Systems and methods for multi-source reference class identification, base rate calculation, and prediction
AssigneePytho, LLC
InventorsPavel Atanasov, Regina Joseph
Life of this patent
Estimating the term…
Abstract
Systems and methods for multi-source reference class identification, base rate calculation, and prediction are disclosed. The systems and method can guide on, then elicit, information about reference class identification on a case-by-case basis, connects to a database in order to calculate historical base rates according to user reference class selections, and collect additional quantitative and qualitative information from users. The systems and methods can then generate predictive estimates based on the combination of the inputs by one or more users.
Claims23
A method comprising:
- receiving a data structure associated with a first database, the data structure being separate from the first database;
- configuring a system executed by a processing device based on the data structure to associate a question with the database, generate reference class information, and generate one or more graphical user interfaces specific to the question and the reference class information;
- receiving, via the system, selection of at least one of reference class criteria or historical reference cases associated with reference class information independently from a plurality of users via a first graphical user interface of the one or more graphical user interfaces, the reference class criteria being generated in response to configuring the system;
- generating, separately and automatically via the system, one or more queries based on the reference class criteria or the historical reference cases to separately define a reference class for each of the plurality of users, the system generating, for each of the plurality of users, at least one of narrower queries and increasing specificity of the reference class as each additional reference class criterion is selected by each of the plurality of users or increasingly broader queries and increasing generalization of the reference class as additional historical cases are selected by each of the plurality of users;
- determining the reference class for each of the plurality of users in response to querying the first database using the one or more queries separately generated for each of the plurality of users;
- identifying a number of cases in the reference class for each of the plurality of users in response to querying the first database using the one or more queries;
- calculating a base rate estimate for each of the plurality of users based on the reference class separately defined by each of the plurality of users, the base rate estimate corresponding to at least one of (a) a frequency of historical event occurrence, (b) time-specific probabilistic estimates derived from survival analysis, or (c) descriptive statistics for a continuous variable associated with the reference class;
- prompting each of the plurality of users to manually adjust the base rated estimate for the question, the manual adjustment accounting for deficiencies in the first database with respect to variables for defining the reference class;
- generating at least one of an aggregated probabilistic estimate or an aggregated quantitative estimate for the question from the base rate and a final probabilistic estimate generated for each of the plurality of users for the question;
- displaying the base rate estimate associated with the reference class or an overall base rate across a plurality of selected different reference classes.
A method comprising:
- determining, for each of a plurality of users, a reference class to which a new case in question belongs in response to receiving inputs associated with reference class information from each of the plurality of users via a first graphical user interface, the inputs including at least one of reference class selection criteria producing a smaller and more specific reference class as each additional reference class criterion is added, or selecting historical reference cases producing a larger and more general reference class as additional historical cases are added;
- querying a first database for each of the plurality of users in response to determining the reference class for each of the plurality of users to update a number of cases in the reference class generated for each of the plurality of users;
- calculating a base rate estimate, the base rate estimate corresponding to at least one of (a) a frequency of historical event occurrence, (b) time-specific probabilistic estimates derived from survival analysis, or (c) descriptive statistics for a continuous variable associated with the reference class;
- generating at least one of an aggregated probabilistic estimate or an aggregated quantitative estimate for the question from the base rate and a final probabilistic estimate generated for each of the plurality of users for the question;
- displaying the base rate estimate associated with the reference class or an overall base rate across a plurality of selected different reference classes;
- and
- storing a reference class definition, the base rate estimate and a final predictive estimate for the user in a second database residing on one or more servers.
A system comprising:
- a non-transitory computer-readable medium storing instructions;
- and
- a processing device programmed to execute the instructions to:
- determine, for each of a plurality of users, a reference class to which a new case in question belongs in response to receiving inputs associated with reference class information from each of the plurality of users via a first graphical user interface, the inputs including at least one of reference class selection criteria producing a smaller and more specific reference class as each additional reference class criterion is added, or selecting historical reference cases producing a larger and more general reference class as additional historical cases are added;
- query a first database for each of the plurality of users in response to determining the reference class for each of the plurality of users to update a number of cases in the reference class generated for each of the plurality of users;
- calculate a base rate estimate, the base rate estimate corresponding to at least one of (a) a frequency of historical event occurrence, (b) time-specific probabilistic estimates derived from survival analysis, or (c) descriptive statistics for a continuous variable associated with the reference class;
- prompt each of the plurality of users to manually adjust the base rate estimate for the question, the manual adjustment accounting for deficiencies in the first database with respect to variables for defining the reference class;
- generate at least one of an aggregated probabilistic estimate or an aggregated quantitative estimate for the question from the base rate and a final probabilistic estimate generated for each of the plurality of users for the question;
- display the base rate estimate associated with the reference class or an overall base rate across a plurality of selected different reference classes;
- and
- store a reference class definition, the base rate estimate and a final predictive estimate for the user in a second database residing on one or more servers.
A non-transitory computer-readable medium comprising instructions that when executed by a processing device causes the processing device to:
- determine, for each of a plurality of users, a reference class to which a new case in question belongs in response to receiving inputs associated with reference class information from each of the plurality of users via a first graphical user interface, the inputs including at least one of reference class selection criteria producing a smaller and more specific reference class as each additional reference class criterion is added, or selecting historical reference cases producing a larger and more general reference class as additional historical cases are added;
- query a first database for each of the plurality of users in response to determining the reference class to update a number of cases in the reference class generated for each of the plurality of users;
- calculate a base rate estimate, the base rate estimate corresponding to at least one of (a) a frequency of historical event occurrence, (b) time-specific probabilistic estimates derived from survival analysis, or (c) descriptive statistics for a continuous variable associated with the reference class;
- prompt each of the plurality of users to manually adjust the base rated estimate for the question, the manual adjustment accounting for deficiencies in the first database with respect to variables for defining the reference class;
- generate at least one of an aggregated probabilistic estimate or an aggregated quantitative estimate for the question from the base rate and a final probabilistic estimate generated for each of the plurality of users for the question;
- display the base rate estimate associated with the reference class or an overall base rate across a plurality selected different reference classes;
- and
- store a reference class definition, the base rate estimate and a final predictive estimate for the user in a second database residing on one or more servers.
Dependencies are inferred from the claim text.