As a master's student studying Applied Data Science at Syracuse University, it was a rewarding experience that challenged my fundamental understanding of how data is essential in today's data economy.

The course focused on the six learning objectives listed below, which were critical in developing the skillset of a competent data scientist.

1. Collect, store, and access data by identifying and leveraging applicable technologies
2. Create actionable insight across a range of contexts (e.g. societal, business, political), using data and the full data science life cycle.
3. Apply visualization and predictive models to help generate actionable insight.
4. Use programming languages such as R and Python to support the generation of actionable insight.
5. Communicate insights gained via visualization and analytics to a broad range of audiences (including project sponsors and technical team leads).
6. Apply ethics in the development, use and evaluation of data and predictive models (e.g., fairness, bias, transparency, privacy).


Scroll to view projects ↓

PROJECT 1

My phone's a switchboard?

Made with:

Azure IoT Hub

Azure Stream Analytics

Azure SQL Database

Power BI

I set out to create a futuristic wonderland of IoT devices. I began by crafting a magical local cloud that would unite all the gadgets and allow them to converse like old friends. The cloud's primary mission was to control lights and switches throughout the kingdom, so I infused it with the power to make bulbs glow and switches click with the flick of a finger. As I completed the project, I couldn't help but grin with excitement, knowing that we had created a system that would turn the mundane into magic with just a tap on a screen.

What I learned:

Using Azure cloud to log the data from the sensors and switches in the network I learnt to Collect, store, and access data by identifying and leveraging applicable technologies.

The IOT data was used to create a forecasting/prediction model that would aid in automation of the device. Through the project it was outside the scope of the project to create an automation, but the model helped with pattern recognization which would have translated to a suggestion shortcut for quick actions in the real world

The end user intended for this project is a normal user with or without technical background. There it was essential to create a visual dashboards that are easy to understand and clear, which taught me to communicate insights gained via visualization and analytics to a broad range of audiences.

PROJECT 2

Keep the guests coming

Made with:

Python

R programming

Seaborn

Our team of data detectives set out to crack the code of hotel guest booking behavior. We combed through mountains of reservation data with the tenacity of a bloodhound, analyzing every booking detail with a hawk-eyed focus. As we sifted through the data, we uncovered some juicy insights that would make any hotel owner swoon. We presented the hotel with our findings and suggested clever strategies to reduce cancellations.

What I learned:

It was a team effort which involved collaborating with peers from diverse background. The skills to communicate and work dispite these differences in experience and background was positive.

Learned how to deliver results to a non-technical audience as part of the final presentation.

Built visually attractive graphs and presentation.

Garnered the skills to perform data collection, cleaning and exploratory data analysis in R.

PROJECT 3

What you selling?

Made with:

Python

Seaborn

Forecasting Algos

The fashionable data scientists set out to solve the mysteries of clothing retail shopping patterns. With the tenacity of a fashionista on a shopping spree, I dug deep into the data, analyzing every stitch and seam of customer behavior. With the help of trusted algorithms, I identified key patterns in customer preferences and even predicted their next purchases. With this knowledge, I could assist the retailer in developing targeted marketing campaigns that would make even the most discriminating fashionista jump for joy.

What I learned:

Deeper understanding of machine learning models to solve real-world problem was attained.

Hands-on with real world data helped understand how to clean noisy data and improve accuracy of models.

Communicate the results to a mix of technical and non-technical stakeholders using visualizations.