In many areas of information science, finding predictive relationships from data is a very important task. Initial discovery of relationships is usually done with a training set while a test set and validation set are used for evaluating whether the discovered relationships hold. More formally, a training set is a set of data used to discover potentially predictive relationships. A test set is a set of data used to assess the strength and utility of a predictive relationship. Test and training sets are used in intelligent systems, machine learning, genetic programming and statistics.
Regression analysis was one of the earliest such approaches to be developed. The data used to construct or discover a predictive relationship are called the training data set. Most approaches that search through training data for empirical relationships tend to overfit the data, meaning that they can identify apparent relationships in the training data that do not hold in general. A test set is a set of data that is independent of the training data, but that follows the same probability distribution as the training data. If a model fit to the training set also fits the test set well, minimal overfitting has taken place. A better fitting of the training set as opposed to the test set usually points to overfitting.
In order to avoid overfitting, when any classification parameter needs to be adjusted, it is necessary to have a validation set in addition to the training and test sets. For example if the most suitable classifier for the problem is sought, the training set is used to train the candidate algorithms, the validation set is used to compare their performances and decide which one to take, and finally, the test set is used to obtain the performance characteristics such as accuracy, sensitivity, specificity, F-measure and so on. Another example of parameter adjustment is hierarchical classification (sometimes referred to as instance space decomposition ), which splits a complete multi-class problem into a set of smaller classiﬁcation problems. It serves for learning more accurate concepts due to simpler classiﬁcation boundaries in subtasks and individual feature selection procedures for subtasks. When doing classiﬁcation decomposition, the central choice is the order of combination of smaller classiﬁcation steps, called the classiﬁcation path. Depending on the application, it can be derived from the confusion matrix and, uncovering the reasons for typical errors and finding ways to prevent the system make those in the future. For example, on the validation set one can see which classes are most frequently mutually confused by the system and then the instance space decomposition is done as follows: firstly, the classification is done among well recognizable classes, and the difficult to separate classes are treated as a single joint class, and finally, as a second classification step the joint class is classified into the two initially mutually confused classes.
Use in artificial intelligence, machine learning, and statistics
In artificial intelligence or machine learning, a training set consists of an input vector and an answer vector, and is used together with a supervised learning method to train a knowledge database (e.g. a neural net or a naive Bayes classifier) used by an AI machine.
In statistical modeling, a training set is used to fit a model that can be used to predict a "response value" from one or more "predictors." The fitting can include both variable selection and parameter estimation. Statistical models used for prediction are often called regression models, of which linear regression and logistic regression are two examples.
In these fields, a major emphasis is placed on avoiding overfitting, so as to achieve the best possible performance on an independent test set that follows the same probability distribution as the training set.
Use in intelligent systems
In general, an intelligent system consists of a function taking one or more arguments and results in an output vector, and the learning method's task is to run the system once with the input vector as the arguments, calculating the output vector, comparing it with the answer vector and then changing somewhat in order to get an output vector more like the answer vector next time the system is simulated.
- Cohen S, Rokach L., Maimon O. Decision-tree instance-space decomposition with grouped gain-ratio In J. Information Sciences, vol. 177, issue 17, pp. 3592–3612. Elsevier. 2007.
- Sidorova, J., Badia, T. "ESEDA: tool for enhanced speech emotion detection and analysis". The 4th International Conference on Automated Solutions for Cross Media Content and Multi-Channel Distribution (AXMEDIS 2008). Florence, November, 17-19, pp. 257–260. IEEE press.