ediwijaya Goto Github PK

analyzing-social-media-data-in-python

Some random codes and stuff of Social Media Analysis of Twitter. It is for my personal learning about Twitter analysis.

blood-donor-prediction-binary-classifier-

In this project, we will predict the whether the person will donor their blood or not based on binary classifier. The datasets model is based on RFM model (RFMTC). The base proportion of donor is is about 76%. However, by using binary classifier, we could predict blood donation by 78.6% (increased by 2.86%).

dashboard

exploring-67-years-of-lego

In this project analyzing the progress of LEGO in past 67 year.

fast-style-transfer

TensorFlow CNN for fast style transfer ⚡🖥🎨🖼

fastbook

Draft of the fastai book

google-play-store-apps-analysis

This project analyze over ten thousand apps on Android App Market on Google Play on the effect of pricing, category, and sentiments toward it (rating and comment). The result shows that user mostly interested in free apps (income mostly comes from advertisements) and app that priced below $30. There's a relation between price and category which shows that medial, business, and family apps are the most expensive (even though there are some apps that mischief or "junk" which excluded from the analysis). Another relationship inferred is that free apps have more tendency to have harsh comment (which inferred low app quality) while paid apps never have that comment (which inferred better app quality).

importing-data-in-python-part-2-

Some methods to import data from Internet using Python 3

international-debt-from-world-bank

This project is to analyze and generate a number of information from World Bank data.

iris_otak_atik

linux-evolution-2018-data

This project will analyze about 17k logs of changes in Linux since April 2005 to late 2017.

real-time-insights-from-twitter

This project is about discovering insights of social media data. The social media used in this project is Twitter. The reason of using Twitter is because of most influential people oare using it. The data used is taken from JSON taken from Twitter. The projects started out with importing and read the data from JSON file. Then, the data is cleaned and taken the most essential parts which are world trends and US trends. Then they are compared to see the common trend which is WeLoveTheEarth hashtag. From this interesting hashtag, I dig a deeper insight by loading more data from Twitter API in JSON file. It was shown that the many big celebrities tweeted this hashtag which leads it to became trend in Twitter. Also, this hashtag is talking about music video of loving Earth by Lil Dicky. Further analysis is conducted. By using Pandas, I convert the retweets file into DataFrame and sort it out. This enable me to easily inferred that the advocation of those celebrity to WeLoveTheEarth trend does not increase their followers number. Another thing to be note is that Lil Dicky's follower is the one who reacted the most of this trends (42.4%). The last but not least thing to be note is the most tweets were English followed by Polish, Italian, and Spanish while there are also some alien (unidentified) language on these tweets.

recommendation-with-ibm

reinforcement-learning-an-introduction

Python Implementation of Reinforcement Learning: An Introduction

stack-overflow-2019-data-analysis

starbucks-capstone-project