Knowledge graph driven content generation

Abstract

Embodiments of the present disclosure relate to knowledge graph driven content generation. Embodiments are provided relating to computer systems, computer program products, and computer-implemented methods for dynamically managing a knowledge graph and its corresponding dataset. Embodiments include identifying new words from a virtual environment, and resolving the meaning of the identified new words with virtual environment exploration. The resolved meaning of the new words is applied to the dynamic expansion of the dataset and corresponding knowledge graph.

Description

Technical Field

[0001] This embodiment relates to an artificial intelligence (AI) platform and associated methods that support the automated management of knowledge graphs (KGs). More specifically, the embodiment relates to the exploration of natural language (NL) dialogues and the identification of one or more new words therein, as well as the management of the KG to resolve the identified new words(s). Background Technology

[0002] Language evolution is the application of evolutionary theory to language research. As a language develops, its vocabulary and usage change. There are generally two stages in language evolution: protologism and neologism. Protologism is the precursor to the neologism stage. More specifically, protologism refers to the stage where new or unestablished words are proposed to a limited group, with the hope that they will become acceptable. Protologisms become neologisms when one or more words or expressions are pronounced differently. In one embodiment, neologisms are also called coinages. New neologisms emerge as language and its usage evolve. Not all neologisms are entirely new. For example, a neologism can be a new use of an old word, or in one embodiment, a neologism can be generated from a new combination of existing words. The following are some examples of neologisms: webinar, malware, and blogosphere, where webinar refers to a seminar or discussion on the internet, malware refers to software designed to interfere with the normal functioning of a computer, and blogosphere refers to blogs or bloggers on the internet. These are just a few examples of new words.

[0003] New words are often driven by cultural and technological changes and are primarily collected or identified through manual detection of new word usage (e.g., slang, idioms, technical terms, etc.). Manual detection or identification requires labor and is therefore expensive. Similarly, machine learning models have been developed to collect new words by crawling social media. These machine learning models use pre-trained language models to extract relationships between words. However, relationship extraction is limited because the model can only extract relationships that are defined or present in one or more labels within the corresponding dataset. Therefore, there is a need in the art to develop an apparatus or process for automatically detecting new words without the limitations of manual or machine learning technology solutions. Summary of the Invention

[0004] These embodiments include systems, computer program products, and methods for dynamically expanding knowledge graphs using virtual environments. This overview is provided to present, in a simplified form, a selection of representative concepts further described in the detailed description. This overview is not intended to identify key or essential features of the claimed subject matter, nor is it intended to be used in any way that would limit the scope of the claimed subject matter.

[0005] In one aspect, a computer system having a processing unit and memory is provided for use with an artificial intelligence (AI) computer platform for the selective and dynamic expansion of a dataset. The processing unit is operatively coupled to the memory and communicates with the AI ​​platform and embedded tools, including a token manager and a bootstrap. The token manager is configured to identify new words in a virtual environment exploration and evaluate the new words against a representation in the dataset. The evaluation includes discovering the absence or presence of the new word in the dataset. In an exemplary embodiment, the evaluation includes the token manager verifying the identified token using two or more virtual explorations, where each exploration generates a corroboration value for the identified token. Through this discovery, if the representation is not found in the dataset, a potential meaning of the token representation of the new word is obtained from the dataset, and if the representation is found in the dataset, a new meaning is extracted from the dataset. The bootstrap is configured to dynamically modify the dataset based on the token manager's discovery. The dynamic modification includes updating the knowledge graph representation of the dataset corresponding to the modified dataset. The knowledge graph update includes adding the identified token as a new node in the knowledge graph, adding the new meaning as a new edge in the knowledge graph, or a combination thereof.

[0006] On the other hand, a computer program product having a computer-readable storage medium is provided, on which program code is stored. The program code is executable by a computer processor to support the selectivity and dynamic expansion of a dataset. Program code is provided to identify new words in virtual environment exploration and to evaluate the new words against a representation in the dataset. The evaluation includes discovering the absence or presence of the new word in the dataset. In an exemplary embodiment, the evaluation includes verifying an identified token using two or more virtual explorations, wherein each exploration generates a confirmation value for the identified token. Through this discovery, if the representation is not found in the dataset, a potential meaning of the token representation of the new word is obtained from the dataset, and if the representation is found in the dataset, a new meaning is extracted from the dataset. Further, program code is provided to dynamically modify the dataset based on the discovery. Dynamic modification includes updating the knowledge graph representation of the dataset corresponding to the modified dataset. Knowledge graph updating includes program code that adds the identified token as a new node in the knowledge graph, adds the new meaning as a new edge in the knowledge graph, or a combination thereof.

[0007] In another aspect, a computer-implemented method is provided for selectively and dynamically expanding a dataset. The method is configured to identify new words in virtual environment explorations and evaluate the new words against a representation in the dataset. Evaluation includes discovering the absence or presence of new words in the dataset. In an exemplary embodiment, evaluation includes a token manager verifying identified tokens using two or more virtual explorations, where each exploration generates a confirmation value for the identified token. Through this discovery, if the representation is not found in the dataset, a potential meaning of the token representation of the new word is obtained from the dataset, and if the representation is found in the dataset, a new meaning is extracted from the dataset. The method is configured to dynamically modify the dataset based on the discovery. Dynamic modification includes updating the knowledge graph representation of the dataset corresponding to the modified dataset. Knowledge graph updates include adding identified tokens as new nodes in the knowledge graph, adding new meanings as new edges in the knowledge graph, or a combination thereof.

