| A study that was done by Nayak et all in 2017 year(24) | A study that was done by Shen et all in 2023 year(23) | A study that was done by Samuel et all in 2024 year(22) | A study that was done by Puri et all in 2024 year(21) | A study that was done by Cao et all in 2019 year(20) | A study that was done by Chen et all in the 2024 year(19) | A study that was done by Kim et all in the 2021 year(18) | A study that was done by Xu et all in the 2024 year (17) |
| Result | Result | Result | Result | Result | Result | Result | Result |
| This study shows that the use of AI in chemical reaction neural networks can increase the accuracy and sensitivity of chemical reactions and optimize chemical processes implying an important role | The result of this study represents that applying AI and dependent techniques to it can enhancement speed of chemical reactions and produce efficient materials via optimized flow chemistry | This research explains that the use of machine learning in the AI approach can help to enhance optimization in chemical reactions via promoting catalysis activity and increase reproduce new chemical materials | Results of this study show that the use of AI and machine learning can help to optimize chemical reactions or enhance precision and reduce time-consume for chemical reactions and this key can help one progress in designing drugs and pharmaceutical technology | They can minimize entropy generation with transfer mass causes optimization of chemical reactions and enhance reproduce yield in one chemical reaction | These results show that with use of machine learning, chemistry engineering, and data science can enhance optimization in chemical reactions with AI, in therefore global model can reproduce new reactions and local models cause an increased yield and selectivity of reaction | These scientists use AI with approach machine learning in converse methane gas to c2 formation and coke they can apply AI to reduce the production of coke and enhance the creation of c2 and amount error reach <5% | Using AI in the optimization of chemical reactions both causes improved economic conditions and environmental performance and also reduces greenhouse gases on earth |
Table 1
| Random forest | Support vector machines | QSAR | EToxPred | The types of machine learning platforms in AI in organic chemistry |
| High accuracy, handling non-linearity, feature importance, robustness, versatility, balanced performance | Interpret high dimensional data, is very effective in resolving optimal decision limitations, effective in | Enhance predictive performance, the inclusion of diverse data sources, ability to handle high dimensional data, | Determine toxicity drug discovery candidate efficiency | Benefit |
| Lack of interpretability, computational complexity, overfitting, imbalanced datasets | Modulating small dataset SVM is very poor in evaluating big chemical data, it can cause noise in classification chemical structures, and unrecognized accurate kernel parameters can approach to cause error | The efficiency of model update Sensitivity to quality of input data, potential for overfitting is prevalent, applicability of domains accessibility, problem in interpretation complexity of data, lack of understanding about molecular interpret data input | Reliability of the training data, interpretability of the model predictions, generalizability, regulatory acceptance | limitations |
| Quantitative- structure-activity prediction, chemical risk assessment, applicable in anticipating chemical domains | Predictive toxicology, structure-activity relationships, computational chemistry, environmental chemistry | Molecular modeling in organic chemistry integrates quantum chemistry with machine learning, enhancement of chemical research | Toxicity prediction, synthetic accessibility, enhancing research efficiency | Applicability in organic chemistry |
Table 2
Tables at a glance