Systems, methods, and media for generating documents using machine learning

Abstract

Mechanisms for generating a document are provided, the mechanisms including: receiving a set of question(s), each of the question(s) corresponding to one or more requirement content items that are related to one or more aspects of a subject of the document; receiving reference data comprising first reference data and second reference data; using a first Al model to generate a template based on one or more of the set of questions and the first reference data, wherein the set of questions are mapped into the template using a mapping procedure that determines the locations to which each question is mapped within the template file; using a second Al model to generate answers to the set of questions based on the second reference data; and replacing each question in the template file by its corresponding generated answer to obtain the document.

Description

SYSTEMS, METHODS, AND MEDIA FOR GENERATING DOCUMENTS USING MACHINE LEARNINGCROSS-REFERENCE TO RELATED APPLICATION

[0001] This application claims the benefit of United States Patent Application No.18 / 935,418, filed November 1, 2024, which claims the benefit of United States Provisional Patent Application No. 63 / 595,182, filed November 1, 2023, and also claims the benefit of United States Provisional Patent Application No. 63 / 595,182, filed November 1, 2023, each of which is hereby incorporated by reference herein in its entirety.BACKGROUND

[0002] While machine learning mechanisms, such as large language models, can generate text for documents, the manner in which it does so is generally in a random format and lacks structure.

[0003] Accordingly, new mechanisms for generating documents using machine learning are desirable.SUMMARY

[0004] Mechanisms, including systems, methods, and media, for document generation using machine learning are provided.

[0005] In some embodiments, methods for generating a document are provided, the methods comprising: receiving a set of question(s), each of the question(s) corresponding to one or more requirement content items that are related to one or more aspects of a subject of the document; receiving reference data comprising first reference data and second reference data; using a first Al model to generate a template based on one or more of the set of questions and the first reference data, wherein the set of questions are mapped into the template using a mapping procedure that determines the locations to which each question is mapped within the template file; using a second Al model to generate answers to the set of questions based on the second reference data; and replacing each question in the template file by its corresponding generated answer to obtain the document.BRIEF DESCRIPTION OF THE DRAWINGS

[0006] It is to be understood that the specific configurations illustrated in the drawings are intended to exemplify embodiments of the invention and are not limiting and that other alternative configurations are possible.

[0007] FIG. 1 is an example of a process for generating a document in accordance with some embodiments.

[0008] FIG. 2 is an example of a process for generating a template in accordance with some embodiments.

[0009] FIG. 3 is an example of a combined process for generating a template and a document in accordance with some embodiments.

[0010] FIG. 4 is another example of a combined process for generating a template and a document in accordance with some embodiments.

[0011] FIGS. 5-9 are examples of templates in accordance with some embodiments.

[0012] FIG. 10 is an example of hardware that can be used in accordance with some embodiments.DETAILED DESCRIPTION

[0013] Mechanisms, including systems, methods, and media, for document generation using machine learning are provided.

[0014] In accordance with some embodiments, these mechanisms use templates in conjunction with one or more machine learning mechanisms to produce documents.

[0015] It should be understood that the invention is not limited in its application to the details of construction and the arrangement of components set forth in this description and / or as illustrated in the accompanying drawings. The invention is capable of other embodiments and of being practiced or of being carried out in various ways. Also, it is to be understood that the phraseology and terminology used herein is for the purpose of description and should not be regarded as limiting. The use of “including”, “comprising”, or “having” and variations thereof herein is meant to encompass the items listed thereafter and equivalents thereof as well as additional items.

[0016] Furthermore, and as described in subsequent paragraphs, the specific configurations illustrated in the drawings are intended to exemplify embodiments of the invention and other alternative configurations are possible.

[0017] Turning to FIG. 1, an example 100 of a process for document generation using machine learning in accordance with some embodiments is shown. Process 100 can be executed in any suitable device such as one or more hardware processors, such as hardware processor 1002 of FIG. 10, in some embodiments.

[0018] As illustrated, after process 100 begins, the process receives a template. The template can have any suitable format and / or content in some embodiments. In some embodiments, the template can include one or more questions (which can refer to questions, queries, requests, and / or any other similar inquiry). In some embodiments, the template can additionally or alternatively contain information about one or more aspects related to the subject of the document to be generated. In some embodiments, the template can be used to guide process 100 in structuring one or more documents to be generated by the process. For example, in some embodiments, the template can indicate where to insert information / answers, and how to format / represent the generated information / answers in a file, how to structure information / answers in the document, etc.

[0019] In some embodiments, a template can be generated automatically or semi- automatically by mapping user-provided questions into a file. For example, in some embodiments, a PDF or a MICROSOFT WORD or EXCEL file can be provided and, within the file, select locations can be determined as locations within the file at which questions are to be mapped. Such locations can be determined for example through labelled fields and / or tagged fields and / or indices and / or keywords and / or designated symbols and / or code embedded within the file, in some embodiments.

[0020] Next, at 104, process 100 can generate information / answers (which can refer to any responses to questions, queries, requests, and / or any other similar inquiries) 106 related to the one or more questions using one or more machine learning mechanisms. In some embodiments, any suitable machine learning mechanisms can be used to generate information / answers 106 in some embodiments. For example, in some embodiments, machine learning mechanisms used can implement any one or more of neural networks, deep learning networks, transformers, decision trees, random forests, decision ferns, K-nearest neighbors (K-NN), K-means, NaturalLanguage Processing (NLP)-based models, Large Language Models (LLMs), Generative Pretrained Transformers (GPT)-based models, Retrieval Augmented Generation (RAG) models, Bidirectional Encoder Representations from Transformers (BERT)-based models, Bidirectional and Autoregressive Transformers (BART)-based models, masked language models, denoising autoencoder based models. In some embodiments, the machine learning mechanisms can use on or more of OPENAI's GPT LLMs (e.g., GPT-3, GPT-4), META’s Large Language Model META Al (e g., LLAMA 2, LLAMA 3), GOOGLE DEEPMIND’s GEMINI, ANTHROPIC’S CLAUDE, TECHNOLOGY INNOVATION INSTITUTE’ S FLACON LLM, STANFORD UNIVERSITY’S ALPACA LLM. In some embodiments, machine learning mechanisms can also comprise one or more machine learning based classification models.

[0021] In some embodiment, machine learning mechanisms used in connection with the processes described herein can be trained in any suitable manner. For example, in some embodiments, the machine learning mechanisms can be trained with training data that includes a set of questions, each question corresponding to a content item that is related to one or more subjects of a document to be created, and a set of content items, in one-to-one correspondence with the questions; such set of questions and corresponding set of content items can be used for example to form training data pairs, e.g., (question, corresponding content item), wherein corresponding content item (e.g., a second element in the pair for example) include information related to one or more answers to the question (e.g., a first element in the pair); optionally, the generated training data may also contain additional information about one or more aspects related a subject of the document to be created (e.g., city, zone related to a location). In some embodiments, the training data can be generated using data that is stored locally on the system, data that is stored on one or more remote servers such as the cloud, user-provided data, data augmentation methods, machine learning techniques and / or a chatbot, or a combination thereof.

[0022] In some embodiments, the information / answers can be generated based on data available from a cloud server or a web server or a local server, or a combination thereof, and / or from one or more software packages running on a local and / or remote (e.g., cloud) computing device (e g., AUTOCAD, CIVIL 3D, AUTODESK, ORACLE, ARCGIS, FFIEC GEOCODING / MAPPING SYSTEM, and / or other commercial software and / or non-commercial software and / or custom software packages).

[0023] Next, at 108, process 100 can replace the one or more questions in the template with the generated information / answers to produce generated document 110. In some embodiments, the generated information / answers can be arranged in the generated document in any suitable format and / or layout. For example, in some embodiments, the generated information / answers can be presented in format / layout specified by the template. In some embodiments, the generated document can include template content together with the generated information / answers. In some embodiments, the generated document can include template content with the generated information / answers contained within the template content.

[0024] At 112, process 100 can then output the generated document in any suitable manner. For example, in some embodiments, process 100 can output the generated document to a display, to a file, to an email, to an API, and / or to any other destination or interface. More particularly, in some embodiments, the generated document can be output in the form of a text file, and / or MICROSOFT WORD file, and / or MICROSOFT EXCEL file and / or PDF file or any other human-readable format. In some embodiments, the generated document can additionally or alternatively be output in a non-human readable format and / or can be intended to be readable by a software package running on a local and / or remote (e.g., cloud) computing device (e.g., AUTOCAD, CIVIL 3D, AUTODESK, ORACLE, ARCGIS, FFIEC GEOCODING / MAPPING SYSTEM, and / or non-commercial software and / or other commercial software and / or custom software packages) and / or interpretable via a machine and / or machine-learning mechanism that can be used to provide recommendations to a user in response to one or more inquiries that are input by the user. In some embodiments, the generated document can also be a combination of human-readable and non-human readable formats.

[0025] Turning to FIG. 2, an example 200 of a process for generating templates in accordance with some embodiments is illustrated. Process 200 can be executed in any suitable device such as one or more hardware processors, such as hardware processor 1002 of FIG. 10, in some embodiments.

[0026] As shown, after process 200 begins, the process receives a set of questions comprising one or more questions. These questions can be received from any suitable source in some embodiments. In some embodiments, the questions can be on any suitable subject. And, in some embodiments, these questions can be received in any suitable format.

[0027] For example, in some embodiments, the questions can be received from a user via a user interface. Any suitable user interface can be used, in some embodiments. For example, the user interface can be a text-based user interface or a graphical user interface (GUI), or a touch screen user interface, or a combination thereof, through which a user can provide the input to the system by inputting text (e.g., in a natural language form and / or in the form of keywords or a combination thereof) into a one or more designated fields of the user interface, and / or one or more command lines, and / or by means of input files, and / or by making one or more selections (e g., selecting a location on a map and / or making a selection through options provided in the user interface such as buttons, drop-down menus, and other selection interfaces) using an input device such as a mouse, or through touch (e.g., by means of a touch screen), or through voice (e.g., by means of a microphone input device by speaking and / or issuing a command through voice) or a combination thereof.

