Can it be done.

Want to do data labeling project, there is no need to pay money to others, otherwise it is thriving leek. In recent years the driverless is very hot, is the high-tech of this century, right, it is how to perceive the complex road conditions like a person? In fact, the reason is very simple, engineers store hundreds of millions of pictures in, and tell it what is on the picture, it will recognize the object.

Scanning an apple, the system retrieves a database at breakneck speed and knows it's an apple. The database is not pre-stored, and behind this high tech is a ponderous feeding of image data to domesticate intelligent machines. This gives rise to an industry: data labeling.

For example, a picture of a red apple into the database, tell the machine that this is the apple, but this is far from enough. Because the machine has no logical thinking, there are green apples, yellow apples, cut in half apples, or flat, round, it will not be able to recognize, which requires a large number of pictures, to enrich the machine's database.

Unmanned driver on the road, identify more things, all angles of all kinds of people, animals, obstacles, signs and so on, to hundreds of millions of labeled pictures. Who will do these simple and boring work? Let the engineers do is the anti-aircraft gun to hit the mosquito, the best way is outsourcing, this pile of pictures you give me well labeled, I pay 10,000 dollars, the right to do.

As long as you can type, you can be competent, you receive the task can also be subcontracted and distributed to others. For example, the mountainous areas with lower wage levels, you can earn the difference, the network world and no restrictions on time and space, even if you lie down in a hot pit head work does not matter.

The two young men took the task in the online platform, and then recruited people in the village to do it, which is to label pictures for big data. For example, labeling cars, trees, obstacles and so on, earn 50,000 a month. Big data requires more picture labeling tasks, such as a supermarket to realize the automatic cash register payment, you need to enter all the commodity pictures in advance, and label the parameters.

Banks want to realize the automation of business, you need to enter a variety of bills, labeled with good information parameters, the machine can automatically identify. Not machine intelligence, it is the people make the machine smarter, AI artificial intelligence has gone through the laboratory process, there will be more AI landing applications in the future, which all need data labeling as the cornerstone of the industry. So, listening to Ma Yun shouted so many years of big data, we can also finally participate in the big data industry.

Only say that the project is good, do not give resources, is like giving smoke does not give fire, I'll share a few more data platforms to you: there is a large number of annotation business above, may know the more people, the more the wind is diluted, enter the industry as early as possible, cultivate the team.