END USER DEVICE (EUD) 103 is any computer system that is used and controlled by an end user (for example, a customer of an enterprise that operates computer 101) and may take any of the forms discussed above in connection with computer 101. EUD 103 typically receives helpful and useful data from the operations of computer 101. For example, in a hypothetical case where computer 101 is designed to provide a recommendation to an end user, this recommendation would typically be communicated from network module 115 of computer 101 through WAN 102 to EUD 103. In this way, EUD 103 can display, or otherwise present, the recommendation to an end user. In some embodiments, EUD 103 may be a client device, such as thin client, heavy client, mainframe computer, desktop computer and so on.

REMOTE SERVER 104 is any computer system that serves at least some data and/or functionality to computer 101. Remote server 104 may be controlled and used by the same entity that operates computer 101. Remote server 104 represents the machine(s) that collect and store helpful and useful data for use by other computers, such as computer 101. For example, in a hypothetical case where computer 101 is designed and programmed to provide a recommendation based on historical data, then this historical data may be provided to computer 101 from remote database 130 of remote server 104.

PUBLIC CLOUD 105 is any computer system available for use by multiple entities that provides on-demand availability of computer system resources and/or other computer capabilities, especially data storage (cloud storage) and computing power, without direct active management by the user. Cloud computing typically leverages sharing of resources to achieve coherence and economies of scale. The direct and active management of the computing resources of public cloud 105 is performed by the computer hardware and/or software of cloud orchestration module 141. The computing resources provided by public cloud 105 are typically implemented by virtual computing environments that run on various computers making up the computers of host physical machine set 142, which is the universe of physical computers in and/or available to public cloud 105. The virtual computing environments (VCEs) typically take the form of virtual machines from virtual machine set 143 and/or containers from container set 144. It is understood that these VCEs may be stored as images and may be transferred among and between the various physical machine hosts, either as images or after instantiation of the VCE. Cloud orchestration module 141 manages the transfer and storage of images, deploys new instantiations of VCEs and manages active instantiations of VCE deployments. Gateway 140 is the collection of computer software, hardware, and firmware that allows public cloud 105 to communicate through WAN 102.

Some further explanation of virtualized computing environments (VCEs) will now be provided. VCEs can be stored as “images.” A new active instance of the VCE can be instantiated from the image. Two familiar types of VCEs are virtual machines and containers. A container is a VCE that uses operating-system-level virtualization. This refers to an operating system feature in which the kernel allows the existence of multiple isolated user-space instances, called containers. These isolated user-space instances typically behave as real computers from the point of view of programs running in them. A computer program running on an ordinary operating system can utilize all resources of that computer, such as connected devices, files and folders, network shares, CPU power, and quantifiable hardware capabilities. However, programs running inside a container can only use the contents of the container and devices assigned to the container, a feature which is known as containerization.

PRIVATE CLOUD 106 is similar to public cloud 105, except that the computing resources are only available for use by a single enterprise. While private cloud 106 is depicted as being in communication with WAN 102, in other embodiments a private cloud may be disconnected from the internet entirely and only accessible through a local/private network. A hybrid cloud is a composition of multiple clouds of different types (for example, private, community or public cloud types), often respectively implemented by different vendors. Each of the multiple clouds remains a separate and discrete entity, but the larger hybrid cloud architecture is bound together by standardized or proprietary technology that enables orchestration, management, and/or data/application portability between the multiple constituent clouds. In this embodiment, public cloud 105 and private cloud 106 are both part of a larger hybrid cloud.

According to the present embodiment, the impact function generation program 150 may be a program capable of receiving input data associated with a target geospatial climate hazard and a corresponding asset, the input data comprising one or more of a first dataset corresponding to a predetermined set of prompts, and a second dataset corresponding to a natural language exchange. Impact function generation program 150 may then generate, based on the first dataset an entity knowledge graph including a series of candidate variables related to the target geospatial climate hazard and the corresponding asset. Next, impact function generation program 150 may generate, based on the second dataset, a universal knowledge graph including a series of candidate function formulas that are most associated with the target geospatial climate hazard and the corresponding asset. Impact function generation program 150 may then generate, using a graph neural network, embeddings corresponding to the entity knowledge graph and the universal knowledge graph respectively. Thereafter, impact function generation program 150 may perform symbolic regression, using the embeddings, to generate one or more impact functions for the target geospatial climate hazard and the corresponding asset. In turn, impact function generation program 150 has provided for improved generating of impact functions for geospatial climate hazards based on user interactions by leveraging a wide breadth of data that includes data received from a user. Described embodiments leverage data including integrated knowledge and observational data, previously modeled data, geospatial context data, and impact formulations from literature to increase the reliability, accuracy, and adaptability of generated climate impact formulas. Described embodiments further provide the ability to generate new model iterations to derive most accurate hazard intensity. Accordingly, presently described embodiments employ methods of generating impact functions for geospatial climate hazards based on user interactions that provide several advantages over conventional modeling platforms such as, for example, increased flexibility towards users interests, robust and transparent risk evaluation, improved efficiency and scalability by automating impact function generation, capability for leveraging fine-grained spatial variability, improved continuous learning by leveraging experience from previous hazard calculations, natural integration of cascading events as described embodiments natively use graphs that can capture linked impact effects, and transferability of skills to different climate hazards and different regions of interest or time periods by leveraging publicly available data.

