Journal: Molecules

Article Title: A Review of Machine Learning and QSAR/QSPR Predictions for Complexes of Organic Molecules with Cyclodextrins

doi: 10.3390/molecules29133159

Figure Lengend Snippet: Summary of predictions for Δ G , ln ⁡ K , and log ⁡ K of complex formation with α-CD, arranged in chronological order.

Article Snippet: Δ G , about 200 [ , ] , 75%/25%, R package caret , PaDEL , ensemble of: Cubist, GBM, MARSplines, RF, polynomial SVR, RBF SVR (Rstudio) , for single models: R 2 test from 0.63 to 0.78, Q 2 LOO from 0.50 to 0.64 , + , [ ] .

Techniques: Software

Journal: Molecules

Article Title: A Review of Machine Learning and QSAR/QSPR Predictions for Complexes of Organic Molecules with Cyclodextrins

doi: 10.3390/molecules29133159

Figure Lengend Snippet: Summary of predictions for Δ G , ln ⁡ K , and log ⁡ K of complex formation with β-CD, arranged in chronological order.

Article Snippet: Δ G , about 200 [ , ] , 75%/25%, R package caret , PaDEL , ensemble of: Cubist, GBM, MARSplines, RF, polynomial SVR, RBF SVR (Rstudio) , for single models: R 2 test from 0.63 to 0.78, Q 2 LOO from 0.50 to 0.64 , + , [ ] .

Techniques: Software, Derivative Assay, Biomarker Discovery, Activity Assay

Journal: PeerJ Computer Science

Article Title: REDf: a deep learning model for short-term load forecasting to facilitate renewable integration and attaining the SDGs 7, 9, and 13

doi: 10.7717/peerj-cs.2819

Figure Lengend Snippet: Architecture of the proposed forecasting model, REDf.

Article Snippet: This discussion is also supported by the visual representation of the comparison of the results shown in . presents an integrated performance comparison of various forecasting models, including REDf, SVR, Prophet, and RFR, across multiple datasets (AEP, COMED, DAYTON, and PJME).

Techniques:

Journal: PeerJ Computer Science

Article Title: REDf: a deep learning model for short-term load forecasting to facilitate renewable integration and attaining the SDGs 7, 9, and 13

doi: 10.7717/peerj-cs.2819

Figure Lengend Snippet: (A) Proposed REDf model, (b) SVR model, (C) Facebook Prophet model, and (D) RFR model.

Article Snippet: This discussion is also supported by the visual representation of the comparison of the results shown in . presents an integrated performance comparison of various forecasting models, including REDf, SVR, Prophet, and RFR, across multiple datasets (AEP, COMED, DAYTON, and PJME).

Techniques:

Journal: Neural Computing & Applications

Article Title: Black box modeling of PIDs implemented in PLCs without structural information: a support vector regression approach

doi: 10.1007/s00521-014-1754-2

Figure Lengend Snippet: Output of the PID algorithm (PLC GE VersaMax ‘PID ISA’ function block) as well as the corresponding simulated outputs of the SVR for K C = 0.66, T I = 5.8, T D = 0.1

Article Snippet: Furthermore, the minor difference between the results of the training and testing of the SVR (90.93–86.96 % for parallel form GE, 91.26–88.62 % for ISA GE, 92.00–89.67 % for ISA Siemens) demonstrates good generalization properties.

Techniques: Blocking Assay

Journal: Neural Computing & Applications

Article Title: Black box modeling of PIDs implemented in PLCs without structural information: a support vector regression approach

doi: 10.1007/s00521-014-1754-2

Figure Lengend Snippet: Results for the SVR, Tf model and NNs, ISA GE

Article Snippet: Furthermore, the minor difference between the results of the training and testing of the SVR (90.93–86.96 % for parallel form GE, 91.26–88.62 % for ISA GE, 92.00–89.67 % for ISA Siemens) demonstrates good generalization properties.

Techniques:

Journal: Neural Computing & Applications

Article Title: Black box modeling of PIDs implemented in PLCs without structural information: a support vector regression approach

doi: 10.1007/s00521-014-1754-2

Figure Lengend Snippet: Results for the SVR, Tf model and NNs, ISA Siemens

Article Snippet: Furthermore, the minor difference between the results of the training and testing of the SVR (90.93–86.96 % for parallel form GE, 91.26–88.62 % for ISA GE, 92.00–89.67 % for ISA Siemens) demonstrates good generalization properties.

Techniques: