Towards Semantic Web

Semantic Web Big Picture
Introduction
The Semantic Web is an evolving collection of knowledge, built to allow anyone on the Internet to add what they know and find answers to their questions. Information on the Semantic web, rather than being in natural language text, is maintained in a structured form which is fairly easy for both computers and people to work with.


Description

The structuring is simple: knowledge is expressed as descriptive statements, saying some relationship exists between one thing and another. "Jane has a mother, Susan" or "Susan is a mother of Jane". An enormous amount of people's knowledge can be expressed in sentences like these. "Part #4521 has a price, $19.95." "George has a city of residence, Washington D.C." "The United States has a president, Barak Obama"

Names

Of course names like "Jane" do not work very well to identify a specific person in the whole world, so on the Semantic Web we identify each of the things (and the relationships) using more complicated unambiguous names. The names we use are web addresses (sometimes called URLs or URIs). This can cause a little confusion (is "http://www.ibm.com" a company or the company's web site?), but it also turns out to be very useful for locating information about the thing.
We also have special names for text strings (literals), and we allow temporary names which function like pronouns. This lets us write "The country with the ISO country code 'US' has a president who has the name 'Barak Obama"" without ever using an identifier for the US.

Layering

Our system cannot directly express things like "Every person was once an embryo", but it can at least convey expressions in other languages, so people and systems which understand those languages can benefit from the Semantic Web as a kind of shared database. Some information will remain in natural languages, like english, which in some cases is easier for people to use. Other information may be stored in languages designed primarily for computers to understand.

Logic

Some facts (like "Mary is a mother" and "A mother is a kind of parent"") lead logically to other facts ("Mary is a parent"). That is often intuitive to people, but can be very hard to explain to a computer. When properly programmed, however, computers can be very helpful in figuring out which facts follow logically from other facts. 
A precise explanation of one's terms and reasoning in some subject area, which can allow computers to help, is called an ontology. Ontologies can be expressed in various languages and carried by the Semantic Web. With them, computers can sometimes act as if they "understand" the information they are carrying. This is where the term "semantic" comes in; on this web, we try to make the meanings so clear that even a computer can understand them.
No logic languages have yet been recommended for the Semantic Web. Some of the experimental languages are RDF Schema, DAML+OIL, and swap/log.
Executable content (computer programs and subroutines) can be considered expressions in a "Turing complete" logic language. Swap/log is both Turing complete and fundamentally modular/reusable; it may turn out to be an essential core in practical Semantic Web clients.

Contributions

The knowledge on the Semantic Web is an aggregate of contributions from many sources, much as the Web is an aggregate of many web sites. The answers to a question you ask, like the results of a web search, will depend to some degree on who you trust, which systems are working at the time you ask, and what kind of search techniques are being used.
As you browse information, a good user interface will make it easy to correct inaccuracies and add your own knowledge as you like. Your additions will be stored in various configurable ways and can be kept private or published as openly as you choose.

Change-Over-Time


The facts stored in the Semantic Web will change over time as the true state of world changes. With each contribution signed and dated, we have a good chance of piecing together a correct and up-to-date picture.
The important point is that people's contributions are factual descriptions asserted to be true at some point in time. Queries are not really "What is the name of the president of the United States?" but "What is the name of the president of the United States according to the latest trustworthy data you have?"

Services

How do you buy a book over the Semantic Web?
  1. You browse/query until you find a suitable offer to sell the book you want.
  2. You add information to the Semantic Web saying that you accept the offer and giving details (your name, shipping address, credit card information, etc). Of course you add it (1) with access control so only you and seller can see it, and (2) you store it in a place where the seller can easily get it, perhaps the seller's own server, (3) you notify the seller about it.
  3. You wait or query for confirmation that the seller has received your acceptance, and perhaps (later) for shipping information, etc.
This approach allows automation of the process, detailed record-keeping, and excellent process abstraction.

Applications

The sky is the limit. What application wouldn't work better with a global knowledge-base under the hood? Well, some might be slower. :-)


Comments

Post a Comment

Popular posts from this blog

Dependency Injection in ASP.NET MVC 5 with StructureMap

Dependency Injection(DI) is a design pattern that let you write loosely coupled testable code. Typically your classes depend on other classes. Take a look at the following example. The above example shows the " HomeController " of an asp.net mvc application. Within the index method, you can see that it requires to create an instance of " Student " class to get the full name of a student. It is very clear that the Index method is tightly coupled with the Student class.  This is a disadvantage when it comes to testing HomeController . Suppose that you have other methods, with similar types of class dependencies. So, each time when you test the code, you have to pass an actual instance Student class. This is tedious. You don't have the luxury of passing a fake student class to easily test your controllers.  So how do we decouple the Student class from HomeController? This is where DI comes in handy. You can inject dependencies in several ways. Today i will sho
Read more

Real-time web with Node.js & Socket.io

Image
If you are a .net developer, you may be familiar with ASP.net SignalR for real-time web development. It's one of the coolest libraries for ASP.net that allows bi-directional communication between a server and a client. For more details refer  http://www.asp.net/signalr .  This blog post is an intro to use Node.js and Socket.io to achieve real-time web communication with pure JavaScript using a simple chat program example. Socket.io is a library that makes our lives easy to work with web sockets. Of course, in the Javascript world, there are hundred other libraries available to implement web sockets,  examples .  But i like the simplicity of Socket.io.  Step 01 : Setting up Node.js Depending on the operating system you are using, download Node.js from the official site  here . Installation process of both Windows and OSX is pretty much the same. Double click the setup file and follow the straightforward instruction. For Linux, either you can do
Read more

First Look at AngularJS 2.0

Image
AngularJS 2.0 developer preview has been released! This blog post is to experience the new features in the framework with a simple angular2 application. Heads Up : AngularJS 2.0 is written in TypeScript. TypeScript is a super-set of ES6 which is a super-set of ES5 which is currently the browsers fully supported version of JavaScript.  So when writing angular2 components, we can use choose among TypeScript, ES6, ES5 or Dart if you will. For this blog post I use TypeScript. Please refer  5 Min Quick Start Guide  to setup angular2 development environment. Once you have setup the environment and followed the guide, you will be able to see a similar result as follows. Now let's add few more features to this. We will create a shopping list where a user can view the list, add a new item or remove a item from the list. So to begin, let's create the listItems array. Change the app.ts file as follows. Let's create the list.html file which is the view for this c
Read more