Matches in SemOpenAlex for { <https://semopenalex.org/work/W4312892766> ?p ?o ?g. }
Showing items 1 to 94 of
94
with 100 items per page.
- W4312892766 abstract "Deep learning algorithms achieve exceptional accuracies in various tasks. Despite this success, those models are known to be prone to errors, i.e. low in robustness, due to differences between training and production environment. One might assume that more model complexity translates directly to more robustness. Therefore, we compare simple, classical models (logistic regression, support vector machine) with complex deep learning techniques (convolutional neural networks, transformers) to provide novel insights into the robustness of machine learning systems. In our approach, we assess the robustness by developing and applying three realistic perturbations, mimicking scanning, typing, and speech recognition errors occurring in inputs for text classification tasks. Hence, we performed a thorough study analyzing the impact of different perturbations with variable strengths on character and word level. A noteworthy finding is that algorithms with low complexity can achieve high robustness. Additionally, we demonstrate that augmented training regarding a specific perturbation can strengthen the chosen models' robustness against other perturbations without reducing their accuracy. Our results can impact the selection of machine learning models and provide a guideline on how to examine the robustness of text classification methods for real-world applications. Moreover, our implementation is publicly available, which contributes to the development of more robust machine learning systems." @default.
- W4312892766 created "2023-01-05" @default.
- W4312892766 creator A5013575494 @default.
- W4312892766 creator A5034494547 @default.
- W4312892766 creator A5035644298 @default.
- W4312892766 creator A5057360338 @default.
- W4312892766 creator A5071457501 @default.
- W4312892766 date "2022-07-18" @default.
- W4312892766 modified "2023-10-04" @default.
- W4312892766 title "Comparing the Robustness of Classical and Deep Learning Techniques for Text Classification" @default.
- W4312892766 cites W1552847225 @default.
- W4312892766 cites W1826790618 @default.
- W4312892766 cites W1986760892 @default.
- W4312892766 cites W2005661126 @default.
- W4312892766 cites W2031913254 @default.
- W4312892766 cites W2034225924 @default.
- W4312892766 cites W2069172670 @default.
- W4312892766 cites W2118020653 @default.
- W4312892766 cites W2140957143 @default.
- W4312892766 cites W2141125852 @default.
- W4312892766 cites W2156909104 @default.
- W4312892766 cites W2160815625 @default.
- W4312892766 cites W2165612380 @default.
- W4312892766 cites W2166183437 @default.
- W4312892766 cites W2167277498 @default.
- W4312892766 cites W2609368435 @default.
- W4312892766 cites W2735556405 @default.
- W4312892766 cites W2766177643 @default.
- W4312892766 cites W2767260595 @default.
- W4312892766 cites W2795386701 @default.
- W4312892766 cites W2799007037 @default.
- W4312892766 cites W2799194071 @default.
- W4312892766 cites W2904139120 @default.
- W4312892766 cites W2963857521 @default.
- W4312892766 cites W2963859254 @default.
- W4312892766 cites W2963969878 @default.
- W4312892766 cites W2964046515 @default.
- W4312892766 cites W2978440502 @default.
- W4312892766 cites W2989201733 @default.
- W4312892766 cites W2996601440 @default.
- W4312892766 cites W2996851481 @default.
- W4312892766 cites W3002654242 @default.
- W4312892766 cites W3105604018 @default.
- W4312892766 cites W3105625590 @default.
- W4312892766 cites W3124185353 @default.
- W4312892766 cites W3156333129 @default.
- W4312892766 cites W3181414820 @default.
- W4312892766 cites W3199178087 @default.
- W4312892766 doi "https://doi.org/10.1109/ijcnn55064.2022.9892242" @default.
- W4312892766 hasPublicationYear "2022" @default.
- W4312892766 type Work @default.
- W4312892766 citedByCount "0" @default.
- W4312892766 crossrefType "proceedings-article" @default.
- W4312892766 hasAuthorship W4312892766A5013575494 @default.
- W4312892766 hasAuthorship W4312892766A5034494547 @default.
- W4312892766 hasAuthorship W4312892766A5035644298 @default.
- W4312892766 hasAuthorship W4312892766A5057360338 @default.
- W4312892766 hasAuthorship W4312892766A5071457501 @default.
- W4312892766 hasConcept C104317684 @default.
- W4312892766 hasConcept C108583219 @default.
- W4312892766 hasConcept C119857082 @default.
- W4312892766 hasConcept C12267149 @default.
- W4312892766 hasConcept C154945302 @default.
- W4312892766 hasConcept C185592680 @default.
- W4312892766 hasConcept C41008148 @default.
- W4312892766 hasConcept C55493867 @default.
- W4312892766 hasConcept C63479239 @default.
- W4312892766 hasConcept C81363708 @default.
- W4312892766 hasConceptScore W4312892766C104317684 @default.
- W4312892766 hasConceptScore W4312892766C108583219 @default.
- W4312892766 hasConceptScore W4312892766C119857082 @default.
- W4312892766 hasConceptScore W4312892766C12267149 @default.
- W4312892766 hasConceptScore W4312892766C154945302 @default.
- W4312892766 hasConceptScore W4312892766C185592680 @default.
- W4312892766 hasConceptScore W4312892766C41008148 @default.
- W4312892766 hasConceptScore W4312892766C55493867 @default.
- W4312892766 hasConceptScore W4312892766C63479239 @default.
- W4312892766 hasConceptScore W4312892766C81363708 @default.
- W4312892766 hasLocation W43128927661 @default.
- W4312892766 hasOpenAccess W4312892766 @default.
- W4312892766 hasPrimaryLocation W43128927661 @default.
- W4312892766 hasRelatedWork W2337926734 @default.
- W4312892766 hasRelatedWork W2803710604 @default.
- W4312892766 hasRelatedWork W3136979370 @default.
- W4312892766 hasRelatedWork W4285106639 @default.
- W4312892766 hasRelatedWork W4311106074 @default.
- W4312892766 hasRelatedWork W4311257506 @default.
- W4312892766 hasRelatedWork W4312417841 @default.
- W4312892766 hasRelatedWork W4320802194 @default.
- W4312892766 hasRelatedWork W4321369474 @default.
- W4312892766 hasRelatedWork W4366224123 @default.
- W4312892766 isParatext "false" @default.
- W4312892766 isRetracted "false" @default.
- W4312892766 workType "article" @default.