Visual Data Exploration:

Today, Internet has become an inseparable part of our day today life.  When a person starts a new project or plans to buy something, the first thing is to search the Internet. Searching Internet is an exploring task. Internet provides all sorts of information and people have to dig useful information from the unlimited data sources. Search engine has become an inseparable part of the internet to achieve this goal. Through key words, a search engine can help people find relevant data source. However, people needs to organize the search results and documents collection process.

Let’s take a close observation on a data exploration process. Assume that a PhD student John Smith starts to prepare a research project “Interactive Visual Explorer”. He will start with a set of categories: goals, problems, approaches, technologies, research directions, and applications. He will further make the details of each category. For example, the technologies can be further divided as: visualization, human computer interaction, data retrieving, Internet searching, and data mining. He will start to use search engine to find papers and documents with the combination of the key words mentioned. John will scan the documents and mark or copy the relevant documents from the Internet. After reading documents, John will further refine other categories. Taking “goals” as instance, he makes following list: humans involved, interactively refine selection criteria, automatic identify potential artifacts (data source for Internet searching), easy for human to select artifacts, organize identify artifacts and free to add new notes in various forms.

Popular search engines like Yahoo! and Google helps us in finding the relevant information from the internet but the results they produce is totally unorganized. Search results should be organized in a way for the users to identify the relation between each individual result. In order to achieve this goal, textual representation of search results won't be an efficient way to represent information. This research aims at solving this problem by visually representing the search results and organizing the results based on how they are related with each other.

Internet data exploration is one of the potential application that could be developed with visual exploration. Other applications are brainstorming, system design, knowledge management, program comprehension, decision-making, biomathematics, and social network.


Yahoo! Visual Search:

Yahoo! Visual Search is a research project of Yahoo! aimed at solving the problem discussed above. This is my personal project working for the upcoming Yahoo! Hack Day. Rest is confidential : )

Sample Screenshots of my Research Prototype: