Discovered public repositories for shujianbu in the GitHub catalog.
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.
A journey of a thousand miles begins with a single step, but starting this week we can march a thousand days forward into a new future. On June 5th, the world will reach a vital moment in history's largest and most successful anti-poverty push -- the 1,000-day mark before the target date to achieve the Millennium Development Goals. The project aims to tell the MDGs story from a more data-driven perspective.
This is a Visual Interface that is trying to solve a journalistic problem - assisting audiences with disabilities to better access news articles online. These people might have difficulty searching, filtering, and reading news on the web. We want to make audible news available to them. Sign Language are detected via Webcam; and the system will recognize different gestures and call Google News API.
What if Facebook's news feed was just that – a stream of information which was purely newsworthy? Posts from reputed news organizations, such as NYTimes, WSJ, and BBC, inform people about what is happening in our world in a simple, uncluttered way. Our team designed The Facebook Tribune – a new way of reading the news in a way that makes more sense to readers. Flash Builder and FB API are the major tools for developing this app.
An introduction of JavaScripts and Github for students at Columbia QMSS's Data Visualization class. Briefing the functionalities provided by github and basics of JavaScript. Presentation using a non-linear Impress.js library. TA materials.
The goal of this project is to take short sequences of visual images, and to determine if the user has placed some body parts in a predetermined sequence of locations and poses. My first combination lock and second creative comb are: 1st - Center Palm + Corner Fist 2nd - Rock-Paper-Scissors Referee.
The main aim of the Mapping Manhattan project was to use Augmented Reality tools to bring to life, the extensive census data that exists on Manhattan. There are stories to be told, detailed information on Race, Education and Wealth to name but three, but the sheer volume of information can be overwhelming. The New York Times created a website for exploring census data through an interactive map, we felt we could go one better and introduce a third axis and new, more sophisticated interaction techniques. So our team at columbia map the data 3-Dimensionally to a 2D map of manhattan and in doing so present the data in a new and meaningful way.
It's a project I did this summer at HKU's Journalism and Media Studies Center. It was just months ahead of China’s biggest political gathering - 18th national congress of the Communist Party of China (CCP) at Beidaihe. How it selects its top leaders seems mysterious to many but is, as always, thoroughly intriguing. The data-mining results cast doubts on some myths about China’s politics and backed up some unspoken beliefs. What’s more, the statistical findings reveal some connectionsbetween some key historical events.
My master's project is an individual work in the form of interactive designs, interviews and analysis, spanning the autumn and spring terms at Columbia Journalism School. Prof. Emily Bell is my supervisor. The project is about the changing history of refugees/asylums in the United States, for the past decades. By employing interactive designs, rather than reading dull tables, this project aims to provide a more intuitive explanation of the refugee subject. In addition, it tries to understand refugees' behavior in choosing residency, from a statistical interpretation.
This small project fetches five articles from both The New York Times and the Wall Street Journal and analyzes the entities, topics, languages, relations, and etc. The tool I used is OpenCalais, which is a rapidly growing toolkit of capabilities that allow people to readily incorporate state-of-the-art semantic functionality within their blog, content management system, website or application. Google has pushed the state of the art in both NER and proposition extraction as part of their Knowledge Graph which extracts structured information from the entire web.
In order to raise the public awareness of the HIV/AIDS prevention and find out effective approaches to alleviate its spreading, our team decides to implement an N-tier web application to visualize related data. By mapping groups of data to a graph, our application should be able to tell a story about recent trends of HIV/AIDS and potential correlated factors. By digging into open data, such as The World Bank data and The Open US data, our application is able to promote government transparency and facilitate the policy makers. Our team plans to design and build a N-tier web application to visualize the HIV prevalence and its correlated factors. The application is an interactive design based on javascript libraries. Users are able to interact with a Choropleth Map page, a Line Chart page, a Bubble Chart page and a customized preference page embedded in this system. By interacting with this application, users can investigate the current trends and prevalence of the global HIV infection.
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.