I graduated from Ondokuz Mayıs University Mechanical Engineering department in 2018. I learned the Python programming language in college. This was the breaking point of my career because it led me to explore the data industry. After graduating, I worked as a business intelligence consultant at Ereteam business intelligence consulting firm in 2019. I got my first professional experience here. During my 6-month working period here, I was trained in data processing and visualization tools. I was assigned to a project at Yapı Kredi Bank and was involved in data modeling and data transfer processes there. In 2020, I started working as a data engineer at Eren Retail company and I worked there about 1.5 years. During my time there, I was involved in implementing data warehousing and reporting systems from scratch on Azure Cloud Platform and the contribution of this to my career development has been enormous. The system I installed here was shared as a sample success story on the Microsoft Turkey website. In 2021, I graduated from Marmara University Management Information Systems master’s program.Now i work as a Business Intelligence Specialist at Solvoyo firm.
In this video, the lecturer shows the project he had done using R. Projects main subject is to analyze the non-emergency requests of city of Chicago.
In short , residents of Chicago can submit requests for non-emergency services, like maintanence, complaints about their living areas etc. This submit requests data is provided freely by Chicago Data Portal. The lecturer fetches that data everyday from API using RSocrata library after that he uses tidyverse package for manipulation and aggregation purposes.He built a interactive dashboard using shiny that includes a map(leaflet) and beautiful user interface that end users can easily filter, select area for explore service requests using that app.
App can be found here : Shiny App
Finally, he mentioned that while he is doing that project he learned new things about his city, new tools thanks to the Chicago Data Portal.Also he said app he built could easily be adapted to other locations with similar infrastructure.
Based on this project I think R is well suited for these types of projects.Easily reproducible reports and shareable contents using just a few packages are awesome.
Mapping 311 Requests Using Shiny, Leaflet, and The Tidyverse
Time Series is a sequance of data being recorded at specific time intervals. Time Series Forecasting uses these data points to make future predictions.
There are many forecasting applications such as ,
Anyone who has historical data is suitable for time series can analyze that data using time series analysis methods and model, forecast, predict.
There are multiple Time Series Forecasting Methods,
In this post, author predicted the sale of airline tickets using the ARIMA model in R.
R Shiny framework can be used to build interactive web applications with R. Using R shiny we can create highly effective reports and visualizations where users can explore a data set.
R shiny is a great framework for building simple web applications without using JavaScript.It has a great user interface, pre-defined widgets for plots, tables etc and works in any R environment.
There are 2 main components in Shiny App structure.
1.UI
UI is basically a frontend that controls what is being displayed, how the components placed etc. In short this component displays how app is presented.
2.Server
Server can be think of as a backbone of shiny app. The server will be where you load and wrangle data before displayed in UI.
In this post the author teaches how to create a shiny app from stratch and publish it on shinyapps.io.
K means clustering is a unsupervised learning algorithm. Unsupervised means that there isn’t any label attacted to data.Main purpose of clustering is to find out similar patterns in the data and group them.In real world scenarios clustering can be used cases like customer segmentation, recommendation engines etc.
In this post K Means Clustering is used to cluster flower species based on similarities between them.Therefore if new data point comes we can easily identify which cluster it belongs to.