Matches in SemOpenAlex for { <https://semopenalex.org/work/W1548698874> ?p ?o ?g. }
- W1548698874 endingPage "163" @default.
- W1548698874 startingPage "137" @default.
- W1548698874 abstract "A widely acknowledged drawback of many statistical modelling techniques, commonly used in machine learning, is that the resulting model is extremely difficult to interpret. A number of new concepts and algorithms have been introduced by researchers to address this problem. They focus primarily on determining which inputs are relevant in predicting the output. This work describes a transparent, advanced non-linear modelling approach that enables the constructed predictive models to be visualised, allowing model validation and assisting in interpretation. The technique combines the representational advantage of a sparse ANOVA decomposition, with the good generalisation ability of a kernel machine. It achieves this by employing two forms of regularisation: a 1-norm based structural regulariser to enforce transparency, and a 2-norm based regulariser to control smoothness. The resulting model structure can be visualised showing the overall effects of different inputs, their interactions, and the strength of the interactions. The robustness of the technique is illustrated using a range of both artifical and “real world” datasets. The performance is compared to other modelling techniques, and it is shown to exhibit competitive generalisation performance together with improved interpretability." @default.
- W1548698874 created "2016-06-24" @default.
- W1548698874 creator A5034097417 @default.
- W1548698874 creator A5077947107 @default.
- W1548698874 date "2002-01-01" @default.
- W1548698874 modified "2023-09-23" @default.
- W1548698874 cites W104267937 @default.
- W1548698874 cites W105756641 @default.
- W1548698874 cites W116941811 @default.
- W1548698874 cites W122505910 @default.
- W1548698874 cites W123991060 @default.
- W1548698874 cites W126994218 @default.
- W1548698874 cites W129216326 @default.
- W1548698874 cites W133826426 @default.
- W1548698874 cites W146336132 @default.
- W1548698874 cites W1492033236 @default.
- W1548698874 cites W1492468531 @default.
- W1548698874 cites W149585525 @default.
- W1548698874 cites W1502092126 @default.
- W1548698874 cites W150254129 @default.
- W1548698874 cites W1508954278 @default.
- W1548698874 cites W1514668021 @default.
- W1548698874 cites W1516573478 @default.
- W1548698874 cites W1523294845 @default.
- W1548698874 cites W1527059713 @default.
- W1548698874 cites W1547246444 @default.
- W1548698874 cites W1549773109 @default.
- W1548698874 cites W1550570395 @default.
- W1548698874 cites W1554663460 @default.
- W1548698874 cites W1563088657 @default.
- W1548698874 cites W1564556749 @default.
- W1548698874 cites W1567512734 @default.
- W1548698874 cites W1574672153 @default.
- W1548698874 cites W1582178871 @default.
- W1548698874 cites W1593996126 @default.
- W1548698874 cites W1594031697 @default.
- W1548698874 cites W1601740268 @default.
- W1548698874 cites W1602485996 @default.
- W1548698874 cites W1602599529 @default.
- W1548698874 cites W1603220789 @default.
- W1548698874 cites W1604293137 @default.
- W1548698874 cites W1604938182 @default.
- W1548698874 cites W161002486 @default.
- W1548698874 cites W1615454278 @default.
- W1548698874 cites W1617735623 @default.
- W1548698874 cites W161994183 @default.
- W1548698874 cites W1631449371 @default.
- W1548698874 cites W1638784138 @default.
- W1548698874 cites W1639247225 @default.
- W1548698874 cites W1646949910 @default.
- W1548698874 cites W1656634586 @default.
- W1548698874 cites W1668978017 @default.
- W1548698874 cites W1687643454 @default.
- W1548698874 cites W1691907524 @default.
- W1548698874 cites W1695561774 @default.
- W1548698874 cites W1698137110 @default.
- W1548698874 cites W1705151579 @default.
- W1548698874 cites W1710315379 @default.
- W1548698874 cites W1746680969 @default.
- W1548698874 cites W1782095941 @default.
- W1548698874 cites W179338030 @default.
- W1548698874 cites W179344782 @default.
- W1548698874 cites W1800232718 @default.
- W1548698874 cites W1810793954 @default.
- W1548698874 cites W1818879346 @default.
- W1548698874 cites W1831940257 @default.
- W1548698874 cites W1835270644 @default.
- W1548698874 cites W184314356 @default.
- W1548698874 cites W1843577658 @default.
- W1548698874 cites W1856330551 @default.
- W1548698874 cites W1858290345 @default.
- W1548698874 cites W1869391892 @default.
- W1548698874 cites W1882532909 @default.
- W1548698874 cites W1915094907 @default.
- W1548698874 cites W193052636 @default.
- W1548698874 cites W1932086218 @default.
- W1548698874 cites W1956866467 @default.
- W1548698874 cites W1963603436 @default.
- W1548698874 cites W1963866822 @default.
- W1548698874 cites W1963908140 @default.
- W1548698874 cites W1964259370 @default.
- W1548698874 cites W1964362866 @default.
- W1548698874 cites W1964793447 @default.
- W1548698874 cites W1965335954 @default.
- W1548698874 cites W1965442628 @default.
- W1548698874 cites W1965681753 @default.
- W1548698874 cites W1966201593 @default.
- W1548698874 cites W1966225015 @default.
- W1548698874 cites W1966254786 @default.
- W1548698874 cites W1966276179 @default.
- W1548698874 cites W1966407672 @default.
- W1548698874 cites W1966426760 @default.
- W1548698874 cites W1966861171 @default.
- W1548698874 cites W1966914996 @default.
- W1548698874 cites W1967059475 @default.
- W1548698874 cites W1967422973 @default.
- W1548698874 cites W1967747901 @default.
- W1548698874 cites W1968201328 @default.