[0028] As another example, in some embodiments, the questions can be received from over an API.

[0029] Next, at 204, process 200 can map the one or more questions from the set of questions into a template 206. The template can exist in any suitable manner, in some embodiments. For example, the template can exist in a file, a data structure, a stream, a memory location, a register, a buffer, etc.

[0030] In some embodiments, the template content can be automatically determined based on the set of questions. For example, in some embodiments, given a set of questions as input, a classification procedure can analyze the set of questions and extract features that are used to determine a class or context of the questions, and, once a class or context is determined, a template corresponding to that class or context can be retrieved and used In some embodiments, the questions can be mapped into the retrieved template via keywords and / or via a mapping procedure that matches keywords to the questions. In some embodiments, certain locations can be determined as locations within the template into which questions are to be mapped. Such locations can be determined for example through labelled fields and / or tagged fields and / or indices and / or keywords and / or designated symbols and / or code embedded within the said template file, in some embodiments.

[0031] In some embodiments, the set of questions can be input (as described above) to a machine learning mechanism that is used to determine a corresponding template. Additionally,in some embodiments, the machine learning mechanism can determine a mapping (e.g., mapping table, or mapping pointers, or mapping indices, or mapping function) providing the location to which each question is mapped within the template. In some embodiments, the machine learning mechanism can directly generate the template together with a mapping indicating the location to which each question is mapped within the template.

[0032] In some embodiments, at 202, process 200 can receive reference data in addition to a set of questions to be used in generating the template. In some embodiments, this reference data can contain documents, portions of documents, figures, drawings, datasets, media and other content, and that reference data can be used together with the set of questions to generate a template at 204. In some embodiments, the reference data can be automatically selected based on questions received in any suitable manner, such as by using a machine learning mechanism.

[0033] In some embodiments, for template generation, given a set of questions (including but not limited to one or more questions, queries, or requests), the set of questions can be analyzed to automatically select or generate a template including static content regions or fields, e.g., regions or fields comprising fixed content that cannot be altered, and / or of dynamic content regions or fields, e.g., regions or fields comprising changeable content that can be changed. Such dynamic content regions and fields can take the form of empty regions or fields that can be later filled with new content; or non-empty regions or fields that can be modified with new content, in some embodiments. FIGS. 5-9 provide examples of such templates, wherein the solid filled regions correspond to static-content regions or fields and the textured filled regions correspond to dynamic-content regions or fields, in accordance with some embodiments. It should be noted that a template can include any suitable number of parts or pages, wherein each part or page can be structured differently in terms of the locations and types of the static and dynamic content regions, in some embodiments. For example, in some embodiments, a single template can include parts or pages with one or more of the structures shown in FIGS. 5-9 combined together to form a single template file. It should be noted that “static content” is used herein to refer to content that cannot be changed and can include text, equations, numbers, images, videos, graphics, drawings, animations, code, scripts, speech, audio, and / or other media or forms of content, in some embodiments. Similarly, “dynamic content” can be used herein to refer to content that can be changed and can include text, equations, numbers, images, videos, graphics, drawings, animations, code, scripts, speech, audio, and / or other media or forms of content, insome embodiments. Tn some embodiments, a template can be defined by the locations and types of each static-content and dynamic-content regions. In some embodiments, the dynamic content can be automatically generated or altered using a machine learning mechanism.

[0034] In some embodiments, a template can be included as part of a document and / or file, or can be stored on a local and / or remote storage device, or a combination thereof. The document and / or file can be a text file, and / or MICROSOFT WORD file, and / or MICROSOFT EXCEL file and / or PDF file or any other human-readable format. The document can also be a non-human readable format and / or can be intended to be readable by a software package running on a local and / or remote (e.g., cloud) computing device (e.g., AUTOCAD, CIVIL 3D, AUTODESK, ORACLE, ARCGIS, FFIEC GEOCODING / MAPPING SYSTEM, and / or other commercial software and / or non-commercial software and / or custom software packages) and / or interpretable via a machine and / or machine-learning based device, in some embodiments. The document can also be a combination of human-readable and non-human readable formats, in some embodiments.

[0035] In some embodiments, a set of questions can be mapped into a selected or generated template via a mapping procedure that determines the “dynamic-content” locations to which each question is mapped within the template. Such mapping procedure can employ for example one or more of the following: a machine learning mechanism, labelled fields, tagged fields, indices, keywords, designated symbols, code embedded within the template, mapping table, mapping pointers, mapping indices, mapping functions. For example, keywords can be used to match questions to fields. In some embodiments, the set of questions or reference data can be input to a machine learning mechanism that is used to select or generate the corresponding template. Additionally, in some embodiments, the machine learning mechanism can determine a mapping (e.g., mapping table, or mapping pointers, or mapping indices, or mapping keywords, or mapping function) providing the “dynamic-content” locations to which each question is mapped within the template. In some embodiments, the machine learning mechanism can directly select or generate a template together with a mapping indicating the “dynamic-content” locations to which each question is mapped within the template.

[0036] Turning to FIG. 3, an example 300 of a combination of the processes of FIGS. 1 and 2 in accordance with some embodiments is shown. Process 300 can be executed in any suitabledevice such as one or more hardware processors, such as hardware processor 1002 of FIG. 10, in some embodiments.

[0037] As illustrated, blocks 302, 304, and 306 of FIG. 3 can be performed in the same or a similar way to blocks 202, 204, and 206 of FIG. 2, and blocks 308, 310, 312, 314, and 316 of FIG. 3 can be performed in the same or a similar way to blocks 104, 106, 108, 110, and 112 of FIG. 1.

[0038] Turning to FIG. 4, another example 400 of a combination of the processes of FIGS. 1 and 2 in accordance with some embodiments is shown. Process 400 can be executed in any suitable device such as one or more hardware processors, such as hardware processor 1002 of FIG. 10, in some embodiments.

[0039] As illustrated, blocks 402, 404, and 406 of FIG. 4 can be performed in the same or a similar way to blocks 202, 204, and 206 of FIG. 2, and blocks 408, 410, 412, 414, and 416 of FIG. 4 can be performed in the same or a similar way to blocks 104, 106, 108, 110, and 112 of FIG. 1. However, as also illustrated in FIG. 4, a first machine learning mechanism can be used at 404 while a second machine learning mechanism can be used at 408, in some embodiments.

[0040] In some embodiments, the mechanisms described herein can be used in connection with land development, environment, zoning, property assessment and / or other domains that are governed by requirements and regulations. In some embodiments, the mechanisms described herein can be used to determine in an automated manner applicable city, county, state, and / or other requirements (e.g., requirements that apply to a property such as (e.g., land, site, lot, building, house, development, to name a few) and / or to determine whether the applicable requirements are met (e.g., determining whether applicable requirements are met by the property and / or determining environmental characteristics and / or environmental requirements that apply to the property, and / or determining entitlement requirements that apply to the property and / or determining zoning requirements and / or related assessment that apply to the property, and / or performing due diligence for the property, and / or determining the property’s neighborhood sentiments about current developments and / or future potential developments in the neighborhood (“neighborhood outreach”)).

[0041] In accordance with some embodiments, the mechanism described herein can generate information and / or documents with the needed information more efficiently and accurately than existing machine learning mechanisms at least in the regard that these mechanisms generateinformation and / or documents based on a determined relevant template format rather than just as an unstructured output. The selected or generated template sets the appropriate context, which helps in increasing the accuracy of the generated information. It also helps in increasing the efficiency by generating accurate and relevant information and / or documents for various tasks.

[0042] The mechanisms of this disclosure may be implemented in a wide variety of devices or apparatuses, including desktops, laptops, local servers, remote servers, tablets, wireless computing devices, wireless handsets, cellular phones, Bluetooth devices, other wired and / or wireless devices. Any components, modules or units have been described to emphasize functional aspects and should not be regarded as limiting. The techniques described herein may also be implemented in hardware, software, firmware, or any combination thereof. Any features described as modules, units or components may be implemented together in an integrated logic device or separately as discrete but interoperable logic devices. In some cases, various features may be implemented as an integrated circuit device, such as an integrated circuit chip or chipset. If implemented in software, the techniques may be realized at least in part by a computer- readable medium comprising instructions that, when executed in a processor, performs one or more of the methods described above. The computer-readable medium may comprise a computer-readable storage medium and may form part of a computer program product, which may include packaging materials. The computer-readable storage medium may comprise random access memory (RAM) such as synchronous dynamic random-access memory (SDRAM), readonly memory (ROM), non-volatile random access memory (NVRAM), electrically erasable programmable read-only memory (EEPROM), FLASH memory, magnetic or optical or solid state data storage media, and the like. The techniques additionally, or alternatively, may be realized at least in part by a computer-readable communication medium that carries or communicates code in the form of instructions or data structures that can be accessed, read, and / or executed by a processor and / or computer and / or computing device comprising one or more processors. The code or instructions may be executed by one or more processors such as central processing units (CPUs), graphics processing units (GPUs), general purpose microprocessors, an application specific integrated circuits (ASICs), field programmable logic arrays (FPGAs), digital signal processors (DSPs) or other equivalent integrated or discrete logic circuitry or a combination thereof. Accordingly, the term “processor,” as used herein may refer to any of the foregoing structure or any other structure suitable for implementation of thetechniques described herein. In addition, in some aspects, the functionality described herein may be provided within one or more dedicated software modules or hardware modules that may be implemented together within a single device or separately as discrete but interoperable devices. This disclosure also contemplates any of a variety of integrated circuit devices that include circuitry to implement one or more of the techniques described in this disclosure. Such circuitry may be provided in a single integrated circuit chip or in multiple, interoperable integrated circuit chips in a so-called chipset.

