Data Visualization With Power-Bi

“Business Intelligence like never before” This was what took over front pages when Microsoft introduced Power-Bi to the world.

Power-Bi is a cloud-based, Business Intelligence Service by Microsoft which provides non-technical Business users with tools for Aggregation, Analysis, Data-Sharing and Visualization.

This software gives users data cleaning abilities as well as interactive data visualization capabilities. It also has a built-in artificial intelligence (AI) tool, the Q&A functionality where you can ask your data simple questions, and Power-BI will try to find the answers for you.

The month was November and we had just concluded our hackathon #30dayswithPowerBi with Cities in Codes where we had to learn about this software and analyze a couple of dataset. During the start of the hackathon, we were split into group. I was in group 3 and each week we got to pick and work on an entirely different dataset from the previous week. We created reports and dashboards using Power BI Desktop and shared on the Power BI Service Platform.

Reports creation is actually easy…import dataset to model, clean data, visualize dataset, and create report… that’s it, the real work is the data preprocessing.

Power-BI Desktop simplifies data evaluation and sharing with scalable dashboards, interactive reports, embedded visuals and more.

It came with a couple of cool features. Yeah, you could import your data from a number of sources like excel, azure, csv, or even scrape off web pages. But that isn’t it, the magic was in the three basic views; the report, data and relationship view. You should check more about the views from here and the numerous cool features about it. Once you are done exploring, processing your data and building your dashboard, all you would need to do is to publish your report on the Power-BI site.

Dashboard building using data visualization is actually the fun and fascinating aspect of data analysis, and Power-Bi had a way of bringing this to life with its captivating patterns and trends.

The more you use it, the better you’ll get at it.

INC 500 Companies data

Dataset Maven analytics. This was the first dataset I got to work on with my group.

The data was about companies on the INC 5000 list in 2019. The dataset had features like the company’s name, industry, founding year, website, and location, revenue as of 2019, % growth and a couple of others. We encountered the problem of the revenue columns since some were in billions and some in millions. We also had changed the formats of most of the columns, and create a new column ‘new_hiree’. This dataset was business inclined and we had to answer a few questions like; the average revenue among companies, revenue by industry, interesting geographic trends, industries with the largest average growth rate.

I explored the MapBox Visual tool to generate a multi-scale analysis: starting with an overview with a cluster map to identify the largest aggregations of companies, then a more detailed layer to explore a few locations. The map view interacts with the other visual elements in my dashboard. That’s beauty of an interactive dashboard.

Dragging, dropping, creating new columns, working with slicers, page filters, or report filters also gave a better sense of your data from different perspectives.

120 years of Olympics history

Dataset Maven analytics.

As a storyteller as well as a data scientist, I’m always curious about how things are connected and how it relates to a wider global context. In this visualization, I explore the last 120 years of Olympics history. Games dating far back to Athens 1896 to Rio 2016. Each row corresponds to an individual athlete competing in an individual event. It included features like the athlete (ID, sex, name, age, height, weight, country) and the event (games, year, city, sport, event, medal).

According to the site’s info. We answered questions like: the percentage of female’s athletes over time, compared summer and winter games, how many athletes, countries and events compete. I got to analyze and visualize country-level trends also as well as who won the most medals along with the trends over time. So, I had to explore these interactions and bring them to life.

This dataset had some duplicate values which had to be dropped, along with a couple of other preprocessing. The analysis did bring a few things to light. The game is male dominated, though female participation has increase since the 1980s. Mid-twenties athletes seems to be the most involved, with the summer game season plateauing and having more games. The number of countries has also increased over the years with almost every country taking part in the summer games. However, countries like the USA, Japan, Germany and Italy are the top 10 with much more participants.

Global Coronavirus (COVID-19) Cases and Deaths

Dataset WHO.

This dataset was interesting though, simple. By simple, I mean the dataset was compiled was actually clean, timely and accurate. I guess this is because of how sensitive the matter was. The dashboard didn’t need to be fancy since it ought to allow generation of answers easier. The trend of the cases made one better understand the infection rate of the coronavirus. I’m fascinated by opportunities to use data to better understand everything in a living, moving way.

Features like the newly reported cases and death along with the cumulative cases and death and the location. In our dashboard we had an overview of the number of global coronavirus cases, confirmed and death and their distribution on a world map. Useful Map box controls like the lasso and geocoding tools let one delve deeper into the data from their specific areas of interest. It gave a dynamic view of the trend that’s constantly changing.

Thank You Note

I am so grateful to have been a part of this hackathon. There is nothing more satisfying then being part of a process from start to finish.

Thanks to Cities in Codes. A special shout out to my team members for making this experience so memorable and motivating.

If you liked this article, click the👏 below so other people will see it here on Medium. Thank you!

Let’s be friends on Twitter. Happy Coding and Learning :)