I am Srinath, and I am a Solutions Architect with a background in ERP and over 25+ years of experience in the industry. Read further to learn about my journey with Great Learning’s PGP Artificial Intelligence and Machine Learning Course.
I like to travel & have mastered the art of not doing anything, always trying to identify patterns everywhere. My career trajectory thus far has involved writing code for the US Insurance domain, ePOS systems, migration of the Legacy system to Oracle EBS, and developing the eChallan system. For the last five years have been actively involved in ML/AI.
Before this course being a Solution Architect was working on Enterprise Mobility for Traffic Solutions in government vertical & Remote Process Automation for some large corporates in the GST domain, allowing them to claim Input Tax Credit on Travel in time. But there were no mechanisms to verify the text with what’s being present on the image & also whether the device is undergoing gradual deterioration. There were too many people involved in doing image verification & also no methods to predict the performance of the device. We were losing money on manpower deployed & we did not have any kind of metrics to improve further.
We used open-cv & tesseract to extract the text from the images reducing the dependency on manpower. We collected the faulty device’s data, labeled them & trained the ML model to predict the state of the device. We used Logistic Regression, Naïve-Bayes & Random Forest Classifier to begin with & finalized on Random Forest Classifier based on the accuracy. Moved on from Binary classification to Multiclass classification.
The degradation of the device starts at 18 months. The performance of the device had direct proportion to the firmware being used. We collected the device’s data & labeled them accordingly. Trained the model based on the labeled data. The output of the ML model was verified with the Customer Support Team & those corrections were incorporated into the model to increase the accuracy. This ultimately helped us in Improved Customer Satisfaction & Saved our money due to reduced manpower. I now have Real-time experience in open-cv, Binary & multi-class classification.