One Size Fits All

// Just found this article I wrote for Linkage long back, with some colleagues in NIIT.

One Size Fits All


Gaurav Bhatnagar, Ritu Dangwal, Renu Gupta, N.N. Ramanathan, Himanshu Tayal

Refer as: Gaurav Bhatnagar, Ritu Dangwal, Renu Gupta, N.N. Ramanathan, Himanshu Tayal , One Size Fits All, Linkage, Volume 6.2, Summer 1998

The Cognitive Engineering Forum (CEF) is a small group in NIIT, searching for ways to make intelligent anthropomorphic interfaces. Currently we are working on the design and construction of an emerging cognitive agent. In the last CEF meeting, the authors were trying to design an algorithm for an emerging learning system, that not only appears to have some chaotic human ability, but is also a strategic learning system.

This kind of system is not new to the R&D labs. The program Fluke, now being tested in many CEG centers as an electronic classmate for NIITians, is a good example of the emerging randomized architecture that we wish to build upon. Fluke is an advanced anthropomorphic system, much, much better than the program Eliza (e.g. see the web-site:

The basic algorithm that we developed turned out to have other applications also. The first implementation of the emerging randomized paradigm, or ERP for short, turned out to have applications in other areas of the lab too. Before going on to the details of the program itself, we will spend some time in looking at the other applications of ERP.

For example, the infant learning multimedia software developed at the R&D lab is an interesting new concept, creating a multimedia learning software for children just born, or about to do so. This software is targeted strategically at new parents, and their new children, and is not the result of corporate randomized management, as it first appears to be. Tests on a three year old child has shown that it aids in the teaching of important motor skills, that are absolutely essential in the IT professional of tomorrow. For example, the youngest known child in the universe who can handle an input device named after the small mammal called "mouse" is a graduate of the infant learning software called MGR.

Another example is of a product developed in the R&D lab of NIIT is the advanced cognitive perception system (code named psycho-mouse). The basic idea is that the chaotic human engineering, especially that of the brain, still sends some recognizable signals, that can be used as an input device for the computer. The feasibility of improving the device to handle intelligent NetCentric perception is being studied. If developed, this device could have an interesting consequence: Email writers will not be able to hide behind the networks, and their thoughts will be bared for all to see.

Finally, the application that is closest to those of us who have an interest in improving humanity for generations to come by means of fundamental research. A random response from the ERP can set our minds thinking for titles of new papers to be written, and further researches in its emerging transformational ability can help us fill in the details. Indeed we almost did that in this paper, except that we changed the title. This title was chosen after the strategic learning methodology of ERP was identified.

This is partly true because of the nature of the human brain itself. We will not go into the theory of how it happens. Suffice it to say that Margaret Boden, of "Creativity" fame has said it all. As Lewis Carroll did not say: When in doubt, trot out Boden. This last statement is false.

I think by now you (the reader) would be really raring to get your hands, or other advanced learning resources, on the program. The basic idea is simple; we will leave the implementation to you.

Choose one word from each of the columns appearing in Table 1.







Table 1

Type it in a suitable text editor, or a typesetting package like MS-Word, and add a few words so the sentence formed appears to make sense. Now try very hard to interpret this sentence. You will find that it is not difficult to make some sense, or at-least some nonsense, out of it. This is due to the properties of the mind, covered in any philosophy of computer science book. If this doesn't work, you probably need to see a computer scientist or a robotist, and re-take the Turing test.