Play semantic web-Introduction

(This is an article posted on BBS. It is pasted as a copy.)

The World Wide Web is undoubtedly a great invention in the 20th century, but it is also flawed. I wonder if you have noticed that many good websites in our school are not easy to find, for example: http://ccst.jlu.edu.cn/jcsb/webroot/index.html. We humans search for information in the form of questions. For example, what is the name of my daughter-in-law, the cousin of my father's mother? (Amount ~, I admit that I am a little spoof), unfortunately, if you do not get the answer in this way, why is this method not feasible, but you must accept the Keyword Method of the traditional search engine. In fact, this is because the current Web infrastructure determines what the above applications are, and the so-called "economic" infrastructure determines the superstructure. Yes, Daniel has long noticed that they have already begun vigorous Semantic Network movements. Here, I plan to use about five articles to briefly introduce semantic Web, let our undergraduate students know this direction. I am new to Semantic Web. If there are any mistakes in the text, I suggest you criticize and correct them. Okay. Let's start the first article.

Oh, by the way, here I want to explain that the following ideas are attributed to Toby Segaran. I just did some localization work, but if it introduced errors (or other problems ), that is my responsibility, and it has nothing to do with Toby Segaran.

========================================================== ==============================================

"People are always asking 'what great technique can I use on this data set? 'When they shocould be asking' what's the best data set I can get? '"
-- Andreas Weigend, former chief scientist at Amazon
 

Introduction -- let's talk about semantics first

Natural language is the language we use to communicate with people in our lives. She is so amazing! You can have a stranger on campus find the building of the Computer College of Jilin University-"qinglou "; you can share your knowledge about music and movies with your friends. You can go to the library, start a book, and learn from it. Even if the author lived for decades, even a hundred years ago (^-^, If you know ancient Chinese ). At this point, some people may say that what is mysterious about natural language can be learned by 3-year-old children. Mysterious? Yes, it's actually mysterious. The reason we ignore her mystery is that we live in a natural language environment. There is a saying: "fish is the last one that knows water ". In fact, there is no better API than natural language.

Here is a simple example. The following are two sentences in the "Main-Sub-object" structure. This is a very simple syntax structure:
1. Li Lei prefers Han Mei.
2. Han Mei loves puppies.

Each sentence here expresses a certain amount of information. The words "Li Lei" and "Han Mei" represent a person respectively. The word "puppy" represents an animal called "Wang" (Note: it is not the one on a certain Snow cake, you can understand, right? If you do not know, enter "wangwangxuebing" on google "). Because you know the meaning of the word "like" (amount ~, You don't know ...), Then, by reading these two sentences, you get some information about the world. With this information, you can now answer some questions, such as "who like puppies ?" (I suddenly remembered that junior high school English listening, ah, I miss it.) This is an example of semantics. Symbols (such as "Li Lei") represent a thing or a concept, A sequence composed of symbols will express a certain meaning.

In an action, semantics is the process of conveying meaning to the result. A symbolic sequence can express a certain meaning (information), and this information will affect our behavior. For example, when you are reading this page, you are integrating the meaning of the text on this page with your previous knowledge. If the meaning of the text I have written is clear, then you may smile occasionally and help you understand the Semantic Web research direction (of course, only a tiny one). You may even choose semantic Web as your future research direction. Then I learned the Semantic Web technology to do some great things (^-^ ).

Although I use natural language as an example, my goal is not to introduce the knowledge of natural language processing. My goal is to introduce how to use semantic information to express, combine, and share knowledge between computers, let machines understand our human knowledge, and let machines help us do more (sorry, we are really lazy ).

If you write a program, even if it has only one variable. Then you will be exposed to semantics. As a programmer, you know that this variable represents a value, and you write a program to respond to changes to these variables. Fortunately, you may have added some comments in your code to explain what this variable represents and how to use it, so that other programmers (including themselves) you can easily understand your code. The relationship between the value, the meaning of the value, and the action of the program is very important, but it is implied in the design of the system.

With a little effort, you can explicitly express the semantic relationship between your data and change the behavior of your system with the meaning of the data. With the explicit expression of semantics, other programs, even those you did not write, can use your data seamlessly (this word is good. Similarly, when your program understands semantic data, your program can process the data, even if the data is not considered when you design your program. (Is it a bit confusing? Well, it doesn't matter. Next, let's look at some simple examples to help you understand what semantic data is)

To be continued ......