System and method for model-based prediction using a distributed computational graph workflow
A system and method for crowdsensing-based determination of risk management and contract validation, comprising at least a heterogeneous mixture of sensors and data-gathering techniques, a sensor fusion suite, and a business operating system, which ingests, transforms if necessary, and analyzes received data and develops and applies models of prediction of consequences of the sensor data and future events based on such data for purposes such as insurance liability and risk assessment, emergency services planning, and financial market predictions, and comparing historical models and data with current data and models to attempt to refine and utilize a more precise predictive model for these purposes.
1. A system for model-based prediction using a distributed computational graph workflow, comprising:
a computing system comprising at least one memory, at least one processor, and at least one non-volatile data storage device; and
a plurality of network-connected physical sensors;
wherein the at least one memory comprises a first plurality of programming instructions stored in the at least one memory and operating on the at least one processor, wherein the first plurality of programming instructions, when operating on the at least one processor, causes the computing system to:
automatically gather telematics data from the plurality of network-connected physical sensors;
record the telematics data and time of receipt in a multidimensional time-series database stored on the at least one non-volatile data storage device;
retrieve a prediction model from the non-volatile data storage device for prediction of a probability of a future event from the received data;
send the telematics data and the prediction model to a distributed computational graph for processing;
receive processed data from the distributed computational graph;
predict the probability of a future event from the processed data;
determine a premium for an insurance product based on at least the predicted probability of the future event;
generate a specific remedial action based on the predicted probability of the future event, wherein the specific remedial action comprises a recommended operational adjustment for communication to an insured party to reduce the probability of the future event;
transmit the specific remedial action to a notification device associated with the insured party;
after transmission of the specific remedial action to the notification device:
automatically gather updated telematics data from the plurality of network-connected physical sensors;
analyze the updated telematics data to determine whether the recommended operational adjustment was implemented by the insured party; and
determine a new premium for an insurance product based on whether the recommended operational adjustment was implemented, as indicated by the updated telematics data; and
wherein the at least one memory further comprises a second plurality of programming instructions stored in the at least one memory and operating on the at least one processor, wherein the second plurality of programming instructions, when operating on the at least one processor, causes the computing system to:
receive the prediction model for processing;
construct the distributed computational graph from the prediction model using the at least one processor, the distributed computational graph representing a data processing workflow and comprising a directed graph with nodes representing data transformations and edges representing messaging between the nodes;
receive the telematics data; and
process the telematics data according to the data processing workflow represented by the distributed computational graph.
2. The system of claim 1 , wherein the first plurality of programming instructions further causes the computing system to further ingest data from a network-enabled financial market crawler that automatically gathers data from financial markets including stock prices and trades.
3. The system of claim 1 , wherein the first plurality of programming instructions further causes the computing system to further ingest data from manual data entry, including human observers during a natural disaster entering data over a network.
4. The system of claim 1 , wherein the first plurality of programming instructions further causes the computing system to further ingest data from network connected cameras, including either or all of video, audio, and image data.
5. The system of claim 1 , wherein the first plurality of programming instructions further causes the computing system to use machine learning techniques to construct one or more predictive models of received data.
6. A method for model-based prediction using a distributed computational graph workflow, comprising the steps of:
automatically gathering telematics data from a network-connected physical sensor;
recording the telematics data and time of receipt in a multidimensional time-series database stored on a non-volatile data storage device;
retrieving a prediction model from the non-volatile data storage device for prediction of a probability of a future event from the received data;
constructing a distributed computational graph from the retrieved prediction model, the distributed computational graph representing a data processing workflow and comprising a directed graph with nodes representing data transformations and edges representing messaging between the nodes;
processing the telematics data according to the data processing workflow represented by the distributed computational graph;
predicting the probability of a future event from the processed data;
determining a premium for an insurance product based on at least the predicted probability of the future event;
generating a specific remedial action based on the predicted probability of the future event, wherein the specific remedial action comprises a recommended operational adjustment for communication to an insured party to reduce the probability of the future event;
transmitting the specific remedial action to a notification device associated with the insured party; and
after transmission of the specific remedial action to the notification device:
automatically gathering updated telematics data from the plurality of network-connected physical sensors;
analyzing the updated telematics data to determine whether the recommended operational adjustment was implemented by the insured party; and
determining a new premium for an insurance product based on whether the recommended operational adjustment was implemented, as indicated by the updated telematics data.
7. The method of claim 6 , further comprising the step of ingesting data from a network-enabled financial market crawler that automatically gathers data from financial markets including stock prices and trades.
8. The method of claim 6 , further comprising the step of ingesting data from manual data entry, including human observers during a natural disaster entering data over a network.
9. The method of claim 6 , further comprising the step of ingesting data from network connected cameras, including either or all of video, audio, and image data.
10. The method of claim 6 , further comprising the step of using machine learning techniques to construct one or more predictive models of received data.