Literature DB >> 21554073

Trade-off between accuracy and interpretability for predictive in silico modeling.

Ulf Johansson1, Cecilia Sönströd, Ulf Norinder, Henrik Boström.   

Abstract

BACKGROUND: Accuracy concerns the ability of a model to make correct predictions, while interpretability concerns to what degree the model allows for human understanding. Models exhibiting the former property are many times more complex and opaque, while interpretable models may lack the necessary accuracy. The trade-off between accuracy and interpretability for predictive in silico modeling is investigated.
METHOD: A number of state-of-the-art methods for generating accurate models are compared with state-of-the-art methods for generating transparent models.
CONCLUSION: Results on 16 biopharmaceutical classification tasks demonstrate that, although the opaque methods generally obtain higher accuracies than the transparent ones, one often only has to pay a quite limited penalty in terms of predictive performance when choosing an interpretable model.

Entities:  

Mesh:

Substances:

Year:  2011        PMID: 21554073     DOI: 10.4155/fmc.11.23

Source DB:  PubMed          Journal:  Future Med Chem        ISSN: 1756-8919            Impact factor:   3.808


  8 in total

Review 1.  Chemical predictive modelling to improve compound quality.

Authors:  John G Cumming; Andrew M Davis; Sorel Muresan; Markus Haeberlein; Hongming Chen
Journal:  Nat Rev Drug Discov       Date:  2013-12       Impact factor: 84.694

2.  Interpretation of machine learning models using shapley values: application to compound potency and multi-target activity predictions.

Authors:  Raquel Rodríguez-Pérez; Jürgen Bajorath
Journal:  J Comput Aided Mol Des       Date:  2020-05-02       Impact factor: 3.686

3.  Representing descriptors derived from multiple conformations as uncertain features for machine learning.

Authors:  Ulf Norinder; Henrik Boström
Journal:  J Mol Model       Date:  2013-03-12       Impact factor: 1.810

4.  Enabling personalized decision support with patient-generated data and attributable components.

Authors:  Elliot G Mitchell; Esteban G Tabak; Matthew E Levine; Lena Mamykina; David J Albers
Journal:  J Biomed Inform       Date:  2020-12-13       Impact factor: 6.317

5.  How feasible is it to abandon statistical significance? A reflection based on a short survey.

Authors:  Fredi Alexander Diaz-Quijano; Fernando Morelli Calixto; José Mário Nunes da Silva
Journal:  BMC Med Res Methodol       Date:  2020-06-03       Impact factor: 4.615

6.  Could graph neural networks learn better molecular representation for drug discovery? A comparison study of descriptor-based and graph-based models.

Authors:  Dejun Jiang; Zhenxing Wu; Chang-Yu Hsieh; Guangyong Chen; Ben Liao; Zhe Wang; Chao Shen; Dongsheng Cao; Jian Wu; Tingjun Hou
Journal:  J Cheminform       Date:  2021-02-17       Impact factor: 5.514

Review 7.  Review of applications and challenges of quantitative systems pharmacology modeling and machine learning for heart failure.

Authors:  Limei Cheng; Yuchi Qiu; Brian J Schmidt; Guo-Wei Wei
Journal:  J Pharmacokinet Pharmacodyn       Date:  2021-10-12       Impact factor: 2.745

8.  Feasibility Study of Constructing a Screening Tool for Adolescent Diabetes Detection Applying Machine Learning Methods.

Authors:  Hansel Hu; Tin Lai; Farnaz Farid
Journal:  Sensors (Basel)       Date:  2022-08-17       Impact factor: 3.847

  8 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.