[0008] These and other features and advantages will become clear from the following detailed description of (a plurality of) presently preferred embodiments in conjunction with the accompanying drawings. Attached Figure Description

[0009] The reference numerals in the accompanying drawings form part of this specification. Unless otherwise expressly indicated, the features shown in the drawings are intended to illustrate some embodiments only, and not all embodiments.

[0010] Figure 1 A system diagram is depicted showing the computing systems and tools of an artificial intelligence platform that support dynamic knowledge graph management and expansion.

[0011] Figure 2 The description is shown as follows Figure 1 The diagram shows the artificial intelligence platform and related tools and their related application programming interfaces.

[0012] Figure 3 A flowchart illustrating the process of selectively and dynamically modifying datasets and corresponding knowledge graphs using a virtual communication environment is depicted.

[0013] Figure 4 A block diagram illustrating an example of a cloud-based supporting computer system / server is provided to achieve the above-mentioned... Figures 1 to 3 The system and process described.

[0014] Figure 5 A block diagram illustrating a cloud computing environment is depicted.

[0015] Figure 6 A block diagram illustrating a set of functional abstraction model layers provided by a cloud computing environment is depicted. Detailed Implementation

[0016] It will be readily understood that the components of the embodiments of the invention as generally described and illustrated in the accompanying drawings can be arranged and designed in a variety of different configurations. Therefore, the following detailed description of embodiments of the apparatus, systems, methods, and computer program products of this invention as presented in the drawings is not intended to limit the scope of the claimed embodiments, but rather represents only selected embodiments.

[0017] Throughout this specification, references to "selected embodiment," "an embodiment," or "embodiment" mean that a particular feature, structure, or characteristic described in connection with that embodiment is included in at least one embodiment. Therefore, the phrases "selected embodiment," "in an embodiment," or "in an embodiment" appearing in various places throughout this specification do not necessarily refer to the same embodiment.

[0018] The illustrated embodiments will be best understood by referring to the accompanying drawings, wherein like parts are indicated by like numbers throughout. The following description is intended to be illustrative only and simply illustrates certain selected embodiments of devices, systems, and processes consistent with the embodiments claimed herein.

[0019] In the field of artificial intelligence computing, natural language processing (such as IBM) Artificial intelligence (AI) computers (or other natural language processing systems) use the acquired knowledge to process natural language. To process natural language, the system can be trained with data derived from databases or knowledge corpora, but the results may be incorrect or inaccurate for various reasons.

[0020] Machine learning (ML), a subset of artificial intelligence (AI), uses algorithms to learn from data and create predictions based on that data. AI refers to the intelligence of information-based machines that can make decisions that maximize the chances of success on a given topic. More specifically, AI can learn from datasets to solve problems and provide relevant recommendations. Cognitive computing is a hybrid of computer science and cognitive science. It utilizes self-learning algorithms that use data minimization, visual recognition, and natural language processing to solve problems and optimize human processes.

[0021] At the heart of AI and related reasoning lies the concept of similarity. Understanding natural language and objects requires reasoning from a relational perspective, which can be challenging. Structures, including static and dynamic structures, prescribe definite outputs or actions for a given set of inputs. More specifically, the determined outputs or actions are based on expressions or inherent relationships within the structure. This arrangement is satisfactory for the chosen environment and conditions. However, it should be understood that dynamic structures are inherently susceptible to change, and the outputs or actions can be changed accordingly. Existing solutions for effectively identifying objects and understanding natural language, as well as handling changes in structure in response to identification and understanding, are extremely difficult at a practical level.

[0022] The automated virtual conversational agent, referred to herein as a chatbot, uses artificial intelligence (AI) as a platform to conduct natural language (NL) interactions between the automated virtual conversational agent and users, typically such as consumers or customers, or even other conversational agents. Interactions can involve product sales, customer service, information retrieval, or other types of interactions or transactions. Chatbots interact with users through dialogue, typically text-based (e.g., online or via text) or auditory (e.g., via telephone). Chatbots are known in the art as question-and-answer components between users and AI platforms. Chatbots can search for answers to presented questions from knowledge sources, ask for clarity, or, in embodiments, guide the user to a person.

[0023] As shown and described herein, systems, computer program products, and methods are provided to automatically update or modify existing datasets, and in embodiments, to update corresponding knowledge graphs to reflect the updated or modified datasets, thereby addressing and supporting new words. It will be understood in the art that a knowledge graph is a representation of a knowledge base or dataset that integrates data using a graph-structured data model or topology. A knowledge graph represents knowledge as content and concepts, and the relationships between these contents and concepts, in a graphical format. In embodiments, the knowledge graph representation of the dataset includes both human- and computer-readable ontologies, where concepts or objects (also referred to herein as content) are represented as nodes, and relationships between concepts or objects are represented as edges or links. As shown and described herein, computer systems, computer program products, and computer-implemented methods are provided to efficiently detect, collect, and validate new words, and to dynamically expand or modify existing or corresponding datasets and knowledge graphs associated with validated new words.

[0024] refer to Figure 1A schematic diagram of an artificial intelligence platform computing system (100) with tools supporting dynamic knowledge graph management and expansion is described. As shown, a server (110) is provided that communicates with multiple computing devices (180), (182), (184), (186), (188), and (190) across a network connection (105). The server (110) is configured with a processing unit (112) that communicates with a memory (116) across a bus (114). The server (110) is shown to have an artificial intelligence (AI) platform (150) configured with one or more tools to support and allow dynamic expansion of a dataset, which in one embodiment is configured or represented as a knowledge graph. The server (110) communicates with one or more of the computing devices (180), (182), (184), (186), (188), and (190) via the network (105). More specifically, computing devices (180), (182), (184), (186), (188), and (190) communicate with each other and with other devices or components via one or more wired and / or wireless data communication links, wherein each communication link may include one or more of wires, routers, switches, transmitters, receivers, etc. In this networking arrangement, the server (110) and network connection (105) enable communication detection, identification, and resolution. Other embodiments of the server (110) may be used with components, systems, subsystems, and / or devices other than those described herein.

