FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3212464206> ?p ?o ?g. }

• W3212464206 abstract "On the one hand, artificial neural networks (ANNs) are commonly labelled as black-boxes, lacking interpretability; an issue that hinders human understanding of ANNs' behaviors. A need exists to generate a meaningful sequential logic of the ANN for interpreting a production process of a specific output. On the other hand, decision trees exhibit better interpretability and expressive power due to their representation language and the existence of efficient algorithms to transform the trees into rules. However, growing a decision tree based on the available data could produce larger than necessary trees or trees that do not generalise well. In this paper, we introduce two novel multivariate decision tree (MDT) algorithms for rule extraction from ANNs: an Exact-Convertible Decision Tree (EC-DT) and an Extended C-Net algorithm. They both transform a neural network with Rectified Linear Unit activation functions into a representative tree, which can further be used to extract multivariate rules for reasoning. While the EC-DT translates an ANN in a layer-wise manner to represent exactly the decision boundaries implicitly learned by the hidden layers of the network, the Extended C-Net combines the decompositional approach from EC-DT with a C5 tree learning algorithm to form decision rules. The results suggest that while EC-DT is superior in preserving the structure and the fidelity of ANN, Extended C-Net generates the most compact and highly effective trees from ANN. Both proposed MDT algorithms generate rules including combinations of multiple attributes for precise interpretations for decision-making." @default.
• W3212464206 created "2021-11-22" @default.
• W3212464206 creator A5022974795 @default.
• W3212464206 creator A5023351148 @default.
• W3212464206 creator A5047705901 @default.
• W3212464206 date "2021-11-18" @default.
• W3212464206 modified "2023-09-27" @default.
• W3212464206 title "Towards Interpretable ANNs: An Exact Transformation to Multi-Class Multivariate Decision Trees." @default.
• W3212464206 cites W128754594 @default.
• W3212464206 cites W1515620500 @default.
• W3212464206 cites W1547466200 @default.
• W3212464206 cites W1585983533 @default.
• W3212464206 cites W1589549454 @default.
• W3212464206 cites W1650780448 @default.
• W3212464206 cites W1673002162 @default.
• W3212464206 cites W188378287 @default.
• W3212464206 cites W1900913856 @default.
• W3212464206 cites W1973722433 @default.
• W3212464206 cites W1981249599 @default.
• W3212464206 cites W1982421201 @default.
• W3212464206 cites W1986733968 @default.
• W3212464206 cites W1989574962 @default.
• W3212464206 cites W1995291953 @default.
• W3212464206 cites W2000257498 @default.
• W3212464206 cites W2006235059 @default.
• W3212464206 cites W2027857686 @default.
• W3212464206 cites W2036677213 @default.
• W3212464206 cites W2040017661 @default.
• W3212464206 cites W2041688042 @default.
• W3212464206 cites W2063046703 @default.
• W3212464206 cites W2063689928 @default.
• W3212464206 cites W2079109804 @default.
• W3212464206 cites W2088203270 @default.
• W3212464206 cites W2094558429 @default.
• W3212464206 cites W2100066980 @default.
• W3212464206 cites W2106075360 @default.
• W3212464206 cites W2107222792 @default.
• W3212464206 cites W2107484060 @default.
• W3212464206 cites W2112337384 @default.
• W3212464206 cites W2112649152 @default.
• W3212464206 cites W2113882472 @default.
• W3212464206 cites W2117850960 @default.
• W3212464206 cites W2125847307 @default.
• W3212464206 cites W2134682899 @default.
• W3212464206 cites W2153125595 @default.
• W3212464206 cites W2154266223 @default.
• W3212464206 cites W2158599326 @default.
• W3212464206 cites W2282821441 @default.
• W3212464206 cites W2332488709 @default.
• W3212464206 cites W2343678873 @default.
• W3212464206 cites W2520774990 @default.
• W3212464206 cites W2535129363 @default.
• W3212464206 cites W2594475271 @default.
• W3212464206 cites W2605409611 @default.
• W3212464206 cites W2764024122 @default.
• W3212464206 cites W2765793020 @default.
• W3212464206 cites W2788403449 @default.
• W3212464206 cites W2789468576 @default.
• W3212464206 cites W2803532212 @default.
• W3212464206 cites W2891503716 @default.
• W3212464206 cites W2898027682 @default.
• W3212464206 cites W2899983622 @default.
• W3212464206 cites W2903137655 @default.
• W3212464206 cites W2912664727 @default.
• W3212464206 cites W2962858109 @default.
• W3212464206 cites W2964043796 @default.
• W3212464206 cites W2964303497 @default.
• W3212464206 cites W2966349618 @default.
• W3212464206 cites W2978740895 @default.
• W3212464206 cites W3035438226 @default.
• W3212464206 cites W3046793792 @default.
• W3212464206 cites W3090855408 @default.
• W3212464206 cites W3093718041 @default.
• W3212464206 cites W3105959310 @default.
• W3212464206 cites W3113524144 @default.
• W3212464206 cites W3123487819 @default.
• W3212464206 cites W3129402291 @default.
• W3212464206 cites W3173097008 @default.
• W3212464206 cites W3190219082 @default.
• W3212464206 cites W40922836 @default.
• W3212464206 hasPublicationYear "2021" @default.
• W3212464206 type Work @default.
• W3212464206 sameAs 3212464206 @default.
• W3212464206 citedByCount "0" @default.
• W3212464206 crossrefType "posted-content" @default.
• W3212464206 hasAuthorship W3212464206A5022974795 @default.
• W3212464206 hasAuthorship W3212464206A5023351148 @default.
• W3212464206 hasAuthorship W3212464206A5047705901 @default.
• W3212464206 hasConcept C10229987 @default.
• W3212464206 hasConcept C104317684 @default.
• W3212464206 hasConcept C113174947 @default.
• W3212464206 hasConcept C119857082 @default.
• W3212464206 hasConcept C124101348 @default.
• W3212464206 hasConcept C134306372 @default.
• W3212464206 hasConcept C154945302 @default.
• W3212464206 hasConcept C183931066 @default.
• W3212464206 hasConcept C185592680 @default.
• W3212464206 hasConcept C204241405 @default.
• W3212464206 hasConcept C2781067378 @default.
• W3212464206 hasConcept C33923547 @default.

• next