introducing the time dimension, also introducing a colorful graph-generator approach. The representation capability of connectograms is tested in comparison with mel-spectrograms (mels) and MFCCs for an environmental sound classification task, as input to state-of-art transfer learning models. Results indicate that connectograms cannot compete with the best-performer mel-spectrogram representations in standalone format, however they significantly improve their classification performance in case they are combined as single layers of hybrid RGB representations. A combination of [mels + mels + connectogram] outperforms either sole representations or their combinations by 2-3%, with 96.46% classification accuracy for ResNet50 classifier model.(c) 2022 Elsevier Ltd. All rights reserved.eninfo:eu-repo/semantics/closedAccessGraph representationSound classificationTime-series classificationComplex networksDeep learningMachine learningArticle10.1016/j.apacoust.2022.1086602-s2.0-85124610723Q1191WOS:000783173100007Q1