If I ignore my first ever ambition of becoming a Fireman as a child; the rest of my aspirations have a creative angle and one thing I always imagined myself doing was showcase my work. With some inspiration from others, I decided to create a portfolio of my Power BI work. By doing this, I am able to learn tonnes as I continue this journey as well as look back.
So here it is! click on the image to be redirected to the dashboards. I hope you all like it 🙂
K-Neighbours Regression Prediction of Diamond Prices
The implementation of Machine Learning has become fun (once i started to understand it more!), and I have recently found a strong interest in implementing it more not only in Power BI, but also in Python too. So here it is! a dashboard which uses all elements! Using the Python visuals in Power BI, I was able to create (and calculate) a K-Neighbours Regression model, which predicts the prices of diamonds 🙂
Linear Regression Model: Salary Prediction
Continuing on with my Machine Learning journey within Power BI, I decided to build a simple linear regression model which predicts salary based on your years of experience. The dataset was from Kaggle 🙂
Live Crypto dashboard
Crypto has been a huge phenomena for the past few years and continues to grow by the second. Using the crypto watch API, I built a live crypto dashboard which updates on a daily basis. I focus on the top 20 biggest cryptos based on market cap. Trend analysis can be conducted based on different Cryptocurrencies and within different periods.
Mental health in tech
After reading several articles regarding the correlation between mental health issues and tech companies, I decided to conduct my own analysis. Using a Kaggle dataset based on US respondents to a mental health survey, there has been an increase in mental health issues in tech; most particularly in companies where benefits are obsolete.
Anime database dashboard
Calling all Anime fans! (or those looking to get into anime 🙂), I created this dashboard which provides all anime dating all the way back to 1963 and is sorted by the rating. The data can be sliced based on source (whether it derives from a manga, novel or musical etc), type (whether it’s a film, show or musical etc), studios and when it premiered. From this dashboard, I hope you find a good anime to dive into! 🙂
Live Stock Analysis
Hi all, I decided to take a short break from Power BI, but I am back and will be posting more regularly. This dashboard is a stock analysis conducted using a live API known as Tiingo. The data itself will refresh automatically on a daily basis.
Machine Learning – Purchase Intent Prediction
Machine Learning (ML) has become a prominent method of data analysis in recent years. It has not only been extremely interesting to see how it can be implemented within various different industries, but also exciting since it seems to be evolving at a rapid rate. One thing I was always wanted to do was use the ML functionality within Power BI – and here it is! the first of many reports in which I will create using ML. This report in particular uses ML to predict a myriad of different possible revenue outcomes from customers!
Adult Census Report
I conducted analysis based on the adult census data. From this dashboard, you can delve into income based on marital status and native countries, as well as the disparities that exist within both race and gender. The data can be sliced even further to provide even deeper analysis.
ESG Benchmark Report
Environment, Social & Governance (ESG) has increasingly become a topic within the financial world, with large corporations beginning to invest more into it. As an ESG Analyst, it has been extremely interesting delving into this topic – most particularly the rating system. Anyone working within this field will tell you – The measurement system itself can be deemed flawed since it is not necessarily robust enough to properly gauge ESG investment. As a result, using a public dataset via a financial API, I decided to create an ESG benchmarking report which focuses on comparing a client’s ESG investment amongst its peers. The quick facts page provides an overview on how much the client is investing in ESG, along with their peers. This is split by region, country and city (drilldown available) and via Holder style and Turnover (click on text to toggle visual). Also, you can see the increase (or decrease) in ESG investment from March 2020 to April 2021. The weightings Analysis is all of the ESG investment within a client. The cards at the top showcases the amount of funds, investment and average ESG fund investment. However, the key figure is the purchasing potential. This is free float and market cap adjusted in order to provide an apple for apples comparison. Further, it provides a client insight on how much their peers can potentially invest in them. However, in this case, the purchasing potential is underweight; meaning there is more investment within the client than in their peers. The ESG Global targeting tab is like a shopping list – it provides a client with all of the institutions and funds that can potentially invest in them; these are all sorted by purchasing potential. Additionally, Artificial Intelligence is embedded to show who and where has the highest purchasing via a decomposition tree. This analysis was extremely interesting especially when looking at ESG investment thorough active and passive funds. Lastly, there is a wide array of filters and drilldowns that can enhance finding answers from ESG investment to contact details.
Top 10 largest internet companies
This new dashboard shows the revenue amassed by the top 10 largest internet companies. I did this by creating my own visual using Charticulator; I definitely will explore this feature more. Also, the data is available from this link; List of largest Internet companies – Wikipedia 🙂
Global Landslides between 1988 and 2016
I decided to get back into my love for geography by exploring landslides that have occurred globally from 1988 to 2016 (note: I am currently in the process of finding up to date data for this dashboard). I really like how there is loads of data to play with, allowing the use and experimentation of different cool visuals (and maps!).
Pulling from Netflix data, this dashboard shows the total amount of films released onto the streaming platform from 2008 – 2020 (note: 2021 data is available, however, I decided to omit it since it would not be full year data). As an avid Netflix-watcher, I thoroughly enjoyed creating this dashboard; it was really interesting to see the gradual increase in films within their ratings. Also, as a finishing touch, I couldn’t create this dashboard without somehow including one of my favourite shows 🙂
Change in Political sentiment
This dashboard shows the gradual change in sentiment by musicians towards political candidates, most particularly Trump and Hillary Clinton. It dates back from 1989 to 2016. It is extremely interesting to not only see how both Trump and Hillary influence the different themes their names come under, but how the disdain towards Hillary slows down over the years, whilst negative sentiment against Trump takes a huge spike during his presidential campaign.
Stocks and Share Price – Live update