[0043] In some embodiments, the mechanisms described herein can be implemented in any suitable general-purpose computer or special-purpose computer. Any such general-purpose computer or special-purpose computer can include any suitable hardware, in some embodiments. For example, as illustrated in example hardware 1000 of FIG. 10, such hardware can include hardware processor 1002, memory and / or storage 1004, an input device controller 1006, an input device 1008, display / audio drivers 1010, display and audio output circuitry 1012, communication interface(s) 1014, an antenna 1016, and a bus 1018.

[0044] Hardware processor 1002 can include any suitable hardware processor, such as a microprocessor, a micro-controller, digital signal processor(s), dedicated logic, and / or any other suitable circuitry for controlling the functioning of a general -purpose computer or a special purpose computer in some embodiments.

[0045] Memory and / or storage 1004 can be any suitable memory and / or storage for storing programs, data, and / or any other suitable information in some embodiments. For example, memory and / or storage 1004 can include random access memory, read-only memory, flash memory, hard disk storage, optical media, and / or any other suitable memory.

[0046] Input device controller 1006 can be any suitable circuitry for controlling and receiving input from input device(s) 1008 in some embodiments. For example, input device controller 1006 can be circuitry for receiving input from an input device 1008, such as a touch screen, from one or more buttons, from a voice recognition circuit, from a microphone, from a camera, from an optical sensor, from an accelerometer, from a temperature sensor, from a near field sensor, and / or any other type of input device.

[0047] Display / audio drivers 1010 can be any suitable circuitry for controlling and driving output to one or more display / audio output circuitries 1012 in some embodiments. For example,display / audio drivers 1010 can be circuitry for driving one or more display / audio output circuitries 1012, such as an LCD display, a speaker, an LED, or any other type of output device.

[0048] Communication interface(s) 1014 can be any suitable circuitry for interfacing with one or more communication networks. For example, interface(s) 1014 can include network interface card circuitry, wireless communication circuitry, and / or any other suitable type of communication network circuitry.

[0049] Antenna 1016 can be any suitable one or more antennas for wirelessly communicating with a communication network in some embodiments. In some embodiments, antenna 1016 can be omitted when not needed.

[0050] Bus 1018 can be any suitable mechanism for communicating between two or more components 1002, 1004, 1006, 1010, and 1014 in some embodiments.

[0051] Any other suitable components can additionally or alternatively be included in hardware 1000 in accordance with some embodiments.

[0052] It should be understood that at least some of the above-described blocks of the process of FIGS. 1-4 can be executed or performed in any order or sequence not limited to the order and sequence shown in and described in the figures. Also, some of the above blocks of the processes of FIG. 1-4 can be executed or performed substantially simultaneously where appropriate or in parallel to reduce latency and processing times. Additionally or alternatively, some of the above described blocks of the processes of FIGS. 1-4 can be omitted.

[0053] In some embodiments, any suitable computer readable media can be used for storing instructions for performing the functions and / or processes described herein. For example, in some embodiments, computer readable media can be transitory or non-transitory. For example, non-transitory computer readable media can include media such as non-transitory magnetic media (such as hard disks, floppy disks, and / or any other suitable magnetic media), non- transitory optical media (such as compact discs, digital video discs, Blu-ray discs, and / or any other suitable optical media), non-transitory semiconductor media (such as flash memory, electrically programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), and / or any other suitable semiconductor media), any suitable media that is not fleeting or devoid of any semblance of permanence during transmission, and / or any suitable tangible media. As another example, transitory computer readable media can include signals on networks, in wires, conductors, optical fibers, circuits, any suitable media thatis fleeting and devoid of any semblance of permanence during transmission, and / or any suitable intangible media.

[0054] In one embodiment a system for automated or semi-automated report generation for one or more aspects of property assessment including a user interface for entering a property address and / or property identifying information and / or other property related information, a computing device, a storage device, and optionally wired and / or wireless communication links for communicating with one or more remote devices. The user interface can be a text-based user interface or a graphical user interface (GUI), or a touch screen user interface, or a combination thereof, through which a user can provide the input to the system by inputting text (e.g., in a natural language form and / or in the form of keywords or a combination thereof) into a one or more designated fields of the user interface, and / or one or more command lines, and / or by means of input files, and / or by making one or more selections (e g., selecting a location on a map and / or making a selection through options provided in the user interface such as buttons, drop-down menus, and other selection interfaces) using an input device such as a mouse, or through touch (e.g., by means of a touch screen), or through voice (e.g., by means of a microphone input device by speaking and / or issuing a command through voice) or a combination thereof. In addition to a property address and / or property identifying information and / or other property related information, the user can optionally provide information about which report template to use and / or which property assessment information is being requested (e g., zoning requirements and / or related assessment, environmental requirements and / or related assessment, entitlement requirements and / or related assessments, appraisal, due diligence, neighborhood outreach, mapping, or a combination thereof). The system may also obtain needed information from data available on a cloud server or a web server or a local server, or a combination thereof, and / or from one or more software packages running on a local and / or remote (e.g., cloud) computing device (e g., AUTOCAD, CIVIL 3D, AUTODESK, ORACLE, ARCGIS, FFIEC GEOCODING / MAPPING SYSTEM, and / or other commercial software and / or non-commercial software and / or custom software packages). The system then outputs a generated report. The report can be a text file, and / or word file, and / or excel file and / or pdf file or any other human- readable format. The generated report can also be a non-human readable format and / or can be intended to be readable by a software package running on a local and / or remote (e.g., cloud) computing device (e.g, AUTOCAD, CIVIL 3D, AUTODESK, ORACLE, ARCGIS, FFIECGEOCODING / MAPPING SYSTEM, and / or non-commercial software and / or other commercial software and / or custom software packages) and / or interpretable via a machine and / or machinelearning based device that can be used to provide recommendations to a user in response to one or more inquiries that are input by the user. The document can also be a combination of human- readable and non-human readable formats.

[0055] In another embodiment, a system for automated or semi-automated report generation for one or more aspects of property assessment including a natural language processing (NLP) based user interface that is implemented by means of machine learning techniques and optionally wired and / or wireless communication links for communicating with one or more remote devices. A user can input questions (e.g., by typing text and / or by speaking and / or by selecting from a set of questions and / or by means of input files and / or by making one or more selections using an input device) about one or more aspects related to property assessment and the system outputs the answers in a form that can be easily understood by the user. The style of the answers (i.e., simple versus complex, using domain-specific terminology or not, using a specific language) may also be adapted based on the user’s needs (“user-centric adaptation”). The natural language processing and / or the user-centric adaptation are performed using Al and / or machine learning methods. Such machine learning methods include for example methods employing neural networks, and / or deep learning, and / or transformers, and / or statistical machine learning techniques, and / or large language models (LLMs), and / or other machine-learning based language models, and / or generative Al techniques, and / or other machine learning techniques that are known in the art.

[0056] In another embodiment, a system for automated or semi-automated report generation for one or more aspects of property assessment including a natural language processing (NLP) based user interface and / or a chatbot implemented by means of machine learning techniques and optionally wired and / or wireless communication links for communicating with one or more remote devices. A user can input questions (e.g., by typing text and / or by speaking and / or by selecting from a set of questions and / or by means of input fdes and / or by making one or more selections using an input device) about one or more aspects related to property assessment and the system outputs the answers in a form that can be easily understood by the user. The style of the answers (i.e., simple versus complex, using domain-specific terminology or not, using a specific language) may also be adapted based on the user’s needs (“user-centric adaptation”).The natural language processing and / or the user-centric adaptation are performed using Al and / or machine learning methods. Such machine learning methods include for example methods employing neural networks, and / or deep learning, and / or transformers, and / or statistical machine learning techniques, and / or large language models (LLMs), and / or other machine-learning based language models, and / or generative Al techniques, and / or other machine learning techniques that are known in the art.

[0057] In another embodiment, a system for automated or semi-automated report generation for one or more aspects of property assessment including a user interface, a computing device, a storage device, and optionally wired and / or wireless communication links for communicating with one or more remote devices and wherein a report template is generated automatically or semi-automatically by mapping user-provided questions into a file. For example, a pdf or word or excel file can be provided and, within said file, select locations can be determined as locations within the said file where questions need to be mapped. Such locations can be determined for example through labelled fields and / or tagged fields and / or indices and / or keywords and / or designated symbols and / or code embedded within the file. The generated report template with the mapped questions is in turn used by the system for automated or semi-automated report generation for one or more aspects of property assessment by replacing the questions in the said report template with corresponding answers that are generated by means of machine learning methods and / or chatbot. Such machine learning methods include for example natural language processing (NLP) methods employing neural networks, and / or deep learning, and / or transformers, and / or statistical machine learning techniques, and / or large language models (LLMs), and / or other machine-learning based language models, and / or generative Al techniques, and / or other machine learning techniques that are known in the art.

[0058] According to one embodiment of the present disclosure, there is provided a system for generating a model to be used for automated or semi-automated report generation for one or more aspects of property assessment, the system comprising: 1) a data unit that receives training data comprising a set of questions, each question corresponding to a content item that is related to one or more aspects of property assessment, and a set of content items, in one-to-one correspondence with the questions; such set of questions and corresponding set of content items can be used for example to form training data pairs, e.g., (question, corresponding content item), wherein corresponding content item (i.e., second element in the pair for example) includeinformation related to one or more answers to the question (i.e., first element in the pair); optionally, the training data may also contain additional information about one or more aspects related to property assessment (e.g., city, zone); 2) a training unit that employs the training data to train an Artificial Intelligence (Al) based algorithm such as a machine learning (ML) algorithm or an Al-based system and / or device implementing a machine learning algorithm to learn a model (“learned model”) that is employed as part of the report generation system. The Al-based system (algorithm) can be a rule-based Al system (algorithm) and / or a machine learning-based system (algorithm). For the machine learning system and / or algorithm, any suitable machine learning technique can be used including, for example, a neural network, a deep learning network, a transformer, a decision tree, a random forest, a decision fern, K-nearest neighors (K-NN), K-means, ensemble of these or a combination thereof, and / or other machine learning techniques that are known in the art. The learned model can be an NLP -based model, a Large Language Model (LLM), a Generative Pretrained Transformer (GPT) based model, a Bidirectional Encoder Representations from Transformers (BERT) based model, a Bidirectional and Autoregressive Transformers (BART) based model, a masked language model, a denoising autoencoder based model, or a combination thereof, and / or other Generative Al models that are known in the art. The learned model can comprise one or more LLMs that are known in the art such as OPENAI's GPT LLMs (e g., GPT-3, GPT-4), META’s Large Language Model META Al (e g., LLAMA 2), TECHNOLOGY INNOVATION INSTITUTE’S FLACON LLM, STANFORD’S ALPACA LLM. The learned model can also comprise a machine learning based classification model.

