“Sriya product and it's predictive capabilities hold a lot of promise for the industrial giants like GE in the areas of Healthcare, Aviation and Energy.”

“Based on my evaluation of Sriya.ai's product, its predictive model shows a lot of potential. Look forward to exploring more.”

“ Looking forward to partnering with Sriya AI to explore the potential of AI/ML in developing enhanced projection models for coastal flooding. Their pilot work for us was very successful in predicting Coastal Flooding in Charleston, SC and Savannah, GA using a multi modal approach. This work will help communities plan and better prepare for future flood events in conjunction with global sea level rise. .”

“ Thank you for the results from ML implementation on our datasets. I am happy to note that even with much smaller datasets than that is used for traditional ML-AI work, Sriya's approach has been able to come up with very informed predictions about the output of both the lime and cement synthesis using our novel calcination and cementing approach. Both the predictions as well as the recommendations for further optimization are accurate. I am also happy to note that your simulations are able to predict the results in a superior manner to other ML algorithms that you have evaluated as well. .”

“ We used Sriya.ai to help us improve predictive power for 30-day delinquencies and charge-offs at Capchase. It is an important problem for Capchase to solve since we are able to lower our loss rates and increase our revenue if we have an accurate assessment of credit risk. The problem in our domain is that the dataset for training is small (as opposed to consumer lending), and therefore, traditional techniques do not work as well. Sriya's technology worked very well on our small dataset to give us results with high accuracy and precision (low false negatives and false positives), which will help us reduce our loss rates while being able to make an offer to more clients. In addition to the simplicity of the output of the algorithm (1 output score called SXI), they provided us with decision trees with actionable insights. Some of those insights were counter-intuitive at first glance (e.g. larger companies default at a higher rate than smaller companies), but made sense when applied through a broader lens of our offering (i.e. we get adverse selection when larger companies turn to us versus borrowing from large banks). I would highly recommend Sriya's solution to anybody looking to extract actionable insights from datasets, no matter how small or large - I think they have the potential to change the trajectory of the financial services industry. ”

“Each time we pose a healthcare problem to Sriya, their algorithms perform incredibly. Looking at published studies and re analyzing their data, Sriya had outperformed Composer in predicting sepsis, Random Forest and others on heart failure readmission. The results are truly game changing. ”

“ We asked Sriya.ai to develop a predictive model for two datasets that are publicly available from UCI Machine Learning archive. One dataset contains 303 records for heart disease diagnosis (https://archive.ics.uci.edu/dataset/45/heart+disease). The second dataset contains 299 records for heart failure clinical records (https://archive.ics.uci.edu/dataset/519/heart+failure
+clinical+records). For the purpose of benchmarking performance, we used the following papers that have previously analyzed these datasets respectively: Nashif, S., Raihan, M. R., Islam, M. R., & Imam, M. H. (2018). Heart disease detection by using machine learning algorithms and a real-time cardiovascular health monitoring system. World Journal of Engineering and Technology, 6(4), 854-873. https://doi.org/10.4236/wjet.2018.64057 Chicco, D., Jurman, G. Machine learning can predict survival of patients with heart failure from serum creatinine and ejection fraction alone. BMC Med Inform Decis Mak 20, 16 (2020). https://doi.org/10.1186/s12911-020-1023-5 We find that SXI has higher accuracy than the benchmark (98.33% vs. 97.53) for the heart disease study. The precision and AUC are also higher than the benchmark. Similarly, SXI has higher accuracy than the benchmark for the heart failure dataset (98.33% vs. 78%). Again, the precision and AUC are also higher than the benchmark. Based on these results, I am happy to report that SXI offers a significant improvement to the standard predictive modeling toolset. As such, I am happy to recommend Sriya’s technology for use in medicine and other fields for the purpose of building predictive models with high accuracy. ”

“ As a medical advisor for Sriya.ai, I am continually impressed by its potential to transform patient care. Sriya.ai precisely predicts which patients are at the highest risk of hospital readmission within 30 days and identifies the specific vulnerabilities contributing to that risk. This empowers healthcare providers to implement proactive, targeted interventions for the patients who need them most. By improving patient outcomes and reducing the strain on hospital resources, Sriya.ai is a game-changer for healthcare systems seeking to deliver more efficient and compassionate care. ”

“ What further impressed us was the comprehensive report provided by the Sriya.AI team, which outlined their methodical approach to mitigating overfitting and underfitting during model training. This report also highlighted the versatility of their models, showcasing their ability to generate outstanding predictions for the mechanical properties of concrete—exceeding the capabilities of conventional regression models used by many researchers in the field of materials science. Given these exceptional results, I enthusiastically recommend Sriya.AI for predictive modeling and optimization, not only for ultra-high temperature ceramics but also for concrete and other materials. Their innovative approach, technical expertise, and commitment to excellence set a high benchmark in the realm of AI-driven materials science. ”

“Utilizing our previously accumulated experimental datasets as a training baseline, Sriya.ai’s algorithms have efficiently demonstrated pathways for achieving our targeted photochemical properties in a “shortest path” manner while increasing the overall fraction of experimental samples fitting within the “optimal process window”. The recommendations arising from Sriya.ai’s algorithms are especially valuable on two fronts: (1) they provide us insights into the relative importance through numerical weighting of each of our process variables towards achieving our targeted slurry photochemical properties; and (2) they provide us recommendations for investigating previously unexplored ranges of process variables that provide the highest likelihood of achieving the targeted photochemical properties while evaluating different slurry formulations. We are highly impressed and enthused that Sriya.ai’s algorithms can serve to greatly increase batch-to-batch performance repeatability and reproducibility of a given production slurry formulation with known small variations in photochemistry, and also greatly accelerate the development and optimization of new slurry formulations that DDM Systems will undoubtedly undertake in the future. We believe that further development of these algorithms will serve to greatly benefit the rapid optimization, process control and yield improvement in advanced manufacturing processes with multivariate statistical process characteristics. ”