OUTCOME PREDICTION
Outcome predictions based on historical data relating to events in a “game” (which can be a sports context, an e-sports contest, a business situation, a medical situation, or other scenarios) can be generated using a watcher process that receives event data and triggers conversion of a “different” last event received into arguments for one or more analysis protocols, which are executed against a data model created by an analysis engine. Methods, systems, articles of manufacture, and the like are described.
1 .- 20 . (canceled)
21 . A computer program product comprising a non-transitory machine-readable medium storing instructions that are configured to, when executed by at least one programmable processor, cause the at least one programmable processor to perform operations comprising:
initiating a watcher process to correspond to a designated event;
based at least on the initiating of the watcher process, automatically pulling, from an event data source via the watcher process, event data, the event data representative of a plurality of occurrences occurring in the designated event, the automatic pulling the event data from the event data source comprising:
periodically generating a request for the event data;
causing transmitting of the request for the event data to the event source, a frequency for the periodic generating of the request for the event data selected based at least in part on a relevant parameter of the designated event such that new event data is generated between a first time of transmitting a first request for the event data and a second time of transmitting a second request for the event data;
determining an occurrence type associated with a most recently pulled occurrence; and
setting the frequency for the periodic generating of the request for the event data based on the occurrence type;
converting the event data for the most recently pulled occurrence of the plurality of occurrences into one or more arguments for one or more analysis protocols;
executing the one or more analysis protocols against a data model created by an analysis engine, the executing comprising the analysis engine predicting an adjusted probability of at least one outcome of an occurrence within the designated event, the predicting reflecting the event data for the most recently pulled occurrence; and
notifying a user application server to update a last occurrence cache accessible by a user application executing on a user machine to display an outcome prediction of the occurrence within the designated event based on the adjusted probability.
22 . The computer program product of claim 21 , wherein the analysis engine predicting the adjusted probability comprises using a tag overlap analysis method to calculate the adjusted probability.
23 . The computer program product of claim 21 , wherein the data source comprises an external event application programming interface.
24 . The computer program product of claim 21 , wherein the operations further comprise tracking, via the watcher process, the plurality of occurrences occurring during the designated event.
25 . The computer program product of claim 21 , wherein the operations further comprise submitting, by a prediction daemon to historical database, new occurrence information comprising at least one of information about the most recently pulled occurrence or analysis results comprising the adjusted probability.
26 . The computer program product of claim 21 , wherein the designated event comprises at least one of a sports contest or a medical condition.
27 . The computer program product of claim 26 , wherein the designated event comprises the medical condition in a patient, and wherein the outcome prediction of the occurrence comprises a predicted result of one or more of a treatment or a change in the medical condition progression.
28 . A computer program product comprising a non-transitory machine-readable medium storing instructions that are configured to, when executed by at least one programmable processor, cause the at least one programmable processor to perform operations comprising:
initiating a watcher process to correspond to a designated event;
based at least on the initiating of the watcher process, automatically pulling, from an event data source via the watcher process, event data, the event data representative of a plurality of occurrences occurring in the designated event;
converting the event data for a most recently received occurrence of the plurality of occurrences into one or more arguments for one or more analysis protocols, the converting occurring based at least in part on the most recently pulled occurrence pulled by the watcher process being new relative to a previously received occurrence;
executing the one or more analysis protocols against a data model created by an analysis engine, the executing comprising the analysis engine predicting an adjusted probability of at least one outcome of an occurrence within the designated game, the predicting comprising a tag overlap method, the adjusted probability reflecting the event data for the most recently pulled occurrence; and
notifying a user application server to update a last occurrence cache accessible by a user application executing on a user machine to display an outcome prediction of the occurrence within the designated event based on the adjusted probability.
29 . The computer program product of claim 28 , wherein the automatic pulling the event data from the event data source comprises:
periodically generating a request for the event data;
causing transmitting of the request for the event data to the event source, a frequency for the periodic generating of the request for the event data selected based at least in part on a relevant parameter of the designated event such that new event data is generated between a first time of transmitting a first request for the event data and a second time of transmitting a second request for the event data;
determining an occurrence type associated with a most recently pulled occurrence; and
setting the frequency for the periodic generating of the request for the event data based on the occurrence type.
