Kyso: Blogging Platform for Data Science

Latest users' studies

andrewjoelpeters / holdover

Analysis of grade scores of different classes of 5th and 6th graders in both Math and English-Language-Arts, visualising the data using seaborn boxplots. The author also looks at the difference in grade performance between these students and "holdovers" - those students who are repeating the year.

29Fork

ac / cunianpr_eda

Curtin ANPR is a program for detecting vehicles entering and exiting a parking pay. A camera captures the entry of a vehicle and the exit of the vehicle from a designated parking spot. The entry time and exit time are also recorded. In this study, the author carries out an in-depth analysis of this dataset.

21Fork

Jupyter notebook analyse the sentiment of any twitter

waldohiding / twitter

Application to analyse the sentiment of any Twitter user's input & to carry out a correlation analysis with the price of Bitcoin. The goal is to find users whose actions on Twitter can relatively predict the short-term price of Bitcoin.

42Fork

Jollker / interview_qualogy

Analysis of Greece’s economic recovery from the euro crisis, looking at the relationship between the employment rate and government debt & GDP.

47Fork

Predicting the heart disease prediction using data mining and machine learning notebook

mohamed / heart-disease-prediction-using-data-mining

Predicting the heart disease using data mining, machine learning and Framingham Heart Study dataset. Multiple demographic, behavioural, medical history and physical risk factors are taken into consideration.

124Fork

In-depth statistical analysis bike share notebook

justin / capstone

In-depth statistical analysis of Chicago's Bike Share Program, plotting customers against subscribers, as well as looking at originations and destinations.

45Fork

romankazinnik / adjoint-state-formulas-markdown

A quick demonstration of how to add implicit equations to an optimisation problem, including an example of fitting a model to data, where the model and data are constrained by an implicit equation.

76Fork

bpunt / nba-draft-analysis-which-college-year-does-the-best-in-the-pros

An exploratory analysis of the NBA leage with a single question the goal of this study: are the players who leave early (before their senior year) for the pros better in the NBA than those who stay all four years?

7Fork

jan_erish / unsupervised-learning-hacker-statistics

Unsupervised machine learning model to create a re-election campaign strategy in Washington, with emphasis on the contrast between East and West.

59Fork

Jupyter Visualising Employees with deferred income

pranav_suri / PyData-EDA

This study uses machine learning to build a POI (Person of Interest) identifier based on financial and email data made public in the aftermath of the Enron scandal of the early noughties. The goal of the project is to identify Enron employees who may have committed fraud (POIs) using a Gaussian Naïve Bayes algorithm.

10Fork

jamesle / fifa18

A brief statistical analysis of FIFA 18 data to determine the strongest contenders for this year's World Cup in Russia. The author looks first at interesting stats on individual players, before diving into a more complex analysis of the strongest national squads.

124Fork

beaukramer / Naive Risk Parity

Implementing a simple naive risk parity strategy by proxying the 4 main asset classes used in risk parity strategies. Diversification carried out on this basis applies weights to different asset classes according to the inverse of their volatility.

44Fork

bro / testnotebook

A really interesting exploratory analysis of San Francisco crime, with some really cool visualisations of the various crimes reported. The data set includes both temporal and spatial data.

25Fork

$different mathematical python$

jedhodson / math-assesment-t3

Math-to-python lovers out there. The author attempts to answer two different mathematical assignments in this post; calaculating the volume of vases and determining the decelaration outcomes of cars travelling at various speeds by breaking.

24Fork

kernalcorn82 / fast-for-loop

This brief but incredibly useful post shows us the fastest ways to loop over a pandas dataframe. This is for those of you that work with very large data sets in pandas.

12Fork

And much more ...

Have a jupyter notebook to share?

Or want to create a notebook?

Or fork and extend open-source studies

Latest users' studies

Analysis of grade scores of different classes of 5th and 6th graders in both Math and English-Language-Arts, visualising the data using seaborn boxplots. The author also looks at the difference in grade performance between these students and "holdovers" - those students who are repeating the year.

Application to analyse the sentiment of any Twitter user's input & to carry out a correlation analysis with the price of Bitcoin. The goal is to find users whose actions on Twitter can relatively predict the short-term price of Bitcoin.

Analysis of Greece’s economic recovery from the euro crisis, looking at the relationship between the employment rate and government debt & GDP.

Predicting the heart disease using data mining, machine learning and Framingham Heart Study dataset. Multiple demographic, behavioural, medical history and physical risk factors are taken into consideration.

In-depth statistical analysis of Chicago's Bike Share Program, plotting customers against subscribers, as well as looking at originations and destinations.

A quick demonstration of how to add implicit equations to an optimisation problem, including an example of fitting a model to data, where the model and data are constrained by an implicit equation.

An exploratory analysis of the NBA leage with a single question the goal of this study: are the players who leave early (before their senior year) for the pros better in the NBA than those who stay all four years?

Unsupervised machine learning model to create a re-election campaign strategy in Washington, with emphasis on the contrast between East and West.

A brief statistical analysis of FIFA 18 data to determine the strongest contenders for this year's World Cup in Russia. The author looks first at interesting stats on individual players, before diving into a more complex analysis of the strongest national squads.

Implementing a simple naive risk parity strategy by proxying the 4 main asset classes used in risk parity strategies. Diversification carried out on this basis applies weights to different asset classes according to the inverse of their volatility.

A really interesting exploratory analysis of San Francisco crime, with some really cool visualisations of the various crimes reported. The data set includes both temporal and spatial data.

Math-to-python lovers out there. The author attempts to answer two different mathematical assignments in this post; calaculating the volume of vases and determining the decelaration outcomes of cars travelling at various speeds by breaking.

This brief but incredibly useful post shows us the fastest ways to loop over a pandas dataframe. This is for those of you that work with very large data sets in pandas.