Self-learning Databases with Video Compression

Designing self-learning multimedia databases

Application Proposal: 2004

Abstract: A new information theory may result in self-learning Autosophy databases that are highly superior to conventional computer databases. Archiving digital documents and live video now requires huge storage facilities and the complex indexing of stored records. That is a consequence of conventional Shannon data storage techniques in which data is considered to have no meaning. A new Autosophy information theory, in contrast, stores data and images according to meaning or content. The result is very high lossless data compression for text, still images and live video. “Autosopher” are self-learning and no-programming text databases, developed in prior research and available for demonstration. This project would add image and video compression to create multimedia databases. The system acts like a “black box” educated similarly to a human child in that it organizes its own internal data storage. Data storage is hyperspace saturating and self-indexing. The more data already stored in the database the less additional memory space required to store additional data. Using Content Addressable Memories data search access speed is about one microsecond per character or pixel regardless of database size. The technology may evolve into self-learning brain-like Autosopher which could eventually replace the programmed data processing computer. More immediately live video compression would enable secure teleconferencing via the Internet as well as low bandwidth high definition television.

Anticipated Applications: A printed circuit module is being developed that plugs into the PC bus to provide real time secure teleconferencing via the Internet. Most data types (including text, fax, high resolution still images and live video sequences) can be highly compressed and encrypted. The module may also be used for self-learning multimedia databases that are highly superior to conventional computer databases. The new databases could store millions of images or text records with near instant search access time. The technology may evolve into a new generation of brain-like Autosopher that could replace the programmed data processing computer. Intelligent robots with vision and other artificial sense organs could replace human labor. Eventually all human knowledge could be combined into huge self-learning, self-organizing and self- repairing databases. Without the development of the new Autosopher such giant databases may not otherwise be possible.

Keywords: Autosophy, lossless data compression, Multimedia Databases, Archiving

Available downloadable documents:

Proposal document – MS Word doc

Slide presentation / Tutorial – MS PowerPoint ppt

Related Publication – Webpage htm