Publications by Selcuk Disci
Plotly for Data Visualization: The Vaccinations Effect on Covid-19
One of the most frequently asked questions these days is whether the vaccine works in reducing the number of cases and deaths. In order to examine that we will build a function so that we can run for every country we want without repeating the same code block. The dataset we’re going to use for this article is from Our World in Data. What we wa...
1663 sym R (2861 sym/3 pcs) 4 img
Wildfires Comparison with ggplot2 dual Y-axis and Forecasting with KNN
In recent days, Turkey has been escalated wildfires. The government and the people were In recent days, Turkey has been escalated wildfires. The government and the people were on the alert and mobilized to put out the fires but, they couldn’t for days. A significant part of people thought that the simultaneous fires had no make sense, and there...
3348 sym R (3741 sym/6 pcs) 12 img
ARIMA Method from {fable}: The Election is Coming for Turkey?
Nowadays, every journalist and intellectual talks about a probable early election in Turkey’s ongoing poor economic conditions. But, is it politically right decision to go early election before the officially announced 23 June 2023 in terms of ruling parties? In order to answer this question, we have to choose some variables to monitor economic...
4660 sym R (4462 sym/11 pcs) 14 img
Dynamic Regression with ARIMA Errors: The Students on the Streets
The higher education students have had trouble being housing in Turkey in recent days. There have been people who even sleep on the streets like a homeless. The government has been accused of investing inadequate dormitories for sheltering the students. Let’s examine the ongoing sheltering problem of students. The dataset we have built for this...
3219 sym R (3551 sym/6 pcs) 12 img
Lagged Predictors in Regression Models and Improving by Bootstrapping and Bagging
Huge lines at gas stations have been seen around Turkey in recent days. Under the reason for this is the rise in the prices so often, in the last few months and the expectation to continue this. Of course, it is known that the rise in the exchange rate (US dollar to Turkish lira) has an important role in the hikes in gas prices. Considering the e...
2247 sym R (3030 sym/8 pcs) 12 img
Comparing Decision Trees
In the last article of the current year, we will examine and compare some of the tree algorithms for the classification. The dataset we are going to use for this will be the answers given to the loan applicants and their evaluated features for it. library(tidyverse) library(tidymodels) library(tsibble) library(C50) library(caret) library(pROC) d...
2282 sym R (3026 sym/5 pcs) 10 img
Association Rules with Interactive Charts
Until today, we have examined the supervised learning algorithms; but this time, we will take a look at a different learning method. The algorithm we just mentioned is association rules which is an unsupervised learning method. The algorithm is referred to as market basket analysis as it usually has been applied to grocery data. The market basket...
3991 sym R (1954 sym/3 pcs) 16 img
Meta-Learning: Boosting and Bagging for Time Series Forecasting
I am always struggled to model the changes in gasoline prices as a categorical variable, especially in a small amount of time-series data. The answer to improving the performance of modeling such a dataset can be to combine more than one model. This method of combining and aggregating the predictions of multiple models is called meta-learning, wh...
3352 sym R (2219 sym/3 pcs) 12 img