Wiki

Pre-pruning is a type of pruning algorithm, which refers to the pruning operation before the decision tree is generated.

A positive definite matrix is a symmetric matrix with all eigenvalues greater than zero.

The positive class refers to the expected class in a binary classification problem. The corresponding class is called the negative class.

Relative majority voting is the simplest voting method. In layman's terms, the minority obeys the majority.

Performance metrics are evaluation criteria used to measure the generalization ability of a model.

An ordinal attribute is an attribute whose possible values have a meaningful order or ranking, but the difference between successive values is unknown. It has a sequence of precedence and size.

One-shot learning refers to the ability of a machine to repeatedly work in different environments without prior knowledge of the new environment scenario after a single demonstration.

Different strategies refer to the strategy for generating new samples that is different from the strategy used when the network updates parameters.

Noise contrast estimation (NCE) is a statistical model estimation method proposed by Gutmann and Hyv¨arinen to solve complex computational problems of neural networks and is widely used in image processing and natural language processing.

There is no free lunch (NFL theorem) means that no learning algorithm can produce the most accurate learner in all fields. That is, for problems in a certain domain, the expected performance of all algorithms is the same.

Newton's method, also known as the Newton-Raphson method, is a method for approximately solving equations in the real and complex domains. It uses the first few terms of the Taylor series of the function f ( x ) to find the roots of the equation f ( y ) = 0.

The negative class refers to the class opposite to the positive class in binary classification.

Natural language processing is an interdisciplinary subject involving artificial intelligence, linguistics, computer science and other disciplines. It explores the problem of letting computers process natural language.

Unsupervised learning is a learning method that does not provide corresponding category labels for the training set.

The sample space is the set of all possible outcomes of an experiment or random trial, and each possible outcome in a random trial is called a sample point.

A self-organizing map (SOM) or self-organizing feature map (SOFM) is an artificial neural network (ANN) that uses unsupervised learning to produce a low-dimensional (usually two-dimensional) discretized representation of the input space of training examples.

A recurrent neural network is a network model used to process sequence data. It means that the current output of a sequence is related to the previous output.

The rectified linear unit (ReLU), also known as the linear rectification function, is a commonly used activation function in artificial neural networks, usually referring to nonlinear functions represented by ramp functions and their variants.

Natural language understanding (NLU) is a technology that obtains the semantic representation of natural language through grammar, semantics, and pragmatics analysis. It is an important step in natural language processing.

Natural language generation (NLG) is a technology that studies how to enable computers to express and write like humans. That is, it can automatically generate a high-quality natural language text based on some key information and its internal expression form in the machine through a planning process.

Nash equilibrium, also known as non-cooperative game equilibrium, is an important strategy combination in game theory, named after economist John Nash.

Named entity recognition (NER), also known as "proper name recognition", refers to the process by which a computer recognizes named entities in a text. It is a basic NLP (natural language processing) task.