[0059] According to another embodiment of the present disclosure, there is provided a system for generating a model to be used for automated or semi-automated report generation for one or more aspects of property assessment, the system comprising: 1) a data generation unit for generating training data comprising a set of questions, each question corresponding to a content item that is related to one or more aspects of property assessment, and a set of content items, in one-to-one correspondence with the questions; such set of questions and corresponding set of content items can be used for example to form training data pairs, e.g., (question, corresponding content item), wherein corresponding content item (i.e., second element in the pair for example) include information related to one or more answers to the question (i.e., first element in the pair); optionally, the generated training data may also contain additional information about one or moreaspects related to property assessment (e.g., city, zone); 2) a training unit that employs the generated training data to train an Artificial Intelligence (Al) based algorithm such as a machine learning algorithm or an Al-based system and / or device implementing a machine learning algorithm to learn a model (“learned model”) that is employed as part of the report generation system. The Al-based system (algorithm) can be a rule-based Al system (algorithm) and / or a machine learning-based system (algorithm). For the machine learning system and / or algorithm, any suitable machine learning technique can be used including, for example, a neural network, a deep learning network, a transformer, a decision tree, a random forest, a decision fem, K-nearest neighors (K-NN), K-means, ensemble of these or a combination thereof, and / or other machine learning techniques that are known in the art. The learned model can be an NLP -based model, a Large Language Model (LLM), a Generative Pretrained Transformer (GPT) based model, a Bidirectional Encoder Representations from Transformers (BERT) based model, a Bidirectional and Autoregressive Transformers (BART) based model, a masked language model, a denoising autoencoder based model, or a combination thereof, and / or other Generative Al models that are known in the art. The learned model can comprise one or more LLMs that are known in the art such as OPENAI's GPT LLMs (e g., GPT-3, GPT-4), META’s Large Language Model META Al (e g., LLAMA 2), TECHNOLOGY INNOVATION INSTITUTE’S FLACON LLM, STANFORD’S ALPACA LLM. The learned model can also comprise a machine learning based classification model. The training data can be generated using data that is stored locally on the system, data that is stored on one or more remote servers such as the cloud, user-provided data, data augmentation methods, machine learning techniques and / or a chatbot, or a combination thereof.

[0060] According to another embodiment of the present disclosure, there is provided a system to be used for automated or semi-automated report generation for one or more aspects of property assessment, the system comprising: 1) an input unit that receives a report template comprising one or more questions; 2) an Al unit that generates answers (“generated answers”) to said one or more questions; 3) a report generation unit that replaces said one or more questions in said report template with said generated answers to generate one or more reports.

[0061] According to another embodiment of the present disclosure, there is provided a system for employing a learned model to be used for automated or semi-automated report generation for one or more aspects of property assessment, the system comprising: 1) an inputunit that receives a report template comprising one or more questions; 2) a machine learning unit that employs the said learned model to generate answers (“generated answers”) to said one or more questions; 3) a report generation unit that replaces said one or more questions in said report template with said generated answers to generate one or more reports.

[0062] According to another embodiment of the present disclosure, there is provided a system to be used for automated or semi -automated report generation for one or more aspects of property assessment, the system comprising: 1) an input unit that receives a set of questions comprising one or more questions; 2) a report template generation unit that maps said one or more questions from said set of questions into a report file to produce a report template; 3) an Al unit that generates answers (“generated answers”) to said one or more questions that are contained in the said report template; 4) a report generation unit that replaces said one or more questions in said report template with said generated answers to generate one or more reports.

[0063] According to another embodiment of the present disclosure, there is provided a system for employing a learned model to be used for automated or semi-automated report generation for one or more aspects of property assessment, the system comprising: 1) an input unit that receives a set of questions comprising one or more questions; 2) a report template generation unit that maps said one or more questions from said set of questions into a report file to produce a report template; 3) a machine learning unit that employs the said learned model to generate answers (“generated answers”) to said one or more questions that are contained in the said report template; 4) a report generation unit that replaces said one or more questions in said report template with said generated answers to generate one or more reports.

[0064] According to another embodiment of the present disclosure, there is provided a system to be used for automated or semi -automated report generation for one or more aspects of property assessment, the system comprising: 1) an input unit that receives a report template comprising one or more questions; 2) an Al unit that generates answers (“generated answers”) to said one or more questions and that replaces said one or more questions in said report template with said generated answers to generate one or more reports.

[0065] According to another embodiment of the present disclosure, there is provided a system for employing a learned model to be used for automated or semi-automated report generation for one or more aspects of property assessment, the system comprising: 1) an input unit that receives a report template comprising one or more questions; 2) a machine learning unitthat employs the said learned model to generate answers (“generated answers”) to said one or more questions and that replaces said one or more questions in said report template with said generated answers to generate one or more reports.

[0066] According to another embodiment of the present disclosure, there is provided a system to be used for automated or semi-automated report generation for one or more aspects of property assessment, the system comprising: 1) an input unit that receives a set of questions comprising one or more questions; 2) a report template generation unit that maps said one or more questions from said set of questions into a report file to produce a report template.

[0067] According to another embodiment of the present disclosure, there is provided a system to be used for automated or semi-automated report generation for one or more aspects of property assessment, the system comprising: 1) an input unit that receives a set of questions comprising one or more questions; 2) an Al unit that produces a report template comprising said one or more questions from said set of questions.

[0068] According to another embodiment of the present disclosure, there is provided a system for employing a learned model to be used for automated or semi-automated report generation for one or more aspects of property assessment, the system comprising: 1) an input unit that receives a set of questions comprising one or more questions; 2) a machine learning unit that employs the said learned model to produce a report template comprising said one or more questions from said set of questions.

[0069] According to another embodiment of the present disclosure, there is provided a system to be used for automated or semi-automated report generation for one or more aspects of property assessment, the system comprising: 1) an input unit that receives a set of questions comprising one or more questions; 2) a reference data unit that receives reference data or location of such reference data; 3) a report template generation unit that generates a report template comprising said one or more questions using said reference data.

[0070] According to another embodiment of the present disclosure, there is provided a system to be used for automated or semi-automated report generation for one or more aspects of property assessment, the system comprising: 1) an input unit that receives a set of questions comprising one or more questions; 2) a reference data unit that determines reference data or location of such reference data; 3) a report template generation unit that generates a report template comprising said one or more questions using said reference data.

[0071] According to another embodiment of the present disclosure, there is provided a system to be used for automated or semi-automated report generation for one or more aspects of property assessment, the system comprising: 1) an input unit that receives a set of questions comprising one or more questions; 2) a reference data unit that receives reference data or location of such reference data; 3) an Al unit that produces a report template comprising said one or more questions using said reference data.

[0072] According to another embodiment of the present disclosure, there is provided a system to be used for automated or semi-automated report generation for one or more aspects of property assessment, the system comprising: 1) an input unit that receives a set of questions comprising one or more questions; 2) a reference data unit that determines reference data or location of such reference data; 3) an Al unit that produces a report template comprising said one or more questions using said reference data.

[0073] According to another embodiment of the present disclosure, there is provided a system to be used for automated or semi-automated report generation for one or more aspects of property assessment, the system comprising: 1) an input unit that receives a set of questions comprising one or more questions; 2) a first Al unit that produces a report template comprising said one or more questions from said set of questions.; 3) a second Al unit that generates answers (“generated answers”) to said one or more questions that are contained in said report template and that replaces said one or more questions in said report template with said generated answers to generate one or more reports.

[0074] According to another embodiment of the present disclosure, there is provided a system for employing a learned model to be used for automated or semi-automated report generation for one or more aspects of property assessment, the system comprising: 1) an input unit that receives a set of questions comprising one or more questions; 2) a reference data unit that receives reference data or location of such reference data; 3) a machine learning unit that employs said learned model and said reference data to produce a report template comprising said one or more questions.

[0075] According to another embodiment of the present disclosure, there is provided a system for employing a learned model to be used for automated or semi-automated report generation for one or more aspects of property assessment, the system comprising: 1) an input unit that receives a set of questions comprising one or more questions; 2) a reference data unitthat determines reference data or location of such reference data; 3) a machine learning unit that employs said learned model and said reference data to produce a report template comprising said one or more questions.

[0076] According to another embodiment of the present disclosure, there is provided a system for employing a learned model to be used for automated or semi-automated report generation for one or more aspects of property assessment, the system comprising: 1) an input unit that receives a set of questions comprising one or more questions; 2) a report template generation unit that maps said one or more questions from said set of questions into a report fde to produce a report template; 3) a machine learning unit that employs the said learned model to generate answers (“generated answers”) to said one or more questions that are contained in said report template and that replaces said one or more questions in said report template with said generated answers to generate one or more reports.

