UK Road Safty Accidents 2015 Project

Author: ptyadana

Database Clinics - MySQL/02.UK Road Safty Accidents 2015/01.UK_Road_Safty_Accidents_Schema_Tables_Load_Data.sql

@@ -0,0 +1,52 @@
+CREATE SCHEMA accidents;
+
+USE accidents;
+
+/* -------------------------------- */
+/* Create Tables */
+CREATE TABLE accident(
+	accident_index VARCHAR(13),
+    accident_severity INT
+);
+
+CREATE TABLE vehicles(
+	accident_index VARCHAR(13),
+    vehicle_type VARCHAR(50)
+);
+
+/* First: for vehicle types, create new csv by extracting data from Vehicle Type sheet from Road-Accident-Safety-Data-Guide.xls */
+CREATE TABLE vehicle_types(
+	vehicle_code INT,
+    vehicle_type VARCHAR(10)
+);
+
+/* -------------------------------- */
+/* Load Data */
+LOAD DATA LOCAL INFILE 'C:\\Users\\Accidents_2015.csv'
+INTO TABLE accident
+FIELDS TERMINATED BY ','
+ENCLOSED BY '"'
+LINES TERMINATED BY '\n'
+IGNORE 1 LINES
+(@col1, @dummy, @dummy, @dummy, @dummy, @dummy, @col2, @dummy, @dummy, @dummy, @dummy, @dummy, @dummy, @dummy, @dummy, @dummy, @dummy, @dummy, @dummy, @dummy, @dummy, @dummy, @dummy, @dummy, @dummy, @dummy, @dummy, @dummy, @dummy, @dummy, @dummy, @dummy)
+SET accident_index=@col1, accident_severity=@col2;
+
+
+LOAD DATA LOCAL INFILE 'C:\\Users\\Vehicles_2015.csv'
+INTO TABLE vehicles
+FIELDS TERMINATED BY ','
+ENCLOSED BY '"'
+LINES TERMINATED BY '\n'
+IGNORE 1 LINES
+(@col1, @dummy, @col2, @dummy, @dummy, @dummy, @dummy, @dummy, @dummy, @dummy, @dummy, @dummy, @dummy, @dummy, @dummy, @dummy, @dummy, @dummy, @dummy, @dummy, @dummy, @dummy, @dummy, @dummy, @dummy, @dummy, @dummy)
+SET accident_index=@col1, vehicle_type=@col2;
+
+
+LOAD DATA LOCAL INFILE 'C:\\Users\\vehicle_types.csv'
+INTO TABLE vehicle_types
+FIELDS TERMINATED BY ','
+ENCLOSED BY '"'
+LINES TERMINATED BY '\n'
+IGNORE 1 LINES;
+
+/* -------------------------------- */

Database Clinics - MySQL/02.UK Road Safty Accidents 2015/02.UK_Road_Safty_Accidents_Create_severity_median_Table.sql

@@ -0,0 +1,4 @@
+CREATE TABLE accidents_median(
+vehicle_types VARCHAR(100),
+severity INT
+);

Database Clinics - MySQL/02.UK Road Safty Accidents 2015/03.motorcycle_accidents_median_severity_calculation.py

@@ -0,0 +1,51 @@
+# UK Road Safty 2015 data
+import pymysql
+
+myConnection = pymysql.connect(
+    host="localhost", user="root", password="root", db="accidents")
+
+cur = myConnection.cursor()
+
+cur.execute(
+    "SELECT vehicle_type FROM vehicle_types WHERE vehicle_type LIKE '%torcycle%';")
+
+cycle_list = cur.fetchall()
+
+
+selectSQL = ('''
+SELECT vt.vehicle_type, a.accident_severity
+FROM accident a
+JOIN vehicles v ON a.accident_index = v.accident_index
+JOIN vehicle_types vt ON v.vehicle_type = vt.vehicle_code
+WHERE vt.vehicle_type LIKE %s
+ORDER BY a.accident_severity;
+''')
+
+
+insert_SQL = ('''INSERT INTO accidents_median
+VALUES(%s, %s);''')
+
+
+for cycle in cycle_list:
+    cur.execute(selectSQL, cycle[0])
+    accidents = cur.fetchall()
+
+    # calculate median severity
+    # divide the length of accidents /2 to find the median of accdients list
+    quotient, remainder = divmod(len(accidents), 2)
+
+    if remainder:
+        # meaning odds number of items in accidents list
+        median_severity = accidents[quotient][1]
+    else:
+        # even numbers of items in accidents list
+        median_severity = (accidents[quotient]
+                           [1] + accidents[quotient + 2][1]) / 2
+
+    print("finding Median Severity for ", cycle[0])
+
+    # insert the calculated median severity into table
+    cur.execute(insert_SQL, (cycle[0], median_severity))
+
+myConnection.commit()
+myConnection.close()

Database Clinics - MySQL/02.UK Road Safty Accidents 2015/04.UK_Road_Safty_Accidents_Data_Analysis.sql

@@ -0,0 +1,35 @@
+/* SIDE NOTE: Compare performance of the query rows by using Explain Icon first (Before Indexing and After Indexing)*/
+
+/* Create index on accident_index as it is using in both vehicles and accident tables and join clauses using indexes will perform faster */
+CREATE INDEX accident_index
+ON accident(accident_index);
+
+CREATE INDEX accident_index
+ON vehicles(accident_index);
+
+
+/* get Accident Severity and Total Accidents per Vehicle Type */
+SELECT vt.vehicle_type AS 'Vehicle Type', a.accident_severity AS 'Severity', COUNT(vt.vehicle_type) AS 'Number of Accidents'
+FROM accident a
+JOIN vehicles v ON a.accident_index = v.accident_index
+JOIN vehicle_types vt ON v.vehicle_type = vt.vehicle_code
+GROUP BY 1
+ORDER BY 2,3;
+
+/* Average Severity by vehicle type */
+SELECT vt.vehicle_type AS 'Vehicle Type', AVG(a.accident_severity) AS 'Average Severity', COUNT(vt.vehicle_type) AS 'Number of Accidents'
+FROM accident a
+JOIN vehicles v ON a.accident_index = v.accident_index
+JOIN vehicle_types vt ON v.vehicle_type = vt.vehicle_code
+GROUP BY 1
+ORDER BY 2,3;
+
+
+/* Average Severity and Total Accidents by Motorcyle */
+SELECT vt.vehicle_type AS 'Vehicle Type', AVG(a.accident_severity) AS 'Average Severity', COUNT(vt.vehicle_type) AS 'Number of Accidents'
+FROM accident a
+JOIN vehicles v ON a.accident_index = v.accident_index
+JOIN vehicle_types vt ON v.vehicle_type = vt.vehicle_code
+WHERE vt.vehicle_type LIKE '%otorcycle%'
+GROUP BY 1
+ORDER BY 2,3;