[0025] The AI ​​platform (150) illustrated herein is configured to receive input (102) from various sources. For example, the knowledge engine (150) may receive input across a network (105) and / or utilize a knowledge base (160), which is also referred to herein as a corpus or data source. As shown, the knowledge base (160) is configured with one or more libraries. For illustrative purposes, the knowledge base (160) is shown herein with two libraries, referred to herein as the first library. A (162 A ) and the second library, library B (162 B However, the number of libraries should not be considered limiting. First library A (162 A A dataset is configured to store one or more datasets, referred to in the art as a collection of related, discrete items of related data, which can be accessed individually or in combination, or managed as a whole entity. As an example, the first library (162...) A ) is shown as having a dataset 0,0 (164 0,0 ), dataset 0,1 (164 0,1 ... and dataset 0,N(164 0,N ), and the second library, library B (162 B The dataset is shown. 1,0 (164 1,0 ), dataset 1,1 (164 1,1 ...and dataset 1,N (164 1,N In the embodiment, the first library, the library A (162A) and the second library, library B (162 B The knowledge base (160) can include either a reduced number of datasets or an expanded number of datasets. Similarly, in one embodiment, the knowledge base (160) can include multiple libraries organized or subordinated by a common subject or topic, although this is not required. In one embodiment, a second library (162) B This can be located in a separate knowledge base (not shown). The first knowledge base (162) A The datasets represented in the diagram are shown separately using the corresponding knowledge graphs. For example, the dataset... 0,0 (164 0,0 Using Knowledge Graph KG 0,0 (166 0,0 The dataset is shown. 0,1 (164 0,1 Using Knowledge Graph KG 0,1 (166 0,1 This shows... dataset 0,N (164 0,N Using knowledge graph KG 0,N (166 0,N The dataset is shown. 1,0 (164 1,0 Using Knowledge Graph KG 1,0 (166 1,0 (As shown in the image). Dataset 1,1 (164 1,1 Using Knowledge Graph KG 1,1 (166 1,1 This shows... dataset 1,N (164 1,N Using knowledge graph KG 1,N (166 1,NAs shown in the figure. In an embodiment, the knowledge base (160) may be pre-populated with a dataset and a corresponding knowledge graph. Similarly, in one embodiment, a knowledge graph may be created for the corresponding dataset, which includes extracting and analyzing one or more phrases from the corresponding knowledge article or dataset using natural language processing (NLP), the extracted and analyzed phrases relating to one or more nouns, such as physical objects. NLP is also used to identify one or more relation words between the extracted phrases. The identified objects are assigned to one or more nodes, and one or more relation words are assigned to one or more corresponding edges. In one embodiment, the relation word describes the connection between objects annotated or referenced in the extracted phrases. Tools for creating knowledge graphs are known in the art and may be utilized in an AI platform (150) in one embodiment.

[0026] The AI ​​platform (150) is equipped with tools that support and allow selective and dynamic modification of datasets, which in this embodiment is an extended form of dataset selection and dynamic modification. In an exemplary embodiment, one or more datasets and their corresponding knowledge graph representations in the knowledge base (160) can be transmitted to the server (110) via a network (105). Tools include, but are not limited to, an exploration manager (152), a token manager (154), a bootstrapper (156), and a neural network manager (158). The AI ​​platform (150) can receive input from the network (105) or utilize the knowledge base (160) to selectively and dynamically manage datasets and corresponding knowledge graphs to support content generation and maintenance.

[0027] Natural Language Processing (NLP) is a branch of computer science, more specifically, Artificial Intelligence (AI), which addresses the ability of computer programs to understand human language when written and spoken. NLP combines computational linguistics with statistical, machine learning, and deep learning models to process human language in the form of text or speech data and understand its full meaning. As will be understood in the art, a chatbot is a computer program that uses AI and NLP to understand conversations (e.g., questions) and respond automatically, thereby simulating human conversation. As shown here, a server (110) is operatively coupled to a dialogue system (170) and a corresponding or embedded virtual dialogue agent (172), such as a chatbot. In an exemplary embodiment, tools (152)-(158) interface with the chatbot (172) as a virtual environment for data exploration. Similarly, in embodiments, tools (152)-(158) interface with multiple virtual environments, such as two or more chatbot interfaces, for verifying new words.

[0028] The exploration manager (152) is configured to explore virtual environments, which in this embodiment are virtual locations, including but not limited to virtual conversational environments such as chatbots, one or more crowdsourced dictionaries, distributed repositories, or combinations thereof. In this embodiment, NLP supports the functionality of the exploration manager (152). In an exemplary embodiment, the exploration manager (152) identifies at least one dataset in the knowledge base (160) that is associated with the virtual environment being explored. For example, a first repository, i.e., a library A (160 A The knowledge base (160) may include one or more datasets that are individually related to various aspects of information technology (IT), with each dataset targeting a different subset of IT. As described above, each dataset in the knowledge base (160) is associated with or has an associated knowledge graph. Each knowledge graph includes nodes and edges, where each node represents a physical object or component referenced or identified in the dataset, and each edge represents a state characteristic of the physical object or component.

