The core element of artificial intelligence is fortune telling _ The core of artificial intelligence is algorithm.

What basis does artificial intelligence need?

Artificial intelligence foundation (AI): 1, three core elements-computing power, algorithm and data (three cornerstones): algorithm, computing power and data are the three core elements of AI, which influence and support each other and form different industrial forms in different industries. With the innovation of algorithm, the enhancement of computing power and the accumulation of data resources, the traditional infrastructure will be upgraded intelligently, which is expected to promote the intelligent innovation of various elements of economic development. Let human society enter intelligence from informationization.

(1) Computing power: In AI technology, computing power is the infrastructure of algorithms and data, supporting algorithms and data, and then affecting the development of AI. The size of computing power represents the strength of data processing power.

(2) Algorithm: The algorithm is the "pushing hand" behind AI.

AI algorithm is a data-driven algorithm, which is the driving force of AI.

(3) Data: In AI technology, data is equivalent to the "feed" of AI algorithm.

Both supervised learning and semi-supervised learning in machine learning need to be trained with labeled data, which gives birth to a large number of data labeling companies, which transform unprocessed raw data into machine-recognizable information. Only after a lot of training and covering as many scenes as possible can we get a good model.

2. Technical basis: (1) Artificial neural network after Renaissance.

Artificial neural network is a kind of functional calculus that imitates the operation of neurons. It can accept the stimulation of external information input, and convert it into output response according to the weight of different stimuli, or change the weight structure of internal functions to adapt to mathematical models in different environments.

(2) Machine learning to operate on massive data.

Scientists have found that to make a machine intelligent, it is not necessary to really give it the ability to think. It can read and store a large amount of data, and it has resolution enough to help human work.

(3) The important application of artificial intelligence: natural language processing.

The research of natural language processing is to make machines "understand" human language, which is one of the important branches of artificial intelligence.

Natural language processing can be divided into two types: computer input and computer output.

One is from human to computer-let the computer convert human language into a form that the program can handle;

The second is to give back to people from the computer-to convert the results of computer calculation into a language that human beings can understand.