[0077] According to another embodiment of the present disclosure, there is provided a system for employing a first learned model and a second learned model to be used for automated or semi -automated report generation for one or more aspects of property assessment, the system comprising: 1) an input unit that receives a set of questions comprising one or more questions; 2) a first machine learning unit that employs said first learned model to produce a report template comprising said one or more questions from said set of questions.; 3) a second machine learning unit that employs said second learned model to generate answers (“generated answers”) to said one or more questions that are contained in said report template and that replaces said one or more questions in said report template with said generated answers to generate one or more reports.

[0078] According to another embodiment of the present disclosure, there is provided a system for employing a learned model to be used for automated or semi-automated report generation for one or more aspects of property assessment, the system comprising: 1) an input unit that receives a report template comprising one or more questions; 2) a machine learning unit that employs the said learned model to generate one or more reports by replacing said one or more questions in said report template with generated answers.

[0079] According to another embodiment of the present disclosure, there is provided a system for employing a learned model to be used for automated or semi-automated report generation for one or more aspects of property assessment, the system comprising: 1) an inputunit that receives a set of questions comprising one or more questions; 2) a report template generation unit that maps said one or more questions from said set of questions into a report file to produce a report template; 3) a machine learning unit that employs the said learned model to generate one or more reports by replacing said one or more questions in said report template with generated answers.

[0080] According to another embodiment of the present disclosure, there is provided a system for employing a first learned model and a second learned model to be used for automated or semi-automated report generation for one or more aspects of property assessment, the system comprising: 1) an input unit that receives a set of questions comprising one or more questions; 2) a first machine learning unit that employs said first learned model to produce a report template comprising said one or more questions from said set of questions; 3) a second machine learning unit that employs said second learned model to generate one or more reports by replacing said one or more questions in said report template with generated answers.

[0081] According to another embodiment of the present disclosure, there is provided a system for employing a learned model to be used for automated or semi-automated report generation for one or more aspects of property assessment, the system comprising: 1) an input unit that receives a set of questions comprising one or more questions; 2) a machine learning unit that employs said learned model to generate one or more reports by generating answers (“generated answers”) to said one or more questions and by mapping said generated answers into one or more report files.

[0082] According to another embodiment of the present disclosure, there is provided a system for employing a learned model to be used for one or more aspects of property assessment, the system comprising: 1) an input unit that receives one or more questions; 2) a reference data unit that receives reference data or location of such reference data; 3) a machine learning unit that employs the said learned model to generate answers (“generated answers”) to said one or more questions using the said reference data. The said reference data can be provided as a document and / or file and / or script and / or code, or using an address of one or more local and / or remote locations (e.g., remote server or website or storage device) at which the data can be obtained, or a combination thereof. The generated answer(s) to each question can be output using a user interface, or by a chatbot, or included in a document and / or file, or stored on a local and / or remote storage device, or a combination thereof. The document and / or file can be a textfile, and / or word file, and / or excel file and / or pdf file or any other human-readable format. The document can also be a non-human readable format and / or can be intended to be readable by a software package running on a local and / or remote (e.g., cloud) computing device (e.g., AUTOCAD, CIVIL 3D, AUTODESK, ORACLE, ARCGIS, FFIEC GEOCODING / MAPPING SYSTEM, and / or other commercial software and / or non-commercial software and / or custom software packages) and / or interpretable via a machine and / or machine-learning based device. The document can also be a combination of human-readable and non-human readable formats.

[0083] According to another embodiment of the present disclosure, there is provided a system for employing a learned model to be used for one or more aspects of property assessment, the system comprising: 1) an input unit that receives one or more questions; 2) a reference data unit that determines reference data or location of such reference data; 3) a machine learning unit that employs the said learned model to generate answers (“generated answers”) to said one or more questions using the said reference data. The reference data can be a document and / or file and / or script and / or code, or can be stored at one or more local (e.g., local server or local storage device) and / or remote locations (e.g., remote server or website or cloud, or remote storage device), or a combination thereof. The reference data can also be obtained from one or more software packages running on a local and / or remote (e.g., cloud) computing device (e.g., AUTOCAD, CIVIL 3D, AUTODESK, ORACLE, ARCGIS, FFIEC GEOCODING / MAPPING SYSTEM, and / or other commercial software and / or non-commercial software and / or custom software packages). The reference data may be a generated report according to techniques of this disclosure. The generated answer(s) to each question can be output using a user interface, or by a chatbot, or included in a document and / or file, or stored on a local and / or remote storage device, or a combination thereof. The document and / or file can be a text file, and / or word file, and / or excel file and / or pdf file or any other human-readable format. The document can also be a non- human readable format and / or can be intended to be readable by a software package running on a local and / or remote (e.g., cloud) computing device (e.g., AUTOCAD, CIVIL 3D, AUTODESK, ORACLE, ARCGIS, FFIEC GEOCODING / MAPPING SYSTEM, and / or other commercial software and / or non-commercial software and / or custom software packages) and / or interpretable via a machine and / or machine-learning based device. The document can also be a combination of human-readable and non-human readable formats.

[0084] According to another embodiment of the present disclosure, there is provided a system for employing a learned model to be used for automated or semi-automated report generation for one or more aspects of property assessment, the system comprising: 1) an input unit that receives a set of questions comprising one or more questions; 2) a reference data unit that receives reference data or location of such reference data; 3) a machine learning unit that employs said learned model to generate one or more reports by generating answers (“generated answers”) to said one or more questions using said reference data and by mapping said generated answers into one or more report fdes.

[0085] According to another embodiment of the present disclosure, there is provided a system for employing a learned model to be used for automated or semi-automated report generation for one or more aspects of property assessment, the system comprising: 1) an input unit that receives a set of questions comprising one or more questions; 2) a reference data unit that determines reference data or location of such reference data; 3) a machine learning unit that employs said learned model to generate one or more reports by generating answers (“generated answers”) to said one or more questions using said reference data and by mapping said generated answers into one or more report fdes.

[0086] According to another embodiment of the present disclosure, there is provided a system for employing a first learned model and a second learned model to be used for automated or semi-automated report generation for one or more aspects of property assessment, the system comprising: 1) an input unit that receives a set of questions comprising one or more questions; 2) a first reference data unit that receives a first reference data or location of such first reference data; 3) a second reference data unit that receives a second reference data or location of such second reference data; 4) a first machine learning unit that employs said first learned model to produce a report template comprising said one or more questions using said first reference data; 5) a second machine learning unit that employs said second learned model to generate answers (“generated answers”) to said one or more questions that are contained in said report template using second reference data; 6) a report generation unit that replaces said one or more questions in said report template with said generated answers to generate one or more reports.

[0087] According to another embodiment of the present disclosure, there is provided a system for employing a first learned model and a second learned model to be used for automated or semi -automated report generation for one or more aspects of property assessment, the systemcomprising: 1) an input unit that receives a set of questions comprising one or more questions; 2) a first reference data unit that receives a first reference data or location of such first reference data; 3) a second reference data unit that receives a second reference data or location of such second reference data; 4) a first machine learning unit that employs said first learned model to produce a report template comprising said one or more questions using said first reference data; 5) a second machine learning unit that employs said second learned model to generate one or more reports by replacing said one or more questions in said report template with generated answers using said second reference data.

[0088] In another embodiment, a method for automated or semi-automated report generation for one or more aspects of property assessment including a user interface for entering a property address and / or other property related information, a computing device, a storage device, and optionally wired and / or wireless communication links for communicating with one or more remote devices. The user interface can be a text-based user interface or a graphical user interface (GUI), or a touch screen user interface, or a combination thereof, through which a user can provide the input to the system by entering text (e.g., in a natural language form and / or in the form of keywords or a combination thereof) into a one or more designated locations, and / or one or more command lines, and / or by means of input files, and / or by making one or more selections (e.g., selecting a location on a map and / or making a selection through options provided in the user interface such as buttons, drop-down menus, and other selection interfaces) using an input device such as a mouse, or through touch (e.g., by means of a touch screen), or through voice (e g., by means of a microphone input device by speaking and / or issuing a command through voice) or a combination thereof. In addition to a property address and / or property identifying information and / or other property related information, the user can optionally provide information about which report template to use and / or which property assessment information is being requested (e.g., zoning requirements and / or related assessment, environmental requirements and / or related assessment, entitlement requirements and / or related assessments, appraisal, due diligence, neighborhood outreach, mapping, or a combination thereof). The system may also obtain needed information from data available on a cloud server or a web server or a local server, or a combination thereof, and / or from one or more software packages running on a local and / or remote (e.g., cloud) computing device (e.g., AUTOCAD, CIVIL 3D, AUTODESK, ORACLE, ARCGIS, FFIEC GEOCODING / MAPPING SYSTEM, and / or other commercialsoftware and / or non-commercial software and / or custom software packages). The system then outputs a generated report. The report can be a text file, and / or word file, and / or excel file and / or pdf file or any other human-readable format. The generated report can also be a non-human readable format and / or can be intended to be readable by a software package running on a local and / or remote (e.g., cloud) computing device (e.g., AUTOCAD, CIVIL 3D, AUTODESK, ORACLE, ARCGIS, FFIEC GEOCODING / MAPPING SYSTEM, and / or other commercial software and / or non-commercial software and / or custom software packages) and / or interpretable via a machine and / or machine-learning based device that can be used to provide recommendations to a user in response to one or more inquiries that are input by the user. The document can also be a combination of human-readable and non-human readable formats.

[0089] In another embodiment, a method for automated or semi-automated report generation for one or more aspects of property assessment including a natural language processing (NLP) based user interface that is implemented by means of machine learning techniques and optionally wired and / or wireless communication links for communicating with one or more remote devices. A user can input questions (e.g., by typing text and / or by speaking and / or by selecting from a set of questions) about one or more aspects related to property assessment and the system outputs the answers in a form that can be easily understood by the user. The style of the answers (i.e., simple versus complex, using domain-specific terminology or not, using a specific language) may also be adapted based on the user’s needs (“user-centric adaptation”). The natural language processing and / or the user-centric adaptation are performed using Al and / or machine learning methods. Such machine learning methods include for example methods employing neural networks, deep learning, transformers, statistical machine learning techniques, and other machine learning techniques that are known in the art.

