The selection is made according to the appropriateness of the alteration to the query context (using a bigram language model), or according to its expected impact on the retrieval effectiveness (using a regression model ).

In this paper, we will use a regression model to predict the impact on retrieval effectiveness.

This method develops a regression model from a set of training data, and it is capable of predicting the expected change in performance when the original query is augmented by this alteration.

5.1 Linear Regression Model

The goal of the regression model is to predict the performance change when a query term is augmented with an alteration.

When building prediction models of human judgments using previously proposed automatic measures, we find that we cannot reliably predict human ratings using a regression model , but we can predict human rankings by a ranking model.

We would also want to include more automatic measures that may be available in the richer corpora to improve the ranking and the regression models .

Similarly, when we use previously proposed automatic measures to predict human judgments, we cannot reliably predict human ratings using a regression model , but we can consistently mimic human judges’ rankings using a ranking model.

Some studies (e.g., (Walker et al., 1997)) build regression models to predict user satisfaction scores from the system log as well as the user survey.

In this study, we build both a regression model and a ranking model to evaluate user simulation.

6.1 The Regression Model

Continuous parameters are modeled with a linear regression model (LR), an M5’ model tree (M5), and a model based on support vector machines with a linear kernel (SVM).

As regression models can extrapolate beyond the [0, 1] interval, the output parameter values are truncated if needed—at generation time—before being sent to the base generator.

Table 3: Pearson’s correlation between parameter model predictions and continuous parameter values, for different regression models .