|
ИСТИНА |
Войти в систему Регистрация |
ИСТИНА ИНХС РАН |
||
Chapter 13. Benchmarking Machine‐Learning Methodsстатья
- Авторы: Baskin I.I., Marcou G., Horvath D., Varnek A.
- Сборник: Tutorials in Chemoinformatics. First Edition. Edited by Alexandre Varnek
- Глава в коллективной монографии
- Год издания: 2017
- Место издания: John Wiley & Sons Ltd United Kingdom
- Первая страница: 209
- Последняя страница: 222
- DOI: 10.1002/9781119161110.ch13
- Аннотация: Nowadays, there exist hundreds of different machine learning methods. This chapter includes a tutorial considering the following machine learning methods for performing regression: zero regression (ZeroR), multiple linear regression (MLR), partial least squares (PLS), support vector regression (SVR), k nearest neighbors (kNN), back-propagation neural network (BPNN), regression tree M5P (M5P), and regression by discretization based on random forest (RD-RF). This part of the tutorial requires the installation of the Partial Least Square method, which is provided in Weka as an additional package. The installation is done in two steps. First, in the Weka main interface, in the menu Tools, choose the Package Manager option. Then in a new window, find the partialLeastSquare package. It is categorized as a preprocessing tool. Finally, click the Install button. In order to achieve increased predictive performance of the QSAR/QSPR models, it is suggested to apply different machine learning methods, to compare results and to select the most appropriate one.
- Добавил в систему: Баскин Игорь Иосифович