[0090] In another embodiment, a method for automated or semi-automated report generation for one or more aspects of property assessment including a natural language processing (NLP) based user interface and / or a chatbot implemented by means of machine learning techniques and optionally wired and / or wireless communication links for communicating with one or more remote devices. A user can input questions (e.g., by typing text and / or by speaking and / or by selecting from a set of questions) about one or more aspects related to property assessment and the system outputs the answers in a form that can be easily understood by the user. The style of the answers (i.e., simple versus complex, using domain-specific terminology or not, using aspecific language) may also be adapted based on the user’s needs (“user-centric adaptation”). The natural language processing and / or the user-centric adaptation are performed using Al and / or machine learning methods. Such machine learning methods include for example methods employing neural networks, deep learning, transformers, statistical machine learning techniques, and other machine learning techniques that are known in the art.

[0091] In another embodiment, a method for automated or semi -automated report generation for one or more aspects of property assessment including a user interface, a computing device, a storage device, and optionally wired and / or wireless communication links for communicating with one or more remote devices and wherein a report template is generated automatically or semi-automatically by mapping user-provided questions into a file. For example, a pdf or word or excel file can be provided and, within said file, select locations can be determined as locations within the said file where questions need to be mapped. Such locations may be determined through labelled fields and / or tagged fields and / or indices and / or keywords and / or designated symbols and / or code embedded within the file. The generated report template with the mapped questions is in turn used by the system for automated or semi-automated report generation for one or more aspects of property assessment by replacing the questions in the said report template with corresponding answers that are generated by means of machine learning methods and / or chatbot. Such machine learning methods include for example natural language processing (NLP) methods employing neural networks, deep learning, transformers, statistical machine learning techniques, large language models (LLM), and / or other machine-learning based language models, and / or generative Al techniques, and / or other machine learning techniques that are known in the art.

[0092] According to one embodiment of the present disclosure, there is provided a method for generating a model to be used for automated or semi-automated report generation for one or more aspects of property assessment, the method comprising: 1) receiving training data comprising a set of questions, each question corresponding to a content item that is related to one or more aspects of property assessment, and a set of content items, in one-to-one correspondence with the questions; such set of questions and corresponding set of content items can be used for example to form training data pairs, e.g., (question, corresponding content item), wherein corresponding content item (i.e., second element in the pair for example) include information related to one or more answers to the question (i.e., first element in the pair); optionally, thetraining data may also contain additional information about one or more aspects related to property assessment (e.g., city, zone); 2) employing the training data to train an Artificial Intelligence (Al) based algorithm such as a machine learning algorithm or an Al-based system and / or device implementing a machine learning algorithm to learn a model (“learned model”) that is employed as part of the report generation system. The Al-based system (algorithm) can be a rule-based Al system (algorithm) and / or a machine learning-based system (algorithm). For the machine learning system and / or algorithm, any suitable machine learning technique can be used including, for example, a neural network, a deep learning network, a transformer, a decision tree, a random forest, a decision fem, K-nearest neighbors (K-NN), K-means, ensemble of these or a combination thereof, and / or other machine learning techniques that are known in the art. The learned model can be an NLP-based model, a Large Language Model (LLM), a Generative Pretrained Transformer (GPT) based model, a Bidirectional Encoder Representations from Transformers (BERT) based model, a Bidirectional and Autoregressive Transformers (BART) based model, a masked language model, a denoising autoencoder based model, or a combination thereof, and / or other Generative Al models that are known in the art. The learned model can comprise one or more LLMs that are known in the art such as OPENAI's GPT LLMs (e g., GPT- 3, GPT-4), META’s Large Language Model META Al (e g., LLAMA 2), TECHNOLOGY INNOVATION INSTITUTE’S FLACON LLM, STANFORD’S ALPACA LLM. The learned model can also comprise a machine learning based classification model.

[0093] According to another embodiment of the present disclosure, there is provided a method for generating a model to be used for automated or semi-automated report generation for one or more aspects of property assessment, the method comprising: 1) generating training data comprising a set of questions, each question corresponding to a content item that is related to one or more aspects of property assessment, and a set of content items, in one-to-one correspondence with the questions; such set of questions and corresponding set of content items can be used for example to form training data pairs, e.g., (question, corresponding content item), wherein corresponding content item (i.e., second element in the pair for example) include information related to one or more answers to the question (i.e., first element in the pair); optionally, the generated training data may also contain additional information about one or more aspects related to property assessment (e.g., city, zone); 2) employing the generated training data to train an Artificial Intelligence (Al) based algorithm such as a machine learning algorithm or an ALbased system and / or device implementing a machine learning algorithm to learn a model (“learned model”) that is employed as part of the report generation system. The Al-based system (algorithm) can be a rule-based Al system (algorithm) and / or a machine learning-based system (algorithm). For the machine learning system and / or algorithm, any suitable machine learning technique can be used including, for example, a neural network, a deep learning network, a transformer, a decision tree, a random forest, a decision fern, K-nearest neighbors (K-NN), K- means, ensemble of these or a combination thereof, and / or other machine learning techniques that are known in the art. The learned model can be an NLP-based model, a Large Language Model (LLM), a Generative Pretrained Transformer (GPT) based model, a Bidirectional Encoder Representations from Transformers (BERT) based model, a Bidirectional and Autoregressive Transformers (BART) based model, a masked language model, a denoising autoencoder based model, or a combination thereof, and / or other Generative Al models that are known in the art. The learned model can comprise one or more LLMs that are known in the art such as OPENAI's GPT LLMs (e g., GPT-3, GPT-4), META’s Large Language Model META Al (e g., LLAMA 2), TECHNOLOGY INNOVATION INSTITUTE’S FLACON LLM, STANFORD’S ALPACA LLM. The learned model can also comprise a machine learning based classification model. The training data can be generated using data that is stored locally on the system, data that is stored on one or more remote servers such as the cloud, user-provided data, data augmentation methods, machine learning techniques and / or a chatbot, or a combination thereof.

[0094] According to another embodiment of the present disclosure, there is provided a method to be used for automated or semi-automated report generation for one or more aspects of property assessment, the method comprising: 1) receiving a report template comprising one or more questions; 2) employing an Al model to generate answers (“generated answers”) to said one or more questions; 3) replacing said one or more questions in said report template with said generated answers to generate one or more reports

[0095] According to another embodiment of the present disclosure, there is provided a method for employing a learned model to be used for automated or semi-automated report generation for one or more aspects of property assessment, the method comprising: 1) receiving a report template comprising one or more questions; 2) employing the said learned model to generate answers (“generated answers”) to said one or more questions; 3) replacing said one ormore questions in said report template with said generated answers to generate one or more reports.

[0096] According to another embodiment of the present disclosure, there is provided a method to be used for automated or semi-automated report generation for one or more aspects of property assessment, the method comprising: 1) receiving a set of questions comprising one or more questions; 2) mapping said one or more questions from said set of questions into a report fde to produce a report template.

[0097] According to another embodiment of the present disclosure, there is provided a method to be used for automated or semi-automated report generation for one or more aspects of property assessment, the method comprising: 1) receiving a set of questions comprising one or more questions; 2) employing an Al model to produce a report template comprising said one or more questions.

[0098] According to another embodiment of the present disclosure, there is provided a method for employing a learned model to be used for automated or semi-automated report generation for one or more aspects of property assessment, the method comprising: 1) receiving a set of questions comprising one or more questions; 2) employing said learned model to produce a report template comprising said one or more questions.

[0099] According to another embodiment of the present disclosure, there is provided a method to be used for automated or semi-automated report generation for one or more aspects of property assessment, the method comprising: 1) receiving a set of questions comprising one or more questions; 2) receiving reference data or location of such reference data; 3) generating a report template comprising said one or more questions using said reference data.

[0100] According to another embodiment of the present disclosure, there is provided a method to be used for automated or semi-automated report generation for one or more aspects of property assessment, the method comprising: 1) receiving a set of questions comprising one or more questions; 2) determining reference data or location of such reference data; 3) generating a report template comprising said one or more questions using said reference data.

[0101] According to another embodiment of the present disclosure, there is provided a method to be used for automated or semi-automated report generation for one or more aspects of property assessment, the method comprising: 1) receiving a set of questions comprising one or more questions; 2) receiving reference data or location of such reference data; 3) employing anAl model to produce a report template comprising said one or more questions using said reference data.

[0102] According to another embodiment of the present disclosure, there is provided a method to be used for automated or semi-automated report generation for one or more aspects of property assessment, the method comprising: 1) receiving a set of questions comprising one or more questions; 2) determining reference data or location of such reference data; 3) employing an Al model to produce a report template comprising said one or more questions using said reference data.

[0103] According to another embodiment of the present disclosure, there is provided a method to be used for automated or semi-automated report generation for one or more aspects of property assessment, the method comprising: 1) receiving a set of questions comprising one or more questions; 2) employing a first Al model to produce a report template comprising said one or more questions from said set of questions.; 3) employing a second Al model to generate answers (“generated answers”) to said one or more questions that are contained in said report template and to replace said one or more questions in said report template with said generated answers to generate one or more reports.

[0104] According to another embodiment of the present disclosure, there is provided a method for employing a learned model to be used for automated or semi-automated report generation for one or more aspects of property assessment, the method comprising: 1) receiving a set of questions comprising one or more questions; 2) receiving reference data or location of such reference data; 3) employing said learned model and said reference data to produce a report template comprising said one or more questions.

[0105] According to another embodiment of the present disclosure, there is provided a method for employing a learned model to be used for automated or semi-automated report generation for one or more aspects of property assessment, the method comprising: 1) receiving a set of questions comprising one or more questions; 2) determining reference data or location of such reference data; 3) employing said learned model and said reference data to produce a report template comprising said one or more questions.

