1.1 Features (metadata):

1. age: age in years
2. sex: sex (1 = male; 0 = female)
3. cp: chest pain type -- Value 1: typical angina -- Value 2: atypical angina -- Value 3: non-anginal pain -- Value 4: asymptomatic
4. rest_bp: resting blood pressure (in mm Hg on admission to the hospital)
5. chol: serum cholestoral in mg/dl
6. restecg: resting electrocardiographic results -- Value 0: normal -- Value 1: having ST-T wave abnormality (T wave inversions )
7. 5. fbs: (fasting blood sugar > 120 mg/dl) (1 = true; 0 = false)
8. thalach: maximum heart rate achieved
9. exang: exercise induced angina (1 = yes; 0 = no)
10. slope: the slope of the peak exercise ST segment -- Value 1: upsloping -- Value 2: flat -- Value 3: downsloping
11. ca: number of major vessels (0-3) colored by flourosopy
12. thal: 0 = normal; 1 = fixed defect; 2 = reversable defect
13. target: diagnosis of heart disease (angiographic disease status) -- Value 0: < 50% diameter narrowing -- Value 1: > 50% diameter narrowing
14. Oldpeak: ST depression induced by exercise relative to rest.

1.2 R code

This is the code that used in data wrangling step before visualising the data

# Load required packages.
library(janitor)
library(tidyr)
library(stringr)
library(readr)
library(forcats)
library(dplyr)
library(tibble)
library(exploratory)

# Set working directory so that the script can read saved data file.
setwd("data path"); jsonlite::toJSON(TRUE)
# Steps to produce the output
exploratory::read_delim_file("https://raw.githubusercontent.com/PacktWorkshops/The-Data-Analysis-Workshop/master/Chapter07/Dataset/heart.csv", delim = NULL, quote = "\"" , col_names = TRUE , na = c('') , locale=readr::locale(encoding = "UTF-8", decimal_mark = ".", tz = "Africa/Cairo", grouping_mark = "," ), trim_ws = TRUE , progress = FALSE) %>%
  readr::type_convert() %>%
  exploratory::clean_data_frame() %>%
    #rename the dataframe for its orginal names
  rename(chest_pain = cp, rest_bp = trestbps, fast_bld_sugar = fbs, rest_ecg = restecg, st_deper = oldpeak, max_hr = thalach, ex_angina = exang, colored_vessels = ca, thalassemia = thal, cholestrol=chol)

2.0.1 Checking the outliers for cholesterol

2.1 cholestrol BoxPlot

In the previous chart, there are a few outliers beyond the 370.

2.2 st_depr

2.3 colored_vessels

2.4 thalassemia

Note: from the preceding boxplots, there are some outliers. However, they will not be imputed as there are few data to start with. i just show them.

---

3.0.1 Age distributions

3.1 Age

We can observe that the youngest patient was 29 years old, while the oldest was 77, and the majority of patients were between 50s and 60s years old. The most common age is 58 years old.

3.2 target count

in the preceding table, 1 means has the disease, while 0 has not. ### Now, I am going to represent them visually with count plot

3.3 target count plot

3.4 sex summary count

3.5 target gpby present absent

From the previous graphs, we can conclude that 72 out of 96 female patients have been diagnosed with heart disease. This scenario is opposite for the male patients, most of them have not been diagnosed with heart disease.

---

4.1 chest pain count

4.2 Presence of Heart Disease by Chest Pain Type

This chart shows us that most of patients have typical angina, and the next is non-anginal pain followed by angina. Most of patients who had typical angina were not diagnose with heart disease. The largest group who had been diagnosed with heart disease had non-anginal pain.

---

5.1 colored vesssels summary

5.2 colored vessels by target

Most of patients who have 0 colored vessels have been diagnosed with heart disease, which implies a strong negative correlation between colored vessels and heart disease.

---

6.1 The total number of each type of slope

6.2 slope summary

## Its distribtion

6.3 heart disease by slope

From the above chart, most patients with a downwards slope of the ST segment post-peak exercise have been diagnosed with heart disease, so there might be a correlation between the two. In addition, most of the patients who had flat slope had not been diagnosed with heart disease.

---

7.1 Creating a scatter plot for detecting the relationship

7.2 Heart Disease based on Age and Maximum Heart Rate gpby target

We can state that although there is no distict pattern as the dots for absent and present are mixed together, quite a few points with a higher heart rate have been diagnosed with heart disease regradless of age. Also, some patients with heart disease might be on the younger side but have a high heart rate.

---

8.0.1 The best course of action is to take a closer look at the relationship between the age and the presence and the absence of the heart disease

We can procced that by categrizing the age group into different bins then plot each group against the presence and the absence of the heart disease.

8.1 presence of heart disease per age group

The previous observation has been confirmed. Some patients who are the youngest have been diagnosed with heart disease when compared to those who have not been diagnosed.

---

9.1 Presence of Heart Disease based on Age and Cholesterol

Again, categorizing the cholesterol against the presence and the abence of

9.2 The presence of heart disease per cholesterol group

There is no much to tell; however, some patient who do have heart disease have high cholesterol. So, cholesterol does not show a strong relationship with heart disease or age.

---

10.1 Visualizing the correlation - Correlations - Heatmap

10.2 Visualizing the correlation - Significance

10.3 Visualizing the correlation - Scatter Matrix

From the previous correlations, slope, maximum heart rate and chest pain have a positive correlation with the target column.