DEV Community: Rahmah

Zamfara Election Result #outlier_scores

Rahmah — Sun, 07 Jul 2024 17:59:42 +0000

This report is based on analysing the dataset of Zamfara election conducted, the analysis should help uncover any potential irregularities and transparency. Through identification of outlier detection in polling unit indicating potential influences or rigging .
Findings
https://docs.google.com/spreadsheets/d/15xf8NkvJtC29DCqvS2WVIGYe7PeoBRl1Pz0SPLCGEK4/edit?usp=drivesdk google sheet shows the conversation of the address into longitude and latitude.

Findings and methodology
https://drive.google.com/file/d/19gGI_FIyR21o2N6gKhjJ_IZYAwaDIv9G/view?usp=drivesdkrl
shows google sheet report on outlier detection score and calculations of each geodesic distance.Harvesine formula was used to calculate the geodesic distance between the coordinates.

Top Outlier scores
After sorting,
115 for PDP in KETA I / MAKARANTA polling unit and its neightbours are: ['BAKIN GULBI / DANFAKO', 'RUGGAR NA ALI / PRIMARY SCHOOL', 'UNG. SARKIN FAWA / DAN FAKO', 'CHEDIYA / BAKIN KASUWA', 'TABKIN KAZAI II / PRIMARY SCHOOL', 'MARKE / PRIMARY SCHOOL']

89 for PDP in TABKIN KAZAI II / PRIMARY SCHOOL polling unit and its neighbours are : ['KETA I / MAKARANTA', 'BAKIN GULBI / DANFAKO', 'RUGGAR NA ALI / PRIMARY SCHOOL', 'UNG. SARKIN FAWA / DAN FAKO', 'CHEDIYA / BAKIN KASUWA', 'MARKE / PRIMARY SCHOOL']

87.5 for APC in SHAMUSHALE I /SHIYAR SABON GARI polling unit and its neighbours are: ['MADAMBAJI / GARKAR HAKIMI']

An image showing the visualisation of each party votes with its respective outlier score with the help of a seaborn and matplotlib.
Conclusion
The analysis was performed using google sheet,Python with libraries such as pandas, numpy and geopy etc.Geospatial analysis of election data successfully identified polling units with significant deviations in voting results which was further elaborated with the outlier scores

Titanic for HNG

Rahmah — Sat, 29 Jun 2024 22:57:51 +0000

Introduction

This analysis about the dataset of passengers that were onboard when the British Luxury Passenger Ship (named Titanic) sank on the 15th of April, 1912.
The purpose for this surface-level analysis is to have the statistical view of the age-range of the passengers on the ship.

The Titanic dataset has the following columns:

PassengerId: This is the identifier for each passenger
Survived: This informs us if a particular passenger survived (with value 1) or not (with value 0)
Pclass: This is the ticket class, either upper (1st), middle (2nd) or lower (3rd)
Age: The age of the passenger
SibSp: The number of siblings for spouse the passenger boarded the ship with
ParCh: The number of parent or children the passenger boarded the ship with.
Fare: The amount the passenger paid for the trip
Cabin: The cabin identifier for the passenger
Embarked: The port the passenger onboards (either Cherbourg, Queenstown, Southampton)

Observation

Missing Data in the dataset
The columns with missing data are: Cabin, Age and Embarked with Cabin having the highest number of missing values with 687 and Embarked having the lowest with 2 while Age sits in the middle with 177.
All other columns don't have missing value(s).

The image below gives the description of each column

Since our purpose is the age of passengers, it can be seen that:

the minimum age is 0.42
the mean(average) age is 29.70
the maximum age is 80

The below is the box plot of Survived against Age.

About HNG

HNG Internship is a remote internship aimed at providing real life experience on projects to interns all over the world. This is the link to the official website.
It also serves as a pool for skilled professional because the interns there would have experience with real life impacting projects, here is the hire link.