[0106] According to another embodiment of the present disclosure, there is provided a method for employing a learned model to be used for automated or semi-automated report generation for one or more aspects of property assessment, the method comprising: 1) receivinga set of questions comprising one or more questions; 2) mapping one or more questions from said set of questions into a report file to produce a report template; 3) employing said learned model to generate answers (“generated answers”) to said one or more questions that are contained in said report template; 4) replacing said one or more questions in said report template with said generated answers to generate one or more reports.

[0107] According to another embodiment of the present disclosure, there is provided a method for employing a first learned model and a second learned model to be used for automated or semi-automated report generation for one or more aspects of property assessment, the method comprising: 1) receiving a set of questions comprising one or more questions; 2) employing said first learned model to produce a report template comprising said one or more questions from said set of questions.; 3) employing said second learned model to generate answers (“generated answers”) to said one or more questions that are contained in said report template; 4) replacing said one or more questions in said report template with said generated answers to generate one or more reports.

[0108] According to another embodiment of the present disclosure, there is provided a method for employing a learned model to be used for automated or semi-automated report generation for one or more aspects of property assessment, the method comprising: 1) receiving a report template comprising one or more questions; 2) employing said learned model to generate one or more reports by replacing said one or more questions in said report template with generated answers.

[0109] According to another embodiment of the present disclosure, there is provided a method for employing a learned model to be used for automated or semi-automated report generation for one or more aspects of property assessment, the method comprising: 1) receiving a set of questions comprising one or more questions; 2) mapping said one or more questions into a report file to produce a report template; 3) employing said learned model to generate one or more reports by replacing said one or more questions in said report template with generated answers.

[0110] According to another embodiment of the present disclosure, there is provided a method for employing a first learned model and a second learned model to be used for automated or semi-automated report generation for one or more aspects of property assessment, the method comprising: 1) receiving a set of questions comprising one or more questions; 2) employing saidfirst learned model to produce a report template comprising said one or more questions; 3) employing said second learned model to generate one or more reports by replacing said one or more questions in said report template with generated answers.

[0111] According to another embodiment of the present disclosure, there is provided a method for employing a learned model to be used for automated or semi-automated report generation for one or more aspects of property assessment, the method comprising: 1) receiving a set of questions comprising one or more questions; 2) employing said learned model to generate one or more reports by generating answers (“generated answers”) to said one or more questions and by mapping said generated answers into one or more report files.

[0112] According to another embodiment of the present disclosure, there is provided a method for employing a learned model to be used for one or more aspects of property assessment, the method comprising: 1) receiving one or more questions; 2) receiving reference data or location of such reference data; 3) employing said learned model to generate answers (“generated answers”) to said one or more questions using said reference data. The said reference data can be provided as a document and / or file and / or script and / or code, or using an address of one or more local and / or remote locations (e.g, remote server or website or storage device) at which the data can be obtained, or a combination thereof. The generated answer(s) to each question can be output using a user interface, or by a chatbot, or included in a document and / or file, or stored on a local and / or remote storage device, or a combination thereof. The document and / or file can be a text file, and / or word file, and / or excel file and / or pdf file or any other human-readable format. The document can also be a non-human readable format and / or can be intended to be readable by a software package running on a local and / or remote (e.g., cloud) computing device (e.g, AUTOCAD, CIVIL 3D, AUTODESK, ORACLE, ARCGIS, FFIEC GEOCODING / MAPPING SYSTEM, and / or other commercial software and / or non-commercial software and / or custom software packages) and / or interpretable via a machine and / or machinelearning based device. The document can also be a combination of human-readable and non- human readable formats.

[0113] According to another embodiment of the present disclosure, there is provided a method for employing a learned model to be used for one or more aspects of property assessment, the method comprising: 1) receiving one or more questions; 2) determining reference data or location of such reference data; 3) employing said learned model to generateanswers (“generated answers”) to said one or more questions using said reference data. The said reference data can be a document and / or file and / or script and / or code, or can be stored at one or more local (e.g., local server or local storage device) and / or remote locations (e.g., remote server or website or cloud, or remote storage device), or a combination thereof. The said reference data can also be obtained from one or more software packages running on a local and / or remote (e.g., cloud) computing device (e.g., AUTOCAD, CIVIL 3D, AUTODESK, ORACLE, ARCGIS, FFIEC GEOCODING / MAPPING SYSTEM, and / or other commercial software and / or noncommercial software and / or custom software packages). The said reference data may be a generated report according to techniques of this disclosure. The generated answer(s) to each question can be output using a user interface, or by a chatbot, or included in a document and / or file, or stored on a local and / or remote storage device, or a combination thereof. The document and / or file can be a text file, and / or word file, and / or excel file and / or pdf file or any other human-readable format. The document can also be a non-human readable format and / or can be intended to be readable by a software package running on a local and / or remote (e.g., cloud) computing device (e g., AUTOCAD, CIVIL 3D, AUTODESK, ORACLE, ARCGIS, FFIEC GEOCODING / MAPPING SYSTEM, and / or other commercial software and / or non-commercial software and / or custom software packages) and / or interpretable via a machine and / or machinelearning based device. The document can also be a combination of human-readable and non- human readable formats.

[0114] According to another embodiment of the present disclosure, there is provided a method for employing a learned model to be used for automated or semi-automated report generation for one or more aspects of property assessment, the method comprising: 1) receiving a set of questions comprising one or more questions; 2) receiving reference data or location of such reference data; 3) employing said learned model to generate one or more reports by generating answers (“generated answers”) to said one or more questions using said reference data and by mapping said generated answers into one or more report files.

[0115] According to another embodiment of the present disclosure, there is provided a method for employing a learned model to be used for automated or semi-automated report generation for one or more aspects of property assessment, the method comprising: 1) receiving a set of questions comprising one or more questions; 2) determining reference data or location of such reference data; 3) employing said learned model to generate one or more reports bygenerating answers (“generated answers”) to said one or more questions using said reference data and by mapping said generated answers into one or more report fdes.

[0116] According to another embodiment of the present disclosure, there is provided a method for employing a first learned model and a second learned model to be used for automated or semi-automated report generation for one or more aspects of property assessment, the method comprising: 1) receiving a set of questions comprising one or more questions; 2) receiving a first reference data and a second reference data or corresponding locations of such first and second reference data; 3) employing said first learned model to produce a report template comprising said one or more questions using said first reference data; 4) employing said second learned model to generate answers (“generated answers”) to said one or more questions that are contained in said report template using second reference data; 5) replacing said one or more questions in said report template with said generated answers to generate one or more reports.

[0117] According to another embodiment of the present disclosure, there is provided a method for employing a first learned model and a second learned model to be used for automated or semi -automated report generation for one or more aspects of property assessment, the method comprising: 1) receiving a set of questions comprising one or more questions; 2) receiving a first reference data and a second reference data or corresponding locations of such first and second reference data; 3) employing said first learned model to produce a report template comprising said one or more questions using said first reference data ; 4) employing said second learned model to generate one or more reports by replacing said one or more questions in said report template with generated answers using said second reference data.Example embodiments

[0118] Example 1. A method of generating requirement information for quality assessment and control, the method comprising:

[0119] receiving training data comprising:

[0120] a set of questions, each question corresponding to a requirement content item that is related to one or more aspects of quality assessment and control;

[0121] a set of requirement content items, in one-to-one correspondence with the questions;

[0122] using the training data and an Artificial Intelligence (Al) based algorithm to generate a learned Al model;

[0123] using the learned Al model to generate requirement information.

[0124] Example 2. The method of example 1, further comprising using the learned AT model to generate requirement information for determining requirements.

[0125] Example 3. The method of example 1, further comprising using the learned Al model to generate requirement information for quality assessment and control.

[0126] Example 4. The method of example 2, further comprising performing quality assessment and control based on the determined requirements.

[0127] Example 5. The method of example 1, wherein the generated requirement information comprises information relating to one or more of requirements, laws, regulations, due diligence, quality assessment and quality control.

[0128] Example 6. The method of example 5, wherein the generated requirement information relates to one or more of land development, environment, zoning, and property assessment.

[0129] Example 7. The method of example 1, wherein the Al based algorithm is a machine learning algorithm and the learned Al model is a learned machine learning model.

[0130] Example 8. The method of example 1, wherein the Al based algorithm comprises one or more of the following: a large language model, a neural network, a deep neural network, a transformer, a decision tree, a random forest, a decision fern, a support vector machine, an adversarial generative network (GAN).

[0131] Example 9. The method of example 1, wherein the training data is generated using one or more of the following: user-provided data, data augmentation, Al-based algorithms, embedding based matching,

[0132] Example 10. The method of example 5, wherein the generated requirement information is provided using a user interface.

[0133] Example 11. The method of example 10, wherein the user interface comprises one or more of the following: text-based user interface, a graphical user interface (GUI), a touch screen user interface.

[0134] Example 12. The method of example 5, wherein the generated requirement information is provided on a storage device.

[0135] Example 13. The method of example 5, wherein the generated requirement information is provided in a file according to a template file.

[0136] Example 14. The method of example 13, wherein the template file is automatically determined by means of a classification procedure based on one or more of the set of questions or reference data.

[0137] Example 15. The method of example 13, wherein the template file is generated using an Al algorithm based on one or more of the set of questions or reference data.

[0138] Example 16. The method of example 13, wherein the set of questions are mapped into the template file using a mapping procedure that determines the locations to which each question is mapped within the template file.

[0139] Example 17. The method of example 15, wherein the mapping procedure employs one or more of the following: labelled fields, tagged fields, indices, keywords, designated symbols, code embedded within the template file.

[0140] Example 18. The method of example 15, wherein the mapping procedure comprises one or more of the following: mapping table, mapping pointers, mapping indices, mapping function.

[0141] Example 19. The method of example 15, wherein the mapping procedure employs an Al algorithm.

