Personal Scalable Solutions Using Machine Lerning

It is absolutely clear that the world is full of structure. In photographs and video, objects, scenes, and events repeat over and over. Objects and scenes, in turn, have structures that characterize them objects have parts and parts have sub-parts. Many current automatic approaches to index visual information rely on specific algorithms constructed by experts. The goal of such algorithms (Visual Detectors) is to automatically label, or index visual content. While these algorithms can often be robust, it is clear that it is not possible to construct a large number of detectors by hand.

Therefore, for future applications it is desirable to build systems that allow the construction of programs that learn, from user input, how to automatically label data without the need of experts. Not only should such systems be scalable but they should also take into consideration users’ interests and subjectivity. This requires recognizing the structure inherent in visual information and exploiting such structure in computational frameworks. Without a doubt machine learning will form the basis of successful, scalable applications in the future. Those applications will change, and the way such algorithms learn will change. Perhaps learning will take place without explicit input from users.

Sensor Networks and its Node tracking scenarios

The tracking of a tagged object through a region of space monitored by a sensor
network. There are many situations where one would like to track the location
of valuable assets or personnel. Current inventory control systems attempt to
track objects by recording the last checkpoint that an object passed through.
However, with these systems it is not possible to determine the current location
of an object. For example, UPS tracks every shipment by scanning it with a
barcode whenever it passes through a routing center. The system breaks
down when objects do not flow from checkpoint to checkpoint. In typical work
environments it is impractical to expect objects to be continually passed  through
checkpoints.

With wireless sensor networks, objects can be tracked by simply tagging them
with a small sensor node. The sensor node will be tracked as it moves through
a field of sensor nodes that are deployed in the environment at known locations.
Instead of sensing environmental data, these nodes will be deployed to sense
the RF messages of the nodes attached to various objects. The nodes can be
used as active tags that announce the presence of a device. A database can be
used to record the location of tracked objects relative to the set of nodes at
known locations. With this system, it becomes possible to ask where an
object is currently, not simply where it was last scanned.

Unlike sensing or security networks, node tracking applications will continually
have topology changes as nodes move through the network. While the connectivity
between the nodes at fixed locations will remain relatively stable, the connectivity
to mobile nodes will be continually changing. Additionally the set of nodes being
tracked will continually change as objects enter and leave the system. It is essential
that the network be able to efficiently detect the presence of new nodes that enter
the network.