I hold a B.S. degree in Mathematics from Boğaziçi University. After graduation, I enrolled Big Data and Business Analytics Certificate Program organized by Istanbul Technical University and Mindset Institute. Here, I learned more about the business applications and I completed a classification project using Python. I attended Professor Salih Durhan’s Machine Learning lectures at Galatasaray University. Over Professor Durhan’s suggestion, I attended Professor İlker Birbil’s lectures on Machine Learning algorithms covering linear regression, logistic regression, tree models, feature selection, neural networks, etc. It included 14 hours of theoretical lessons and 14 hours of practical training. I am currently enrolled in a master’s degree program in Industrial Engineering at Boğaziçi University. I had always been fond of solving problems using symbolic languages. I am also interested in machine learning and I find the linear algebra behind it fascinating. You can contact me via ilayda.celenk@boun.edu.tr or my Linkedin profile.
In this Youtube video, Steffen Moritz, from the Institute of Data Science of TH Cologne is talking about missing data in time series. He introduces the imputeTS package and suggests that it is useful for dealing with missing data. Features of the package are imputation, visualization, stats and datasets. Imputation functions are used for replacing missing values. It is also possible to compare imputation algorithms. He gives three example functions called ggplot_na_distribution, ggplot_na_imputations and ggplot_na_gapsize. Using these functions visualization and imputation is much easier. It is simpler to adjust the most important parameters compared to ggplot2.
useR! 2020: Visualization of missing data and imputations in time series (S. Moritz), lightning
This is a post from blog.rstudio.com . Lou Bajuk introduces RStudio Cloud. Basically it is an online R Studio. The only requirement is a browser, since there is no need for hardware or installation. There are paid and nonpaid plans for different kinds of user types. It allows users to share their projects easily. Being accesible makes teaching R easy. Members can create projects on their own personal workspace and complete assignments and their progress can also be viewed by instructors.
Do, Share, Teach, and Learn Data Science with RStudio Cloud
This is another post created by Lou Bajuk. Both R an Python languages are used for data science and this has become a huge dilemma. Lou Bajuk explains pros and cons of both. Python is considered as a general-purpose programming language and it has been used for machine learning but R is prefered for most of the introductory courses to statistics and data science. Therefore he suggests that they should work together in order to benefit from both.
R vs. Python: What's the best language for Data Science?.
This is a relatively old post but I believe simple introductions are very crutial to help us understand how the tools work. Tidyverse is package is a set of packages such as ggplot2, dplyr, tidyr, readr, etc. For almost every analysis these packages are used and this makes tidyverse a very collective and useful package. It makes it very simple to install core packages at once. Hadley Wickham explains how to install and loaad tidyverse. He also explains use of some core tidyverse packages as well. It is easy to check updates using tidyverse_update().
Tidyverse 1.0.0.