TY - JOUR A1 - Schaffner, Ellen A1 - Schiefele, Ulrich T1 - The prediction of reading comprehension by cognitive and motivational factors - does text accessibility during comprehension testing make a difference? JF - Learning and individual differences N2 - This study examined the unique contributions of various predictors to reading comprehension measured either without or with access to the text during testing. Reasoning ability, prior knowledge, and decoding skills were assumed to have stronger contributions to comprehension without text access than with text access, whereas current motivation should be more strongly associated with comprehension measured with access to the text. Metacognitive strategy knowledge and test anxiety were expected to be equally associated with comprehension in the two test conditions. Participants were 424 eighth- and ninth-grade students. They were presented with several instruments measuring cognitive and motivational predictors and read a text on a mathematical topic; then half of them took a test on comprehension either without or with text access. Based on multiple-group structural equation modeling, results indicated that reasoning ability, decoding ability, and metacognitive strategy knowledge significantly predicted comprehension only in the without-text condition, whereas achievement motivation and test anxiety significantly predicted comprehension only in the with-text condition. The unique contributions of intrinsic motivation to comprehension were significant, but did unexpectedly not differ between the without-text and the with-text condition. KW - Reading comprehension KW - Reasoning ability KW - Prior knowledge KW - Metacognitive strategy knowledge KW - Current motivation KW - Test anxiety Y1 - 2013 U6 - https://doi.org/10.1016/j.lindif.2013.04.003 SN - 1041-6080 VL - 26 IS - 8 SP - 42 EP - 54 PB - Elsevier CY - Amsterdam ER - TY - JOUR A1 - Perscheid, Cindy T1 - Comprior BT - Facilitating the implementation and automated benchmarking of prior knowledge-based feature selection approaches on gene expression data sets JF - BMC Bioinformatics N2 - Background Reproducible benchmarking is important for assessing the effectiveness of novel feature selection approaches applied on gene expression data, especially for prior knowledge approaches that incorporate biological information from online knowledge bases. However, no full-fledged benchmarking system exists that is extensible, provides built-in feature selection approaches, and a comprehensive result assessment encompassing classification performance, robustness, and biological relevance. Moreover, the particular needs of prior knowledge feature selection approaches, i.e. uniform access to knowledge bases, are not addressed. As a consequence, prior knowledge approaches are not evaluated amongst each other, leaving open questions regarding their effectiveness. Results We present the Comprior benchmark tool, which facilitates the rapid development and effortless benchmarking of feature selection approaches, with a special focus on prior knowledge approaches. Comprior is extensible by custom approaches, offers built-in standard feature selection approaches, enables uniform access to multiple knowledge bases, and provides a customizable evaluation infrastructure to compare multiple feature selection approaches regarding their classification performance, robustness, runtime, and biological relevance. Conclusion Comprior allows reproducible benchmarking especially of prior knowledge approaches, which facilitates their applicability and for the first time enables a comprehensive assessment of their effectiveness KW - Feature selection KW - Prior knowledge KW - Gene expression KW - Reproducible benchmarking Y1 - 2021 U6 - https://doi.org/10.1186/s12859-021-04308-z SN - 1471-2105 VL - 22 SP - 1 EP - 15 PB - Springer Nature CY - London ER - TY - GEN A1 - Perscheid, Cindy T1 - Comprior: facilitating the implementation and automated benchmarking of prior knowledge-based feature selection approaches on gene expression data sets T2 - Zweitveröffentlichungen der Universität Potsdam : Reihe der Digital Engineering Fakultät N2 - Background Reproducible benchmarking is important for assessing the effectiveness of novel feature selection approaches applied on gene expression data, especially for prior knowledge approaches that incorporate biological information from online knowledge bases. However, no full-fledged benchmarking system exists that is extensible, provides built-in feature selection approaches, and a comprehensive result assessment encompassing classification performance, robustness, and biological relevance. Moreover, the particular needs of prior knowledge feature selection approaches, i.e. uniform access to knowledge bases, are not addressed. As a consequence, prior knowledge approaches are not evaluated amongst each other, leaving open questions regarding their effectiveness. Results We present the Comprior benchmark tool, which facilitates the rapid development and effortless benchmarking of feature selection approaches, with a special focus on prior knowledge approaches. Comprior is extensible by custom approaches, offers built-in standard feature selection approaches, enables uniform access to multiple knowledge bases, and provides a customizable evaluation infrastructure to compare multiple feature selection approaches regarding their classification performance, robustness, runtime, and biological relevance. Conclusion Comprior allows reproducible benchmarking especially of prior knowledge approaches, which facilitates their applicability and for the first time enables a comprehensive assessment of their effectiveness T3 - Zweitveröffentlichungen der Universität Potsdam : Reihe der Digital Engineering Fakultät - 010 KW - Feature selection KW - Prior knowledge KW - Gene expression KW - Reproducible benchmarking Y1 - 2022 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:517-opus4-548943 SP - 1 EP - 15 PB - Universitätsverlag Potsdam CY - Potsdam ER -