shujianbu/DJShowcases
DJChina Showcases是配合数据新闻网 2.0上线推出的一个小平台,收集数据新闻及相关行业的优秀作品集,以跟踪国外新闻编辑室以及独立工作室为主,希望提供给从业人员一个快速索引学习的平台。
This is a textual analysis project, fetching stories from the current winners and finalists of the Pulitzer Prizes website (www.pulitzer.org). It read and stored stories from 12 different fields of the finalists between 1995 and 2012 using Python. Topic modeling, Clustering and other techniques are used to do textual data mining. The final presentation is a collection of interactive heat-map and bar charts that shows how journalism pieces have been changing and affecting the society.
Public repository record indexed from GitHub. Explore verified star velocity metrics, source code repositories, and curated developer tool directories across the GitHubRepo ecosystem.
DJChina Showcases是配合数据新闻网 2.0上线推出的一个小平台,收集数据新闻及相关行业的优秀作品集,以跟踪国外新闻编辑室以及独立工作室为主,希望提供给从业人员一个快速索引学习的平台。
Phinch is an open-source framework for visualizing biological data, funded by a grant from the Alfred P. Sloan foundation. This project represents an interdisciplinary collaboration between Pitch Interactive, a data visualization studio in Berkeley, CA, and biological researchers at UC Davis.
This is a project created for The Wall Street Journal. After seeing projects that mapped the workout routes of residents in cities across Europe and the U.S., I felt like doing one for the cities in Asia. Down Jones & Company owns the copyright and reserves other rights.
A Magic Grant project sponsored by the Brown Institue for Media Innovation at Columbia and Stanford University. We are building a system to track the censorship and inappropriate journalistic practice post-publication. We started this project because we want to better inform the readers and involve the public in monitoring online publications, to keep journalism accountable.