[0029] A token manager (154), operably coupled to the exploration manager (152), serves as a tool for identifying one or more tokens representing new words in the explored virtual environment. Tokens are referred to in the art as instances of character sequences grouped together as semantic units. In an exemplary embodiment, the token manager (154) is responsible for tokenizing data in the explored virtual environment, a process of converting data segments into strings. For example, in an embodiment, a data segment may be a word or phrase, and a single token represents a single word or phrase. In connection with virtual environment exploration, the token manager (154) may encounter words and phrases that can represent or are actually new words. In addressing the challenge of new words, or in an embodiment, a possible new word is one whose meaning is not understood or identified by the encountered word or phrase. To address new word or possible new word encounters, the token manager (154) identifies an appropriate or corresponding dataset from a data source (160) or across a network (105) and processes the new word or possible new word against the dataset. In an exemplary embodiment, the token manager (154) tokenizes encountered new words or potential new words and processes the tokens in light of the corresponding dataset or the dataset tokenized in the embodiment. For example, the token manager (154) may find potential new words in the dataset and represent them in the corresponding knowledge graph, and discard the encounter as a new word, or it may not find a node representing a potential new word in the knowledge graph and designate the potential new word as a new word. Thus, the token manager (154) uses new word identifiers to bridge virtual environment exploration.

[0030] In the fields of NLP and virtual environment exploration, it is understood that words or phrases may be misspelled. A token manager (154) is configured to address this aspect, and more specifically, to check the spelling of tokens to ensure that the natural language representation is correct, which in this embodiment includes spell correction. If the token manager (154) performs spell correction on one or more tokens, the token manager evaluates the corrected words or phrases against a dataset. Thus, in one embodiment, the spelling of (multiple) tokens is validated to ensure the correct processing of new words.

[0031] In addition to identifying initial or potential new words, the token manager (154) is configured to perform an evaluation of potential new words. As described herein, and Figure 3 As further illustrated, the token manager (154) evaluates potential new words against a corresponding dataset, or, in one embodiment, against a knowledge graph representation of the dataset. In an exemplary embodiment, the token manager (154) tokenizes potential new words, such that the evaluation utilizes the token representation of the new word. It should be understood that the evaluation will result in the token manager (154) finding that there is no representation of the new word in the knowledge graph, i.e., no corresponding knowledge graph node, or that a representation exists but there is no corresponding meaning or association in the knowledge graph, i.e., there is a corresponding knowledge graph node but no corresponding edge. If no representation of the new word is found in the knowledge graph, the token manager (154) identifies the potential meaning of the new word from the dataset, and if a new word is found in the knowledge graph but no corresponding edge is found or identified, the token manager (154) extracts the new meaning of the new word from the knowledge graph. Details regarding the discovery of new words in the knowledge graph are provided below. Figure 3 It is shown and described in the text.

[0032] In an exemplary embodiment, the token manager (154) verifies a potential new word or subjectes a potential new word to a verification process that includes at least two explorations, each exploration being configured to generate a confirmation value for the potential new word. In an embodiment, the at least two explorations may be chatbots, referred to herein as sub-chatbots, virtual locations different from chatbots, or combinations of different forms of virtual locations (172). For example, in an embodiment, one or more explorations may be virtual locations, such as, but not limited to, one or more social media websites, one or more crowdsourced dictionaries, distributed repositories, etc. As shown here by way of example, the dialogue system (170) is configured with multiple sub-virtual locations (174), each of which is in the form of a chatbot, shown herein as chatbot 0 (1740), chatbot 1 (1741)... chatbot N (174 NThe number of sub-locations shown here is for illustrative purposes and should not be considered limiting. In an exemplary embodiment, a sub-chatbot represents an exploration or discovery location configured to validate new words. For example, in an embodiment, the sub-chatbot enables a sub-user to submit validation values ​​for potential new words. Validation values ​​received via the sub-location are processed by a guide (156) and used as a factor regarding knowledge graph updates. For example, in an exemplary embodiment, validation values ​​are processed relative to a configurable threshold, and in an embodiment, if a validation value meets or exceeds the threshold, the new word is validated and accepted, and if the validation value does not meet the threshold, the new word is not accepted. Thus, the sub-chatbot, or one or more sub-virtual environments in an embodiment, is utilized by the token manager (154) to validate new words and selectively update the dataset and corresponding knowledge graph.

[0033] As shown, the guide (156) is operatively coupled to the token manager (154). The guide (156) is configured to modify the dataset and manage modifications to the knowledge graph based on the identification and evaluation of new words. In an exemplary embodiment, dataset modification is dynamic. Since the knowledge graph is a representation of the dataset, dynamic modification of the dataset includes updating the knowledge graph, or updating the knowledge graph after dynamic modification of the dataset, adding new words as new nodes in the knowledge graph, adding new meanings as new edges in the knowledge graph, or combinations thereof.

