Matching Techniques –
Matching is the process of checking and comparing two or multiple structures to find their likenesses or comparisons. These structures will represent a wide range of objects including physical things, words or phrases in a language, complete classes of things, common concepts, relations between complex entities, and the like. The representations will show in one or more of the formalisms such as FOPL, any networks, or some other scheme, and matching may involve comparing the component parts of such structures. Matching is used in a wide application of programs for different reasons. It may help to control the sequence of operations, to find or classify objects, to determine the best of a number of different alternatives, or to get items from a database. It is an important operation such complexity programs as speech recognition and natural language concept, vision, learning, automated reasoning, planning things, automatic programming, and domain experts systems, as well as many others. In its simple form, matching is like a process of comparing two structures and also patterns for equality.
The match can fails if given the patterns differ in any point of view. For example, a match between the two strings means character strings acdebfba and acdebeba fails on an exact match since the strings is vary in the sixth character positions. In most complex problem the matching process may give permission for transformations in the patterns in order to achieve a best equality match. The transformation will be a simple change of any variables to constants, or amount to ignoring components during the match operation. For example, a pattern matching variable such a x may be used to permit correct matching between the two patterns (a b (c d) e) and (a b? x e) now by binding? x to (c,d). matching is normally restricted in some way, however, as is the case with the unification of two classes where consistent bindings is permitted.
We will one examples of many problems where correct matches are not good way, and also form of partial matching is more important. In such examples, one can take interest in finding a best match between pairs of structures. This is the case in object classification problems, for example, if object descriptions are subject to corruption by either noise or distortion, at that time, a measure of the degree of match also be essential. Remaining types of partial matching may require to find a match between certain key elements while ignoring all other elements in the pattern. A human language input unit must be flexible enough to find any of the three statements as expressing a choice of preference or for the low-calorie food item.
But in practice, the theoretical requirements for good knowledge representations can usually be achieved by dealing appropriately with a number of practical requirements: The representations need to be complete – so that everything possibly need to represented can easily be represented. It must be computable – implementable with standard computing procedures. It should make the essential objects and relations explicit and accessible –so it is simple to see what is going on, and how the components interact to each other . It should suppress irrelevant detail so rarely used details don’t introduce unnecessary complications, but are still available whenever it is need. They should expose any natural constraints – so that it is easy to express how can first object or relation influences another object. They should be transparent to clear concept – so you can simply understand what is being said and what is right. The implementation of this should be concise and fast – so that data can be stored also you can retrieve it and manipulated rapidly. The four fundamental components of a good representation are as follows-:
The lexical part – It determines the symbols or words which are used in the representation’s vocabulary.
The structural or syntactic part – It describes the constraints on how the best and simple symbols can be arranged, i.e., a grammar.
The semantic part – It establishes a way of associating the term like a real-world meaning with the representations.
The procedural part – It specifies the access procedures that enables best ways of creating and good to modifying representations and answering questions using them, example how we can generate and compute things with the representation.
Knowledge Representation in Natural Language –
Advantages of natural language –
It is very very expressive. We can easily express virtually everything in natural language (realworld situations, ideas, emotions, reasoning, pictures). humans use it as their knowledge of representation of choice
Disadvantages of natural language –
Both the syntax and semantics are very difficult and complex so not fully understood. There small uniformity in the structure of sentences or other things.
Call the Trainer and Book your free demo Class
Call now!!! | SevenMentor Pvt Ltd.
© Copyright 2021 | Sevenmentor Pvt Ltd.