Table 11 Differences in physical and chemical data predictions

Falls under the predictions of mass, capacity, density, response time, and size

Falls under prediction of the electronegativity, ionic potential, bonding energy, electrochemical characteristics, fragmentation, and structural attributes

There is no more uncomplicated process

The process is little complicated

Involvement of fewer descriptors in the prediction process

Involvement of a large number of descriptors in the prediction process

1D, 2D, and topological descriptors are mostly utilized for the prediction [271]

All topological, electronic, 1D, 2D, and 3D descriptors were utilized for the prediction [272]

The database size is typically smaller

The database is larger and more complex

Canonical SMILES are includes in the data set, mostly, 2D chemical structure is sufficient for the prediction

Data sets used for prediction includes canonical SIMILES, ChEMBL, and 2D and 3D chemical structures

Predictions scores typically produce immediate results

Prediction outcomes should not be direct; instead, they correlate with other outcomes to produce final outcomes

There are fewer accessible methods and approaches; most are QSPR-based

Numerous techniques and strategies are employed, such as QSAR and QSRR

Mathematical calculations only involved such as regression and correlations

There are included quantum chemical calculations like TD-DFT and DFT

The final outcomes are unaffected by the less accurate physical data predictions in some cases

Less accurate chemical data prediction will have an impact on the outcomes