[0034] The neural network manager (158) can train an artificial neural network (ANN) with the dataset using a knowledge graph representation of the dataset, and retrain the ANN in response to updates to the corresponding knowledge graph. In this embodiment, ANN training is optional. As shown in the knowledge base (160), each dataset-knowledge graph pair is illustrated with an operatively coupled ANN, referred to herein as a model. For example, the dataset... 0,0 (164 0,0 Knowledge Graph 0,0 (166 0,0 ) Displayed as a model 0,0 (168 0,0 ), dataset 0,1 (164 0,1 Knowledge Graph 0,1 (166 0,1 ) Displayed as a model 0,1 (168 0,1 ), ... dataset 0，N (164 0，N Knowledge Graph 0，N (166 0，N ) Displayed as a model 0，N N(168) 0，N ), dataset 1,0 (164 1,0Knowledge Graph 1,0 (166 1,0 ) Displayed as a model 1,0 (168 1,0 ), dataset 1,1 (164 1,1 Knowledge Graph 1,1 (166 1,1 ) Displayed as a model 1,1 (168 1,1 ), ... dataset 1，N (164 1，N Knowledge Graph 1，N (166 1，N ) Displayed as a model 1,1 (168 1,N As is known in the art, an ANN is configured with multiple layers, including an input layer, one or more inner layers (also referred to herein as hidden layers), and an output layer. The purpose of training an ANN is that, when complete, the ANN can receive input data and generate output data that classifies the received input data. For example, regarding image identification and classification, an ANN can receive an image as input data, convert the image into a set of pixels, process the pixel set and corresponding pixel values ​​through the ANN, and generate output data corresponding to the image, which classifies the interpretation of the received image. ANNs are not limited to image recognition. In one embodiment, an ANN can be trained to recognize other expressive media, such as audio, and therefore the scope of ANNs should not be considered limiting. The output from the ANN, along with synchronous evaluation, indicates the selective emission of control signals, also referred to herein as encoded actions, where the control signals are directed to physical devices or components of physical devices.

[0035] In an exemplary embodiment, a control signal facilitates or causes a change in the state of an object, physically transforming the object from a first state to a second state. As illustrated herein by way of example, a physical hardware device (178) is operatively coupled to a server (110) across a network (105). In an embodiment, the device (178) may be operatively coupled to one or more of the server (110) or systems (180), (182), (184), (186), (188), and (190). In an exemplary embodiment, the control signal selectively controls the operatively coupled physical hardware device (178), or in an embodiment controls a process controlled by software or a combination of the physical hardware device (178) and software, selectively modifying physical functional aspects of the device (178). In an embodiment, the device (178) may be a first physical device operatively coupled to an internal component, or in an embodiment, it may be a second physical device, and the emitted first signal may modify the operating state of the internal component or the second device. For example, the first device (178) may be a product dispenser, and control signals may modify or control the product dispensing rate to suit the rate at which the second device receives dispensed products. In one embodiment, the guide (156) calculates control actions based on the generated context and constructs or configures control signals aligned with or commensurate with the calculated control actions. In an exemplary embodiment, control actions may be applied as feedback signals to directly control event injection, thereby maximizing the likelihood of event or operational states of the device (178). In an embodiment, the ANN configures and generates control signals in response to dynamic updates of the corresponding knowledge graph. Thus, the guide (156) interfaces with the corresponding ANN to selectively generate and transmit control signals, thereby selectively controlling the physical state of the operatively coupled device (178), software, or a combination thereof.

[0036] As described herein, the AI ​​platform (150) and associated tools (152)-(158) are operatively coupled to a data source (160), which includes one or more libraries containing one or more datasets, knowledge graphs, and ANNs. As described herein, the system and associated tools leverage AI to support dynamic knowledge graph management and, in embodiments, dynamically signal (also referred to herein as control signals) to control or modify physical hardware devices, software-controlled processes, or combinations thereof. As shown, in various embodiments, the network (105) may include local and remote network connections, enabling the AI ​​platform (150) to operate in environments of any size, including local and global, such as the Internet. Additionally, the AI ​​platform (150) serves as a front-end system that makes various kinds of knowledge extracted or represented from network-accessible sources and / or structured data sources available. In this way, processes populate the AI ​​platform (150), which also includes input interfaces to receive requests and respond accordingly.

[0037] The AI ​​platform (150) and associated tools (152)-(158) utilize data sources (160) to support the dynamic management of one or more knowledge graphs, and use the knowledge graphs and corresponding trained or retrained ANNs to coordinate one or more actions for device and / or process optimization. Devices processing data received over a network (105) can be provided by servers (110) (e.g., IBM). The server and the corresponding AI platform (150) are used for processing. As shown in this document, the AI ​​platform (150) works with embedded tools (152)-(158) to explore the virtual environment interface to identify and parse new words, dynamically update the dataset and corresponding knowledge graph based on the new word parsing, and generate one or more signals in the embodiment to physically modify the state of physical objects.

[0038] In some illustrative embodiments, the server (110) may be an IBM product available from International Business Machines Corporation in Armonk, New York. The system adds mechanisms to the illustrative embodiments described below. Tools (152)-(158), collectively referred to below as AI tools, are shown as being contained in or integrated into the AI ​​platform (150) of the server (110). AI tools may be implemented in a separate computing system (e.g., 190), or in one embodiment, they may be implemented in one or more systems connected to the server (110) via a network (105). Wherever embodied, the AI ​​tool functionality parses new words and reflects the parse through dynamic optimization of the dataset and the corresponding knowledge graph.

[0039] The types of devices and corresponding systems that can utilize the artificial intelligence platform (150) range from small handheld devices such as handheld computers / mobile phones (180) to mainframe systems such as mainframe computers (182). Examples of handheld computers (180) include personal digital assistants (PDAs), personal entertainment devices such as MP4 players, portable televisions, and CD players. Other examples of information processing systems include pen or tablet computers (184), laptop or notebook computers (186), personal computer systems (188), and servers (190). As shown, various devices and systems can be networked together using a computer network (105). Types of computer networks (105) that can be used to interconnect various devices and systems include local area networks (LANs), wireless local area networks (WLANs), the Internet, the public switched telephone network (PSTN), other wireless networks, and any other network topologies that can be used to interconnect devices and systems. Many devices and systems include non-volatile data storage, such as hard disk drives and / or non-volatile memory. Some devices and systems can use separate non-volatile data repositories (e.g., servers (190) utilize non-volatile data repositories (190) A ), and mainframe computers (182) utilize non-volatile data storage (182) A Non-volatile data repository (182) A () can be an external component of various devices and systems, or it can be an internal component of a device or system.