[0142] Example 20. The method of example 16, further comprising generating answers to the set of questions using the learned Al model and replacing each question in the template file by its corresponding generated answer to obtain a requirement information file.

[0143] Example 21. The method of example 1, further comprising generating requirement information by using the learned Al model to generate answers to said one or more questions.

[0144] Example 22. The method of example 21, further comprising using reference data together with the learned Al model to generate answers to said one or more questions.

[0145] Example 23. A method of generating requirement information for quality assessment and control, the method comprising:

[0146] receiving a learned Al model that is trained to generate requirement information;

[0147] receiving a set of questions comprising one or more questions, each question corresponding to one or more requirement content items that are related to one or more aspects of quality assessment and control;

[0148] using the learned Al model to generate requirement information.

[0149] Example 24. The method of example 23, further comprising using the learnedAl model to generate requirement information for determining requirements.

[0150] Example 25. The method of example 23, further comprising using the learnedAl model to generate requirement information for quality assessment and control.

[0151] Example 26. The method of example 24, further comprising performing quality assessment and control based on the determined requirements.

[0152] Example 27. The method of example 23, wherein the generated requirement information comprises information relating to one or more of requirements, laws, regulations, due diligence, quality assessment and quality control.

[0153] Example 28. The method of example 27, wherein the generated requirement information relates to one or more of land development, environment, zoning, and property assessment.

[0154] Example 29. The method of example 23, wherein the learned Al model is a learned machine learning model.

[0155] Example 30. The method of example 23, wherein the learned Al model comprises one or more of the following: a large language model, a neural network, a deep neural network, a transformer, a decision tree, a random forest, a decision fern, a support vector machine, an adversarial generative network (GAN).

[0156] Example 31. The method of example 27, wherein the generated requirement information is provided using a user interface.

[0157] Example 32. The method of example 31, wherein the user interface comprises one or more of the following: text-based user interface, a graphical user interface (GUI), a touch screen user interface.

[0158] Example 33. The method of example 27, wherein the generated requirement information is provided on a storage device.

[0159] Example 34. The method of example 27, wherein the generated requirement information is provided in a file according to a template file.

[0160] Example 35. The method of example 34, wherein the template file is automatically determined by means of a classification procedure based on one or more of the set of questions or reference data.

[0161] Example 36. The method of example 34, wherein the template file is generated using an Al algorithm based on one or more of the set of questions or reference data.

[0162] Example 37. The method of example 34, wherein the set of questions are mapped into the template file using a mapping procedure that determines the locations to which each question is mapped within the template file.

[0163] Example 38. The method of example 37, wherein the mapping procedure employs one or more of the following: labelled fields, tagged fields, indices, keywords, designated symbols, code embedded within the template file.

[0164] Example 39. The method of example 37, wherein the mapping procedure comprises one or more of the following: mapping table, mapping pointers, mapping indices, mapping function.

[0165] Example 40. The method of example 37, wherein the mapping procedure employs an Al algorithm.

[0166] Example 4E The method of example 36, further comprising generating answers to the set of questions using the learned Al model and replacing each question in the template file by its corresponding generated answer to obtain a requirement information file.

[0167] Example 42. The method of example 23, further comprising generating requirement information by using the learned Al model to generate answers to said one or more questions.

[0168] Example 43. The method of example 42, further comprising using reference data together with the learned Al model to generate answers to said one or more questions.

[0169] Example 44. A method of generating requirement information for quality assessment and control, the method comprising: receiving a set of questions comprising one or more questions, each question corresponding to one or more requirement content items that are related to one or more aspects of quality assessment and control; using a first Al model to generate a template file based on one or more of the set of questions, wherein the set of questions are mapped into the template file using a mapping procedure that determines the locations to which each question is mapped within the template file; using a second Al model to generate answers to the set of questions; andreplacing each question in the template file by its corresponding generated answer to obtain a requirement information file.

[0170] Example 45. A method of generating requirement information for quality assessment and control, the method comprising: receiving a set of questions comprising one or more questions, each question corresponding to one or more requirement content items that are related to one or more aspects of quality assessment and control; receiving reference data comprising first reference data and second reference data; using a first Al model to generate a template file based on one or more of the set of questions and the first reference data, wherein the set of questions are mapped into the template file using a mapping procedure that determines the locations to which each question is mapped within the template file. using a second Al model to generate answers to the set of questions based on the second reference data; and replacing each question in the template file by its corresponding generated answer to obtain a requirement information file.

[0171] Although the invention has been described and illustrated in the foregoing illustrative embodiments, it is understood that the present disclosure has been made only by way of example, and that numerous changes in the details of implementation of the invention can be made without departing from the spirit and scope of the invention, which is limited only by the examples that follow. Features of the disclosed embodiments can be combined and rearranged in various ways.

Claims

What is claimed is:

1. A system generating a document, comprising: memory; and at least one hardware processor collectively configured to at least: suse a first one or more Al algorithms to generate one or more answers to one or more questions included in a template; and replace the one or more questions in the template with the one or more answers to produce the document.

2. The system of claim 1, wherein, in using the first one or more Al algorithms to generate the one or more answers to the one or more questions, the at least one hardware processor makes use of reference data.

3. The system of claim 1, wherein the at least one hardware processor is further configured to: receive the one or more questions; and generate the template based on the one or more questions.

4. The system of claim 3, wherein each of the one or more questions corresponds to one or more requirement content items that are related to one or more aspects of a subject of the document.

5. The system of claim 3, wherein, in generating the template based on the one or more questions, the at least one hardware processor uses a second one or more Al algorithms.

6. The system of claim 5, wherein, in using the second one or more Al algorithms, the at least one hardware processor makes use of reference data.

7. The system of claim 3, wherein, in generating the template based on the one or more questions, the at least one hardware processor uses a classification procedure.

8. The system of claim 3, wherein, in generating the template based on the one or more questions, the at least one hardware processor maps the one or more questions into the template based on at least one of a labelled field, a tagged field, an index, a keyword, a designated symbol, a mapping table, a mapping pointer, a mapping function, and code embedded within the template.

9. The system of claim 1, wherein the first one or more Al algorithms comprises one or more of the following: a large language model; a neural network; a deep neural network; a transformer; a decision tree; a random forest; a decision fem; a support vector machine; and an adversarial generative network (GAN).

10. The system of claim 5, wherein the second one or more Al algorithms comprises one or more of the following: a large language model; a neural network; a deep neural network; a transformer; a decision tree; a random forest; a decision fem; a support vector machine; and an adversarial generative network (GAN).

11. A method of generating a document, comprising: using a first one or more Al algorithms to generate one or more answers to one or more questions included in a template; and replacing the one or more questions in the template with the one or more answers to produce the document using at least one hardware processor.

12. The method of claim 11, wherein using the first one or more Al algorithms to generate the one or more answers to the one or more questions makes use of reference data.

13. The method of claim 11, further comprising: receiving the one or more questions; and generating the template based on the one or more questions.

14. The method of claim 13, wherein each of the one or more questions corresponds to one or more requirement content items that are related to one or more aspects of a subject of the document.

15. The method of claim 13, wherein generating the template based on the one or more questions comprises using a second one or more Al algorithms.

16. The method of claim 15, wherein using the second one or more Al algorithms to generate the template makes use of reference data.

17. The method of claim 13, wherein generating the template based on the one or more questions comprises using a classification procedure.

18. The method of claim 13, wherein generating the template comprises mapping the one or more questions into the template based on at least one of a labelled field, a tagged field, an index, a keyword, a designated symbol, a mapping table, a mapping pointer, a mapping function, and code embedded within the template.

19. The method of claim 11, wherein the first one or more Al algorithms comprises one or more of the following: a large language model; a neural network; a deep neural network; a transformer; a decision tree; a random forest; a decision fern; a support vector machine; and an adversarial generative network (GAN).

20. The method of claim 15, wherein the second one or more Al algorithms comprises one or more of the following: a large language model; a neural network; a deep neural network; a transformer; a decision tree; a random forest; a decision fem; a support vector machine; and an adversarial generative network (GAN).

21. A non-transitory computer-readable medium containing computer executable instructions that, when executed by a processor, cause the processor to perform a method for generating a document, the method comprising:using a first one or more Al algorithms to generate one or more answers to one or more questions included in a template; and replacing the one or more questions in the template with the one or more answers to produce the document.

22. The non-transitory computer-readable medium of claim 21, wherein using the first one or more Al algorithms to generate the one or more answers to the one or more questions makes use of reference data.

23. The non-transitory computer-readable medium of claim 21, wherein the method further comprises: receiving the one or more questions; and generating the template based on the one or more questions.

24. The non-transitory computer-readable medium of claim 23, wherein each of the one or more questions corresponds to one or more requirement content items that are related to one or more aspects of a subject of the document.

25. The non-transitory computer-readable medium of claim 23, wherein generating the template based on the one or more questions comprises using a second one or more Al algorithms.

26. The non-transitory computer-readable medium of claim 25, wherein using the second one or more Al algorithms to generate the template makes use of reference data.

27. The non-transitory computer-readable medium of claim 23, wherein generating the template based on the one or more questions comprises using a classification procedure.

28. The non-transitory computer-readable medium of claim 23, wherein generating the template comprises mapping the one or more questions into the template based on at least one ofa labelled field, a tagged field, an index, a keyword, a designated symbol, a mapping table, a mapping pointer, a mapping function, and code embedded within the template.

29. The non-transitory computer-readable medium of claim 21, wherein the first one or more Al algorithms comprises one or more of the following: a large language model; a neural network; a deep neural network; a transformer; a decision tree; a random forest; a decision fern; a support vector machine; and an adversarial generative network (GAN).

30. The non-transitory computer-readable medium of claim 25, wherein the second one or more Al algorithms comprises one or more of the following: a large language model; a neural network; a deep neural network; a transformer; a decision tree; a random forest; a decision fern; a support vector machine; and an adversarial generative network (GAN).

Images