Referring now to FIG. 2, an operational flowchart for an illustrative process 200 of generating impact functions for geospatial climate hazards based on user interactions according to at least one embodiment is provided. FIG. 3 will be discussed with reference to FIG. 2.

Specifically, FIG. 3 illustrates exemplary system architecture 300 for generating impact functions for geospatial climate hazards based on user interactions according to at least one embodiment. System architecture 300 may include a user interface 310, a user interaction module 320 including a chatbot 330 for gathering and receiving input data, data processing modules 340 for processing the received input data and leveraging an accessible library 345 to identify most relevant candidate variables and candidate function formulas usable for impact function generation, graph construction modules 350 for converting the processed data into knowledge graphs, graph neural networks (GNNs) 360 for generating embeddings corresponding to the graphs, and an impact function generator 370 for performing symbolic regression on the generated embeddings to generator impact functions for a target geospatial climate hazard and a corresponding asset. FIG. 3 may also include a feedback module 380 configured to gather and store feedback data within accessible library 345, and an output display module 390 configured to output the generated impact functions for the target geospatial climate hazard and the corresponding asset to a user via user interface 310.

Returning to FIG. 2 and illustrative process 200, at step 202, impact function generation program 150 may receive input data associated with a target geospatial climate hazard and a corresponding asset, the input data comprising one or more of a first dataset corresponding to a predetermined set of prompts, and a second dataset corresponding to a natural language exchange. For example, in embodiments, as shown in FIG. 3, impact function generation program 150 may be configured to leverage an exemplary user interaction module 320 including a chatbot 330 to prompt a user to provide input data via a suitable user interface 310. In embodiments, user interaction module 320 may be configured to receive input data corresponding to a first dataset and a second dataset. In other embodiments, user interaction module 320 may be configured to receive only the first or the second dataset.

In embodiments, to receive the input data corresponding to the first dataset, exemplary user interaction module 320 of impact function generation program 150 may be configured to prompt the user with a predetermined set of questions, such as, for example, multiple choice questions, to gather desired information and data related to a target geospatial climate hazard and a corresponding asset of interest. For example, exemplary user interaction module 320 of impact function generation program 150 may be configured to prompt a user to input a time and date, a hazard type, an asset type (building type, vehicle type, or any other suitable asset type), a time horizon to consider, and any other suitable datapoints that may be useful in predicting the impact of the target geospatial climate hazard on the target asset.

In embodiments, to receive the input data corresponding to the second dataset, exemplary user interaction module 320 of impact function generation program 150 may be configured to leverage an exemplary chatbot 330 (see FIG. 3) to conduct a natural language exchange with a user to gather data related to user interests. In embodiments, the conducted natural language exchange may be preconfigured to discuss topics such as, for example, damage types and descriptions, historical precedents, characteristics of the target asset, contextual land use, damage units of interest, infrastructure, structural defenses, and any other suitable topics that may be useful for generating suitable impact functions for the target geospatial climate hazard and the corresponding asset. In embodiments, a data processing module 340 (see FIG. 3) may further process the first and second datasets to extract certain useful properties, keywords, dimensions, tags, or other useful data that may be subsequently leveraged for searching and filtering through historical data as will be described in greater detail below.

Next at 204, impact function generation program 150 may generate, based on the first dataset an entity knowledge graph including a series of candidate variables related to the target geospatial climate hazard and the corresponding asset. At this step, impact function generation program 150 may leverage a graph generation module 350 to generate an entity knowledge graph based on previously collected data products and models, contained in an accessible library 345, which may be used to determine candidate variables or terms which are strongly associated with the target geospatial climate hazard and the corresponding asset. This may be accomplished by extracting properties, dimensions, tags, keywords, and other data gathered from the first dataset using an exemplary data processing module 340 (See FIG. 3) and filtering historical data in accessible library 345 with the extracted properties, dimensions, tags, keywords. For example, extracted properties and keywords identified by data processing module 340 may be used to filter through historical data of observations and computational simulations describing earthquakes and associated impacts. It should be noted that this includes historical data, rather than externally our publicly available data. The historical data may be accessed from an accessible library 345 to identify a series of suitable candidate variables, which may include, for example, ‘Magnitude Value’, ‘Building Height’, ‘Building Material’, ‘Geographic building location’, and other variables that are strongly associated with the target climate hazard ‘Tl’ associated with an earthquake. In this example, the historical data in accessible library 345 that was leveraged, may have included, for example, historical earthquakes, historical observed damages, historical earthquake simulation data, and any other relevant historical data related to earthquakes. In other words, the first dataset provides ‘keywords’ to search and filter accessible library 345 which includes insights and data from both historical observations and computational simulations describing hazard types and impacts. The filtered data and models may then be linked in an entity knowledge graph generated by graph construction module 350 of impact function generation program 150 according to the user interest.

At 206, impact function generation program 150 may generate, based on the second dataset, a universal knowledge graph including a series of candidate function formulas that are most associated with the target geospatial climate hazard and the corresponding asset. At this step, the natural language exchange described in step 202 may be leveraged as input for a comprehensive search through accessible library 345 to extract impact function formulations related to the user interest. Information from the extracted formulations may then be included in a universal knowledge graph, linking, for example, nodes of hazard types, asset types and damages via edges of hazard intensity. In embodiments, data processing module 340 of impact function generation program 150 may be configured to access and additionally search any suitable external sources containing relevant and useful impact function formulations. Accordingly, data processing module