Friday, October 31, 2014

About Python

Python is a widely used general-purpose, high-level programming language. Its design philosophy emphasizes code readability, and its syntax allows programmers to express concepts in fewer lines of code than would be possible in languages such as C++ or Java.
In the class exercise, I wrote my first phython program, analysis the words frequency of a text. Do not need to load lots of extra packages, just several lines of program, the problem is solved. That made me fully felt the convenience of using python as the tool to programming.

Python is also the most popular programming language now. Python interpreters are available for installation on many operating systems, allowing Python code execution on a majority of systems. I prefer to using Ecplise.

Friday, October 17, 2014

Social Media Analytics Based on Human Behavior

I started learning the knowledge about data mining when I was a junior student. And since then, I thought data mining technology is based on mathematics, whatever it is used to do the social network recommendation system or movie rating system. I believed that there may be a significant number of programmers who have the same point of view as me.

The situation had continued to the time that I had the first lesson of social media analytics. This wonderful course lead me to the brand new field of data mining. For instance, if we want to do a social network friends recommendation, we need to analysis the behavior of users before we use the formulaic algorithm to analysis those users' data. Social media analysis should be based on the human behavior.

Friday, October 3, 2014

Sentiment Classification and Human Emotion

Maybe the most complex thing in the world is human emotion. It is also the most difficult part of Artificial Intelligence.
Kobian, a "humanoid" robot, which can express human emotions, has been unveiled by researchers at Waseda University in Japan.


This robot can simulate human emotions, but only seven, including happiness, 
anger, sadness, cheering, sorrow, fear and shock. However, human emotions can not be classified simply as this. Human emotions are often consist of several kinds of feelings. Studying the consumer sentiment could contribute to the discovering of the potential profits. A growing number of businesses are tracking social-media outlets such as Facebook and Twitter to analyze the consumer.

Sentiment classification can be divided into three levels, attribute level, sentence level and document level. We could analyze the polarity of the sentence through the dictionary-based method. Then we could analyze the polarity of a text. The polarities can be divided into three kind, negative, positive and neutral. Then we could get people's attitude towards something.

Saturday, September 20, 2014

Big Data VS the Human Brain

With the birth of ENIAC in 1946, a new era has arrived!
In the past 70 years, there has been a significant number of controversies around the question, which is more intelligent,human brain or computer. In the early decades, computer was just used to calculate, just like ENIAC,was initially designed to calculate artillery firing tables for the United States Army's Ballistic Research Laboratory. Though it had a speed of one thousand times that of electro-mechanical machines, people still did not think it can go beyond the wisdom of human, it was just a tool.

In 1996, IBM released a chess-playing computer named Deep Blue. This machine beat the world champion Garry Kasparov on May 11, 1997 in a chess competition. It is the first time that machine intelligence beyond the wisdom of human. However, that proved nothing, due to the factor that Deep Blue was just developed to playing chess. In another word, Deep Blue could not think and solve any other problem but playing chess.


The most important thing we could acquire from Deep Blue is that when a computer stores enough amounts of knowledge, it could make predictions about the matter through a program that pre-defined by human,similarly to human thinking mechanism. This is the 'Big Data'!

The machine could not simulate a human thinking, but it could obtain the basic rules from mass data through data mining. Those basic rules without incidental personal emotions and prejudices are much more closer to the truth than human thinking.

It is the era of Big Data, a better time!