Karthik Sarma

As a dedicated tech student specializing in Artificial Intelligence and Data Sciences, I am passionate about harnessing the power of cutting-edge technology to solve complex problems and drive innovation. My academic journey has equipped me with a robust understanding of machine learning algorithms, data analysis, and predictive modeling. Through hands-on projects and rigorous coursework, I have developed a keen ability to transform raw data into actionable insights, creating intelligent systems that can learn and adapt. Eager to contribute to the evolving landscape of AI and data science, I am committed to applying my skills to real-world challenges, collaborating with experts in the field, and continuously expanding my knowledge to stay at the forefront of technological advancements.

PROJECTS
EDA and Pre-Processing on Banking Dataset :
• Used a banking dataset and used the application multiple Python libraries including numpy, pandas, matplotlib to do all the necessary steps for Exploratory Data Analysis and Preprocessing of the Data.
• Found out useful information such as the deposits done by different classes of marital status and the differences in deposit based on age differences by using hitograms, boxplots and correlation matrix
Model building on Crop dataset:
• Built a project where a crop dataset was trained on multiple models and the best model was then determined to make recommendations for the crop model
• After completion of fitting the data under various models found that the best model for this dataset is ’Naive Bayes’ with an accuracy ’0.955’ which shows this model is perfectly fit for the dataset
Model Recommendation on Credit Card Payment Due:
• Made a project where a Credit Card monthly dues dataset was selected and the best model for assessing it was selected after detailed analysis
• Exploratory Data Analysis using all Python libraries such as numpy, pandas, matplotlib, seaborn and a retailed report was finally extracted.
• The best model for this data was ’Random Forest Classifier’ with an accuracy of ’0.813’
• Finally converted the best model into a .pkl file that is downloadable Tesla Stock Visualization Project:
• I used a stock data for Tesla stocks for the year 2010 between the months June, July to determine the various trends in the stocks by EDA( Exploratory Data Analysis) and find various correlations using heatmap and usage of other plots.