[0040] The devices and systems used to support the artificial intelligence platform (150) can take many forms, some of which are... Figure 1 As shown in the diagram. For example, an information processing system can take the form of a desktop computer, server, portable computer, laptop computer, notebook computer, or other form of computer or data processing system. Furthermore, the devices and systems(s) ...

[0041] Application Programming Interface (API) is understood in this field as a software intermediary between two or more applications. About Figure 1 The AI ​​platform (150) shown and described herein can utilize one or more APIs to support one or more of the tools (152)–(158) and their associated functionalities. References Figure 2A block diagram (200) is provided illustrating the tools (252)–(258) and their associated APIs. As shown, multiple tools are embedded within the artificial intelligence platform (205), including the Explorer Manager (152) associated with API0 (212), shown herein (252); the Token Manager (154) associated with API1 (222), shown herein (254); the Guide (156) associated with API2 (232), shown herein (256); and the Neural Network Manager (158) associated with API3 (242), shown herein (258). Each API can be implemented using one or more languages ​​and interface specifications.

[0042] API (212) provides functional support for exploring one or more virtual environments (e.g., virtual dialogue environments, chatbots, one or more social media websites, one or more crowdsourced dictionaries, distributed repositories, or combinations thereof). API1 (222) provides functional support for identifying and evaluating one or more new words present in the explored virtual environment. API2 (232) provides functional support for dynamically revising the dataset and corresponding knowledge graph based on the evaluation of (multiple) new words. API3 (242) provides functional support for training an ANN using the dataset and retraining the ANN using a selectively updated knowledge graph. As shown, each of APIs (212), (222), (232), and (242) is operatively coupled to an API orchestrator (260), or orchestration layer, which is understood in the art to serve as an abstraction layer to transparently connect individual API threads together. In embodiments, the functionality of individual APIs may be combined or combined. Therefore, the configuration of the APIs shown herein should not be considered limiting. Thus, as shown herein, the functionality of the tools may be implemented or supported by their respective APIs.

[0043] refer to Figure 3A flowchart (300) is provided to illustrate the process of selectively and dynamically modifying a dataset and corresponding knowledge graph using a virtual communication environment. The dataset is processed to train or support an AI model (302). In an exemplary embodiment, the dataset processing in step (302) transforms the dataset (i.e., unstructured data) into a knowledge graph (i.e., structured data) with nodes and edges. The knowledge graph gives the dataset shape and structure, enabling the data in the knowledge graph to be queried. An artificial neural network (ANN) can be trained using the knowledge graph representation of the dataset (304), and the ANN is retrained in an embodiment. In an embodiment, ANN training is optional. The virtual environment in the form of a chatbot uses machine learning and deep learning elements of AI to develop an increasingly granular knowledge base for natural language understanding to discern user needs and uses AI tools to determine the user's(s) goals(s), such as the goals the user is trying to achieve. Using the chatbot environment, input from the user is elicited (306). In an embodiment, the user input follows NLP, and the words in the input are tokenized. It should be understood in the art that tokenization is the process of separating a string sequence into fragments such as words, keywords, phrases, symbols, and other elements, which are individually referred to as tokens. Tokens can be individual words, phrases, or, in embodiments, sentences. During tokenization, some characters may be discarded. After receiving and tokenizing in step (306), the token(s) are verified, referred to herein as a first verification or first verification process, to determine whether the token is represented as a node in the knowledge graph (308). In an exemplary embodiment, the verification at step (308) is an initial evaluation of the provided token representation. If the verification at step (308) does not produce a match in the knowledge graph, a new node (312) is created for the token representation. In an embodiment, spell verification of the token(s) may be performed at step (310) before the creation of the new node to ensure that the representation in natural language is correct; in an embodiment, this may include spell correction. Therefore, the token verification at steps (308) and (310) is used to selectively modify the knowledge graph by creating one or more nodes.

[0044] As shown in the figure, the initial evaluation at step (308) identifies whether the verified token(s) are represented in the knowledge graph. After creating a new node or finding a matching node in the knowledge graph at step (312), the meaning for the topic node is identified (314). In an exemplary embodiment, the meaning of the node is used to define one or more edges in the knowledge graph. The meaning of the token may be obvious from at least one of the corresponding edges, or the meaning of the token may not be obvious. The token may be in the form of a new word, i.e., a newly created word or phrase whose meaning is inherently difficult to determine. A verification process (316) (also referred to herein as a second verification or second verification process) is shown here to illustrate the meaning of the token to be checked. The second verification process at step (316) explores request feedback from one or more virtual environments. In an exemplary embodiment, the virtual environment is a chatbot, which may be the same chatbot platform that identifies the node and subjectes the node to processing, or it may be a separate chatbot environment used to request the definition and interpretation of the token(s). In an embodiment, the second verification process at step (316) is in the form of crowdsourcing to obtain interpretations or interpretation data for the topic token(s). For example, in an embodiment, the second verification process at step (316) may require automated crawling and scraping of social media sites, crowdsourced dictionaries, and / or distributed repositories. Similarly, in an embodiment, the second verification process at step (316) requests definitions and interpretations of the tokens(s)(s)(s) from two or more users different from the user who provided the tokens(s). Therefore, the second verification process is used to verify the meaning of the tokens(s)(s)(s)(s)(s)) through interaction or a set of interactions.

