Divisha Choudavaram | A Graduate Student at Northeastern University.

About Me

I am a Data Professional with a Master’s in Data Analytics from Northeastern University and a Bachelor's in Computer Science from India. After completing my Bachelor's, I worked at S&P Global, where I worked with financial services data. My primary responsibilities included providing data solutions through custom SQL queries, transforming financial data from the company’s data warehouses to various systems, and creating dashboards to help stakeholders better interpret the data. Following this, I pursued my Master’s, during which I worked as a Data Analyst at World Education Services. In this role, I provided full-stack analytical solutions, including gathering requirements, developing KPIs, visualizing them through dashboards, and conducting statistical analysis and modeling. Currently, I am working at ML 4 Cyber as a Data Scientist, where I am developing and implementing machine learning models, analyzing cybersecurity data to improve the company's ML tool. Additionally, I participated in various internships and personal projects using a wide range of analytical tools.

Through my experiences and coursework, I have gained proficiency in SQL, Python, R, Tableau, Power BI, Looker, Excel, Spark, Databricks, various SQL infrastructures like MS SQL, MySQL, Oracle, and PostgreSQL, as well as cloud platforms for data transformation, including GCP, Azure, AWS, and Snowflake. I am now seeking a data-oriented role where I can apply the skills I developed through my professional experiences and Master’s degree in a professional setting.

My Resume

Skills

Programming Languages

Data Analytic Tools

Cloud Platforms

Big Data Technologies

My Work

Car Insurance Claims Dashboard

The Car Insurance Claims Dashboard is a Power BI tool designed to provide insights into various aspects of car insurance claims for "AutoGuard." It empowers stakeholders to explore patterns in claims processes by enabling the analysis of key metrics such as Claims Paid, Premiums Earned, and Underwriting Profits, segmented by factors like Region, Gender, Age, Car Make, and Model. The data is cleaned, transformed, and modeled using Power BI features such as Power Query and DAX.

Visit Site →

Spotify Song Analysis Dashboard

The Spotify Music Analysis Dashboard is a PowerBI visualization tool . It is crafted from original Spotify song data sourced directly from the Spotify API. This dataset encompasses song attributes, including popularity, artist information, genre classification, and various key characteristics defining each track. The dashboard unveils insights into factors affecting the popularity of the song and top music trends by genre, artist, and listening patterns. An ETL pipeline is built, which utilizes Python libraries like Pandas and Spotipy to interact with the Spotify API, retrieve data, clean it up, transform it into a usable format, and then load it into an SQLite database. Finally, it's loaded into Power BI for dashboarding to present the insights in a clear and visually appealing way.

Visit Site →

Whatsapp Chat Analyser

WhatsApp Chat Analyzer is a Streamlit app offering insights from WhatsApp conversations. It provides comprehensive analytics on total messages, words, and media shared, identifying busy days and top users. The app employs NLP techniques like Parts of Speech Tagging, Sentiment Analysis, and Topic Modeling for linguistic insights, prevalent themes, and emotional tone. This blend of statistical analysis and NLP provides nuanced insights into communication dynamics and sentiment within WhatsApp conversations.

Visit Site →

NYC Trip Analysis Dashboard

The NYC Trip Analysis Dashboard utilizes trip records data provided by the Taxi and Limousine Commission from 2016 onwards, encompassing a staggering 615 million records. This comprehensive dashboard offers an in-depth analysis of taxi trips in New York City, revealing key insights such as the busiest pickup and drop-off locations, peak travel times, and fare variations throughout the day. By offering this information, the dashboard provides valuable insights into the city's traffic dynamics. Additionally, it conducts a thorough analysis of cab services, including Yellow taxis, Green taxis, and ride-sharing platforms such as Uber, Lyft, Juno, and Via. To achieve this, a big data pipeline was developed, leveraging Google Storage for storing raw data, Apache Spark in Databricks for data processing, Google BigQuery for storing processed data, and Looker for interactive dashboard.

Visit Site →

Credit Card Customer Segmentation

This project aims to analyze credit card customer behavior using machine learning techniques. Initially, K-means clustering is used to segment customers into groups based on their characteristics. Then, various classification algorithms are tested to predict customer behavior, with Gradient Boosting identified as the most accurate model. A user-friendly Streamlit app is developed for users to input their information and receive predictions about their cluster.

Visit Site →

Sales Management Dashboard

A Sales Management Power BI dashboard provides comprehensive insights into sales performance, fulfilling the business request for an executive sales report from sales managers. The dashboard allows for tracking internet sales, identifying top-performing customers and products, and monitoring sales against the budget. It also provides a detailed overview of sales per customer and per product. The data model was built using SQL, including data cleansing and transformation steps.

Visit Site →

Time Series Analysis and Forecasting of US Foreign-Born Population

Analyzed US foreign-born population data from 1860 to 2021, examining various factors such as employment, poverty, education attainment, state-wise distribution, and metropolitan status. Forecasted future trends using ARIMA modeling. Developed a regression model using employment and poverty rates and other factors to predict changes in the foreign-born population w.r.t those factors.

Visit Site →

Data Visualization and Analysis of At-Risk Youth Data

Collaborated with KARA, a non-profit organization, to analyze data on at-risk youth. Using Python, worked on cleaning non-granular, unstructured, and inconsistent data from 22 Excel worksheets. Additionally, utilized Tableau to create an interactive dashboard that provided visual representations of the analyzed data. The analysis of the data revealed specific areas with poor metrics related to child abuse and neglect. These findings were instrumental in supporting KARA's efforts in writing a book that focuses on child abuse and its impact on both children and the communities they live in.

Visit Site →

Contact

If there are any potential opportunities related to Data Analyst, Data Scientist, or Business Analyst positions, please feel free to reach out to me via email or through my provided social media channels below. I am excited to connect with you.