Prasanna Chandran

Results-driven Lead Data Analyst with expertise in transforming complex raw data into actionable insights using SQL and Python. Skilled in developing dynamic Tableau and Power BI dashboards to drive data-driven decision-making. Known for uncovering key data trends and developing innovative solutions to solve business problems.

Capstone project in Supply Chain Management domain:

Obtained material purchasing data for a real time manufacturing company. Identified their  monthly material usage. Forecasted material usage by applying Machine Learning with Hierarchical  Project Approach: Data Cleansing | Missing Value | Outlier Treatment | Exploratory Data Analysis |  Feature Engineering | Model Building |Hyperparameter Tuning | Model Validation | Forecast |  Time Series. Later, suggested safety stock levels to be maintained by the company.   Snowflake, MYSQL, MS SQL, POSTGRESQL  properties of sale and factors affecting it. Tableau’s data visualization tools were used to create  Models: HTS Forecasting: ARIMA, SNAIVE, ETS, NNETAR, XGBOOST, CATBOOST, LightGBM. Based on model results, final forecasts were made using catboost model. The models are validated using the metrics: MAPE, MAE, RMSE, SMAPE.  

Salary Prediction using ML & deploying the model in Heroku (End to End):

Data cleansing | EDA | Feature Engineering | Predictive model using XGBoost | Deploying the model on Heroku using Flask API.

House Price : Advanced Regression Techniques:

Date Cleaning | Missing value & Outlier Value treatment | Exploratory Data Analysis &visualizations|  One hot encoding | Feature Engineering | Hyper parameter Tuning | Lasso, Ridge, Elastic Net  Regression, Support Vector Regression, Gradient Boosting Regression, LGBM, XGBoost, Bagging &  Stacking regressor.

Web scrapping with Python:

Web scrapped online bookstore using python - Selenium | Requests | BeautifulSoup | Scrapy | xml.

Factor Analysis on Consumer Perception towards Cereal Brands:

Consumer perceptions towards a brand can be built on various parameters depending upon the  product. In this project, I have used Factor Analysis to evaluate and identify the most significant attributes that contribute towards the consumer perception of the different cereal brands.

Visualizing House Sales for Boston Real Estate in Tableau:

In this project, I have explored the art of problem-solving with the aid of visual analytics to identify properties of sale and factors affecting it. Tableau’s data visualization tools were used to create interactive dashboards to uncover hidden insights for sellers of the properties as well as prospective  buyers. https://public.tableau.com/profile/prasanna3111#!/vizhome/BostonCondostorybyPrasanna/BostonCondo-StorybyPrasanna

Evaluated impact of change in temperature of Cold Storage products during the year:

In this project, I have computed the mean cold storage temperature in different seasons and used  hypothesis testing to evaluate if any corrective action is required from the plant's side or the  procurement's side.

Evaluated and inferred significant predictors affecting the customer loan preferences:

In this project, I have found Education and Income are the most significant predictors affecting the  customer response on seeking personal loan for ‘Thera bank’. Business strategies can be made by  concentrating more on these variables to increase the potential customers seeking personal loan.

Created multiple dashboards using Power BI:

Performed ETL & data visualization using Power BI | Extracted data to power BI by connecting to  multiple Microsoft office tools, SQL | Transformed data using Power Query | Performed Data Modelling using DAX functions, measures, mfunction and connecting multiple tables. My first blog on power BI is featured by Analytics-vidhya in their medium publication. https://medium.com/analytics-vidhya/a-complete-beginners-guide-on-power-bi-ed7d54a0b73e