[0045] In an exemplary embodiment, and as shown herein, the requested definition or interpretation from the second verification process (316) may be evaluated, or in an embodiment, evaluated by a process (318). For example, a platform for requesting verification of token(s) may receive feedback from multiple users. It should be understood that token verification may include data that matches the meaning of the token obtained at step (314), data that does not match the obtained token meaning, or a combination of matching and non-matching data. In an exemplary embodiment, the evaluation process at step (318) identifies the number or percentage of requested interpretation definitions that match the token meaning based on the number of interpretations received. For example, in one embodiment, the evaluation process (318) may include a percentage threshold for matching meanings. In an embodiment, the threshold is a configurable value. If the token meaning is evaluated and verified at step (318), the knowledge graph is modified or otherwise extended to reflect the verification (320), and the process returns to step (302) to retrain the ANN using the modified knowledge graph. Similarly, if the evaluation in step (318) does not verify the meaning of the token, the process returns to step (314) to continue or restart the verification process.

[0046] As shown in the figure, this document uses one or more AI interactive environments to support and implement the automatic modification of the knowledge graph. In an exemplary embodiment, as shown in step (306), words are continuously collected from the chatbot environment when input is elicited from the user. Verification of the collected words and their corresponding meanings or interpretations is performed through a second verification process, which in embodiments may be the same or different chatbot platforms. In response to the verification process, the knowledge graph is selectively and dynamically modified, such as expanded, followed by training or retraining an ANN to support the chatbot platform. Thus, as shown herein, automatic knowledge graph expansion is inherently dynamic and leverages AI, and in embodiments, an ANN, to support subsequent or future chatbot interactions.

[0047] The embodiments shown and described herein may take the form of a computer system used with an intelligent computing platform for identifying new words through exploration of a virtual environment, dynamically parsing the new words, and propagating the parsed new words to their corresponding datasets and knowledge graph representations. The aspects of tools (152)-(158) and their associated functionality may be implemented in a single-location computer system / server, or, in the embodiments, configured in a cloud-based system with shared computing resources. References Figure 4 A block diagram (400) is provided illustrating an example of a computer system / server (402), hereinafter referred to as a host (402) in a cloud computing environment (410), to implement regarding Figures 1 to 3 The systems, tools, and processes described. The host (402) can operate with a wide variety of other general-purpose or special-purpose computing system environments or configurations. Examples of well-known computing systems, environments, and / or configurations suitable for use with the host (402) include, but are not limited to, personal computer systems, server computer systems, thin clients, fat clients, handheld or laptop devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputer systems, mainframe computer systems, and file systems (e.g., distributed storage environments and distributed cloud computing environments) that include any of the foregoing systems, devices, and their equivalents.

[0048] The host (402) can be described in the general context of computer system executable instructions such as program modules executed by a computer system. Generally, a program module may include routines, programs, objects, components, logic, data structures, etc., that perform a specific task or implement a specific abstract data type. The host (402) can be implemented in a distributed cloud computing environment where the task is performed by a remote processing device linked via a communication network. In a distributed cloud computing environment, program modules can reside in both local and remote computer system storage media, including memory storage devices.

[0049] like Figure 4 As shown, the host (402) is illustrated as a general-purpose computing device. Components of the host (402) may include, but are not limited to, one or more processors or processing units (404), such as a hardware processor, system memory (406), and a bus (408) that couples various system components, including the system memory (406), to the processor (404). The bus (408) represents one or more bus structures of any type, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, and a processor or local bus using any of the various bus architectures. By way of example and not limitation, such architectures include the Industry Standard Architecture (ISA) bus, the Micro Channel Architecture (MCA) bus, the Enhanced ISA (EISA) bus, the Video Electronics Standards Association (VESA) local bus, and the Peripheral Component Interconnect (PCI) bus. The host (402) typically includes various computer system readable media. Such media can be any available media accessible to the host (402), and it includes both volatile and non-volatile media, removable and non-removable media.

[0050] The memory (406) may include computer system readable media in the form of volatile memory, such as random access memory (RAM) (430) and / or cache memory (432). By way of example only, the storage system (434) may be provided for reading from and writing to a non-removable, non-volatile magnetic medium (not shown and generally referred to as a "hard disk drive"). Although not shown, a disk drive for reading from and writing to a removable, non-volatile disk (e.g., a "floppy disk") and an optical disk drive for reading from and writing to a removable, non-volatile optical disk such as a CD-ROM, DVD-ROM, or other optical media may be provided. In this case, each may be connected to the bus (408) via one or more data media interfaces.

[0051] A program / utility (440) having at least one set of program modules (442), along with an operating system, one or more applications, other program modules, and program data, may be stored in memory (406) as an example, not a limitation. Each of the operating system, one or more applications, other program modules, and program data, or some combination thereof, may include an implementation of a networking environment. Program modules (442) typically perform the functions and / or methods of various embodiments to dynamically organize activities across one or more domains to minimize risk. For example, the set of program modules (442) may include, for example, programs such as... Figure 1 The tools (152) to (158) mentioned above.

