This is our first step: the first functional demo of iGlue released! Click here to launch (Firefox and Safari only):
We affectionately call iGlue the internet’s semantic superglue. It is an online interface that is a search engine and content organizer, capable of intelligently rearranging and defragmenting the internet’s broken up data piles, establishing interlinked entities based hierarchies in the process. It organizes interrelated images, videos, people, concepts, notions, ideas and geographical places into cohesive data structures.
The heart of iGlue is what we call the hyper-data model. With the help of hyper-data any text, data, image, video or any other database entry can become a data junction. iGlue reads and understand the web pages meaningful content and then with a single click it organizes and builds an interlinked net of data relating to that chosen content. The building blocks of this data net are the individual data junctions linked together like Lego pieces.
Entities in the thus built multi dimensional data matrix can be called upon either from a text, (moving) image, time or space (map) perspective with ease.
Operation:
In this first iteration of iGlue the following features are available:
Identification of text elements: iGlue recognizes and identifies text within the contents of web pages as illustrated in the demo (in the current release it understand names, geographic positions and institutions in the sample New York times article). To start iGlue’s recognition engine simply click on the green „iGlue annotate” button in the upper right hand corner. - Data linked to entities: Data linked to the identified and highlighted test elements can easily be called up simply by clicking on the highlighted text. This will start the iGlue interface. Once the interface is open, recognized data can also be access from the „Annotated Content” tab.
- Chosen content related data: Content specific, organized data can be viewed in the left hand column of the iGlue interface.
You can chose content specific pictures, videos and geographic position on a map by switching between the right hand side widow’s tabs. This is the media interface.
At the bottom of the interface is a drop down Timetable where all relevant data belonging to the entity can be viewed in sliding timeline. By clicking on an event in the left hand column, that even is highlighted in the timeline, showing its location. Naturally this works in reverse as well. Click on an event in the timeline and see its corresponding entity in the left had column of the interface.
- Connection between the various elements of the interface: By clicking on any of the light green text elements in the left hand column, we can limit the information shown on the map, the timeline and the media interface to that one particular entity. For example if on John F. Kennedy’s data tab you click on the „Death: 22 November 1963, Parkland Hospital, assassination” line you will see on the map the location of Parkland hospital, on the timeline the date and on the media interface the images and videos relating to that event.
- Depth of linked entities: This current release of iGlue provides on one layer of data depth, in other words you can only get information on the entities it recognizes in the text, in this example the New York Times Article. In subsequent releases of iGlue all elements of the chosen entity will be accessible information wise in new data tabs. In other words, staying with the example of Kennedy, in depth data (biographical, image, video and timeline wise) on any member of his family, associate, schools he attended, places he visited, etc.
Application areas of iGlue:
- Information search and organization: Content creators and users can both benefit from high added value through the use of iGlue. In addition, thanks to the locally presented information, there is no need to leave the webpage to do a copy-paste operation in order to find information on the topic of interest. The organized presentation of data can be effective used in any number of fields such as education, research, culture or commerce.
- Targeted advertising: By understanding the text elements themselves better targeted, user interest specific advertising becomes possible.
- Joint utilization of ’info-bricks’: The iGlue semantic database is ideal for collaborative expansion, and thanks to the portability of its data junctions it is an ideal tool for application where repetitive and voluminous data needs to be processed, such as in institutional databases (governmental and corporate), collections (such as libraries) and large databases (geographic, research, etc.).
iGlue can even incorporate the data matrix it builds into a larger network net, what we could call the collective intelligence of the internet. In other words the machine learns from what we post on the web, so that in the end we may learn from it, as presented by the organized coherent content of iGlue.
Our goal is the creation if a tool that allows the creation of an information galaxy in place of a web built of small, individual data-islands. The creation of a tool that is open to and is expendable by all people who value the freedom of the creative process.
Compatibility:
This current release of the iGlue product demonstrator requires one of the following browsers:
- Firefox
- Safari
- Chrome
- Explorer
Current state of completion:
This release of iGlue can be considered an Alpha version (it only hints at many of the functions that will become available shortly.)
In the mean time we would love to receive any and all feedback from users on the current state of iGlue. We are also interested in features requests. Feedback can be posted here in the blog or sent to in4 at in4.hu. Thank you.
Development team
iGlue is developed by in4, Ltd. a Hungarian company. ( www.in4.hu ),
Contact us: in4@in4.hu
in4, Ltd. is backed by Power of the Dream Venture Inc., a public corporation – Ticker symbol PWRV ( www.pdv.hu )
