Recommendations based on predictive model

1. A computer-implemented method for generating options, an option being a set of variables representing a state of a system at a given point of time, the method comprising:

collecting a set of data representing similar events by collecting observations of similar events and/or generating observations of the similar events from a set of simulations, each observation describing an event by a set of variables related to the system, the observations having common variables;

selecting a model capable of receiving inputs and providing output in response to an input, the model being configured for predicting, based on an observation, a target value, a target value being a value of a target variable of an observation, a target variable being a variable providing information for understanding, predicting and/or reacting to behavior of the system;

training the selected model based on the set of data representing similar events;

generating options that represent hypothetical events, each option being a set of variables representing a state of the system at a given point of time;

computing target values by applying the trained model to the generated options;

computing an index for indexing the generated options and target values associated with the trained model;

querying the index for obtaining a selection of a set of options, the selection being performed according to a specific constraint, the constraint being a set of equalities and/or inequalities on values of variables of an option; and

returning, as a result of the query, a subset of the set of the generated options, the subset being ranked according to the target values associated with each option.

2. The computer-implemented method of claim 1 , wherein the set of data representing similar events comprises a set of observations wherein each observation is defined with variables:

identifying target variables on the set of observations, a target variable being a variable;

identifying measurable variables in the set of observations, a measurable variable being a variable that can be measured in a current observation of a current event that is similar to the past events; and

identifying actionable variables on the set of observations, an actionable variable being a variable that can be acted upon based on a decision of a user.

3. The computer-implemented method of claim 2 , wherein generating the options that represent hypothetical events comprises:

defining each variable of each observation as continuous segments or as discrete sets of possible values of the said each variable; and

generating the options by generating values on the defined segments or sets of possible values for the respective variables.

4. The computer-implemented method of claim 3 , wherein generating the options further comprises, for each option to be generated:

selecting an event among the similar events represented by the set of data; and

setting a variable of the option to be generated based on the value of the variable of the selected event.

5. The computer-implemented method of claim 3 , wherein the step of generating values is performed with the use of a random variate generator.

6. The computer-implemented method of claim 2 , comprising:

determining dependencies between the variables of the observations of the set; and

removing events among the generated events whose variables do not satisfy the determined dependencies.

7. The computer-implemented method of claim 6 , wherein one or more variables of one or more observations are transmitted in real time.

8. The computer-implemented method of claim 7 , further comprising:

generating new options from the transmitted one or more variables;

computing new target values by applying the trained model to the new options generated; and

updating the index by indexing the generated new options and the computed new target values.

9. The computer-implemented method of claim 7 , wherein the constraint according to which a selection is performed by querying the index applies to the variables transmitted in real-time and is set using the values transmitted in real-time.

10. A non-transitory computer readable storage medium having recorded thereon a computer program that when executed by a computer causes the computer to implement a method for generating options, an option being a set of variables representing a state of a system at a given point of time, the method comprising:

collecting a set of data representing similar events by collecting observations of similar events and/or generating observations of the similar events from a set of simulations, each observation describing an event by a set of variables related to the system, the observations having common variables;

selecting a model capable of receiving inputs and providing output in response to an input, the model being configured for predicting, based on an observation, a target value, a target value being a value of a target variable of an observation, a target variable being a variable providing information for understanding, predicting and/or reacting to behavior of the system;

training the selected model based on the set of data representing similar events;

generating options that represent hypothetical events, each option being a set of variables representing a state of the system at a given point of time;

computing target values by applying the trained model on the generated options;

computing an index for indexing the generated options and target values associated with the trained model;

querying the index for obtaining a selection of a set of options, the selection being performed according to a specific constraint, the constraint being a set of equalities and/or inequalities on values of variables of an option; and

returning, as a result of the query, a subset of the set of the generated options, the subset being ranked according to the target values associated with each option.

11. A server comprising:

processing circuitry coupled to a memory, the memory having recorded thereon a computer program for generating options, an option being a set of variables representing a state of a system at a given point of time, the processing circuitry implementing the computer program by being configured to:

collect a set of data representing similar events by collecting observations of similar events and/or generating observations of the similar events from a set of simulations, each observation describing an event by a set of variables related to the system, the observations having common variables;

select a model capable of receiving inputs and providing output in response to an input, the model being configured for predicting, based on an observation, a target value, a target value being a value of a target variable of an observation, a target variable being a variable providing information for understanding, predicting and/or reacting to behavior of the system;

train the selected model based on the set of data representing similar events;

generate options that represent hypothetical events, each option being a set of variables representing a state of the system at a given point of time;

compute target values by applying the trained model on the generated options;

compute an index for indexing the generated options and target values associated with the trained model;

query the index for obtaining a selection of a set of options, the selection being performed according to a specific constraint, the constraint being a set of equalities and/or inequalities on values of variables of an option; and

return, as a result of the query, a subset of the set of the generated options, the subset being ranked according to the target values associated with each option.

12. The server of claim 11 , wherein the server is connected to a client computer from which the query on the said index is generated.