Natural Language Processing

WHAT IS NATURAL LANGUAGE PROCESSING?

Natural language processing is the branch of artificial intelligence which deals with the translating natural language to machine understandable form. Natural language is any language that human being speak like English, German etc. our aim is develop a robot system which understand the natural language and takes act upon that action.

 

WHY NATURAL LANGUAGE?

You may ask why natural language because there is no need to learn natural language everyone can easily speak natural language.

GOALS

• Design a system that can

Our goal is to develop a system which can Interpret the input in natural language and execute upon reach human input. Interactively learn objects, attributes, skills, tasks. Robot should be able to learn object properties like shape, size, color etc. robot should be able to recognize the objects. For example when I give instruction to robot “hey to the kitchen and bring water bottle” in this case robot first find location of kitchen and robot should identify the water bottle.

Examples

Example of this system is wet lab assistant , wet lab is chemistry lab where liquid chemicals are  handled. In wet lab we can use robot to perform experiment first, things can be taught to robot by demonstration.

NL TO ROBOT CONTROL SYSTEM

 

Interactive learning is a broad problem, with components including natural language (NL) understanding, user interface design, active learning, learning by demonstration, gaze and gesture tracking, and probabilistic world modelling. Human instruction-giving is a rich area; modalities such as speech, gesture, gaze, and demonstration are all natural mechanisms by which humans teach, and learn from, one another.. The integration of natural language instruction with teaching language into a formal representation capable of representing a robot and its operation in an environment and, second, mapping the formal representation to actions and perceptions in the real world.

 

INTERACTIVE GROUNDING

Parse language, gestures, gaze, and body motion in to formal presentation. As discuss in above example we can teach to robot how to shake test tube in wet lab using body language. In this case the robot should be able to recognize

 

 

 

 

LEARNING OBJECT ATTRIBUTES

Our goal in this work is to extend the framework described in Sec. II to incorporate visual precepts  in order to learn about completely new colour and shape attributes. This requires taking full advantage of physically grounded sensors and actuators in order to learn about objects in the environment.

 

PARSER

Parsing is the method of determining whether a particular string belongs of CFG. PDA is a machine for  CFG’s. it is procedure that compasses the grammar against input sentence to produce a parsed structure called parse tree.

 

GRAMMAR

 

DEFINITION:

A language can be generated from its grammar G = (V,Σ, S, P).

1. where V is set of variables


2. Σ  is set of terminal symbols, which appear at the end  of generation.


3. S is start symbol.


4. P is set of production rules.


5. The corresponding language of G is L(G).

 

PARSE TREE:

A set of derivations applied to generate a word can be represented as tree.

Natural Language Processing