30 . The computer program product of claim 28 , wherein the data source comprises an external event application programming interface.
31 . The computer program product of claim 28 , wherein the operations further comprise tracking, via the watcher process, the plurality of occurrences occurring during the designated event.
32 . The computer program product of claim 28 , wherein the operations further comprise submitting, by a prediction daemon to historical database, new occurrence information comprising at least one of information about the most recently pulled occurrence or analysis results comprising the adjusted probability.
33 . The computer program product of claim 28 , wherein the designated event comprises at least one of a sports contest or a medical condition.
34 . The computer program product of claim 33 , wherein the designated event comprises the medical condition in a patient, and wherein the outcome prediction of the occurrence comprises a predicted result of one or more of a treatment or a change in disease progression.
35 . A computer-implemented method comprising:
initiating a watcher process to correspond to a designated event;
based at least on the initiating of the watcher process, automatically pulling, from an event data source via the watcher process, event data, the event data representative of a plurality of occurrences occurring in the designated event;
converting the event data for a most recently received occurrence of the plurality of occurrences into one or more arguments for one or more analysis protocols, the converting occurring based at least in part on the most recently pulled occurrence pulled by the watcher process being new relative to a previously received occurrence;
executing the one or more analysis protocols against a data model created by an analysis engine, the executing comprising the analysis engine predicting an adjusted probability of at least one outcome of an occurrence within the designated game, the predicting comprising a tag overlap method, the adjusted probability reflecting the event data for the most recently pulled occurrence; and
notifying a user application server to update a last occurrence cache accessible by a user application executing on a user machine to display an outcome prediction of the occurrence within the designated event based on the adjusted probability.
36 . The computer-implemented method of claim 35 , wherein the automatically pulling the event data from the event data source comprises:
periodically generating a request for the event data;
causing transmitting of the request for the event data to the event source, a frequency for the periodic generating of the request for the event data selected based at least in part on a relevant parameter of the designated event such that new event data is generated between a first time of transmitting a first request for the event data and a second time of transmitting a second request for the event data;
determining an occurrence type associated with a most recently pulled occurrence; and
setting the frequency for the periodic generating of the request for the event data based on the occurrence type.
37 . The computer program product of claim 35 , wherein the designated event comprises a medical condition in a patient, and wherein the outcome prediction of the occurrence comprises a predicted result of one or more of a treatment or a change in disease progression.
38 . A system comprising:
computer hardware configured to execute at least one computer program comprising a plurality of computerized instructions, the plurality of computerized instructions configured to, when executed by the computer hardware, cause the computer hardware to perform operations comprising:
initiating a watcher process to correspond to a designated event;
based at least on the initiating of the watcher process, automatically pulling, from an event data source via the watcher process, event data, the event data representative of a plurality of occurrences occurring in the designated event;
converting the event data for a most recently pulled occurrence of the plurality of occurrences into one or more arguments for one or more analysis protocols, the converting occurring based at least in part on the most recently pulled occurrence pulled by the watcher process differs from a previously pulled occurrence;
executing the one or more analysis protocols against a data model created by an analysis engine, the executing comprising the analysis engine predicting an adjusted probability of at least one outcome of an occurrence within the designated event, the predicting the adjusted probability comprising a tag overlap analysis method, the adjusted probability reflecting the event data for the most recently pulled occurrence; and
notifying a user application server to update a last occurrence cache accessible by a user application executing on a user machine to display an outcome prediction of the occurrence within the designated event based on the adjusted probability.
39 . The system of claim 38 , wherein the automatically pulling the event data from the event data source comprises:
periodically generating a request for the event data;
causing transmitting of the request for the event data to the event source, a frequency for the periodic generating of the request for the event data selected based at least in part on a relevant parameter of the designated event such that new event data is generated between a first time of transmitting a first request for the event data and a second time of transmitting a second request for the event data;
determining an occurrence type associated with a most recently pulled occurrence; and
setting the frequency for the periodic generating of the request for the event data based on the occurrence type.
40 . The computer program product of claim 38 , wherein the designated event comprises a medical condition in a patient, and wherein the outcome prediction of the occurrence comprises a predicted result of one or more of a treatment or a change in disease progression.