[0052] The host (402) can also communicate with one or more external devices (414), such as a keyboard, pointing device, etc.; a display (424); one or more devices that enable a user to interact with the host (402); and / or any device that enables the host (402) to communicate with one or more other computing devices (e.g., a network card, modem, etc.). This communication can occur via an input / output (I / O) interface (422). Furthermore, the host (402) can communicate with one or more networks, such as a local area network (LAN), a general wide area network (WAN), and / or a public network (e.g., the Internet), via a network adapter (420). As shown, the network adapter (420) communicates with other components of the host (402) via a bus (408). In embodiments, multiple nodes of a distributed file system (not shown) communicate with the host (402) via the I / O interface (422) or via the network adapter (420). It should be understood that, although not shown, other hardware and / or software components may be used in conjunction with the host (402). Examples include, but are not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data archive storage systems.

[0053] In this document, the terms “computer program medium,” “computer-usable medium,” and “computer-readable medium” are generally used to refer to media such as main memory (406), including RAM (430), cache (432), and storage systems (434), such as removable storage drives and hard disks installed in hard disk drives.

[0054] The computer program (also known as computer control logic) is stored in memory (406). The computer program may also be received via a communication interface such as a network adapter (420). When run, such a computer program enables the computer system to perform the features of this embodiment discussed herein. Specifically, when run, the computer program enables the processing unit (404) to perform features of the computer system. Therefore, such a computer program represents the controller of the computer system.

[0055] Computer-readable storage media can be tangible devices capable of retaining and storing instructions for use by an instruction execution device. Computer-readable storage media can be, for example, but not limited to, electronic storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination of the foregoing. A non-exhaustive list of more specific examples of computer-readable storage media includes the following: portable computer disks, hard disks, dynamic or static random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), magnetic storage devices, portable optical disc read-only memory (CD-ROM), digital multifunction disc (DVD), memory sticks, floppy disks, mechanical encoding devices, such as punched cards or raised structures in slots on which instructions are recorded, and any suitable combination of the foregoing. Computer-readable storage media as used herein should not be construed as transient signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses through fiber optic cables), or electrical signals transmitted through wires.

[0056] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to a suitable computing / processing device, or downloaded via a network (e.g., the Internet, a local area network, a wide area network, and / or a wireless network) to an external computer or external storage device. The network may include copper cables, optical fibers, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to a computer-readable storage medium within the respective computing / processing device.

[0057] Computer-readable program instructions used to perform the operations of this embodiment may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, Smalltalk, C++, etc., and conventional procedural programming languages ​​such as the "C" programming language or similar programming languages. The computer-readable program instructions may execute entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer, partially on a remote computer, or entirely on a remote computer, server, or server cluster. In the latter case, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or wide area network (WAN), or connected to an external computer (e.g., via the Internet provided by an Internet service provider). In some embodiments, electronic circuitry including, for example, programmable logic circuitry, a field-programmable gate array (FPGA), or a programmable logic array (PLA) may execute the computer-readable program instructions by personalizing the electronic circuitry with state information from the computer-readable program instructions in order to perform aspects of the embodiment.

[0058] In one embodiment, the host (402) is a node in a cloud computing environment. As is known in the art, cloud computing is a service delivery model for enabling convenient, on-demand network access to a shared pool of configurable computing resources (e.g., networks, network bandwidth, servers, processing, memory, storage, applications, virtual machines, and services) that can be rapidly provisioned and released with minimal management effort or interaction with service providers. This cloud model may include at least five features, at least three service models, and at least four deployment models. Examples of such features are as follows:

[0059] On-demand self-service: Cloud consumers can automatically and unilaterally provide computing power, such as server time and network storage, as needed, without requiring manual interaction with the service provider.

[0060] Wide network access: The capability is available on the network and accessed through standard mechanisms that facilitate the use of heterogeneous thin or thick client platforms (e.g., mobile phones, laptops, and PDAs).

[0061] Resource pooling: pooling computing resources from providers to serve multiple consumers using a multi-tenant model, where different physical and virtual resources are dynamically allocated and reallocated based on demand. Location independence is significant because consumers typically do not have control or knowledge of the exact location of the resources provided, but are able to specify the location at a higher level of abstraction (e.g., country, state, or data center).

[0062] Rapid flexibility: In some cases, capacity can be automatically, quickly, and flexibly provided to rapidly shrink and rapidly expand. For consumers, the capacity available for supply often appears unlimited and can be purchased in any quantity at any time.

[0063] Metering services: Cloud systems automatically control and optimize resource usage by leveraging metering capabilities at an abstraction layer appropriate to service types (e.g., storage, processing, bandwidth, and active user accounts). Resource usage can be monitored, controlled, and reported, providing transparency to both service providers and consumers.

[0064] The service model is as follows:

[0065] Software as a Service (SaaS): The capability offered to consumers is the ability to use the provider's applications running on cloud infrastructure. Applications can be accessed from various client devices through thin client interfaces such as web browsers (e.g., web-based email). Consumers do not manage or control the underlying cloud infrastructure, including networks, servers, operating systems, storage, or even individual application capabilities, with possible exceptions such as limited user-specific application configuration settings.

[0066] Platform as a Service (PaaS): This provides consumers with the ability to deploy applications created or acquired by the consumer onto cloud infrastructure, using programming languages ​​and tools supported by the provider. Consumers do not manage or control the underlying cloud infrastructure, including networks, servers, operating systems, or storage devices, but they have control over the configuration of the deployed applications and, possibly, the application hosting environment.

[0067] Infrastructure as a Service (IaaS): This provides consumers with the capabilities of processing, storage, networking, and other basic computing resources that enable them to deploy and run arbitrary software, which may include operating systems and applications. Consumers do not manage or control the underlying cloud infrastructure, but they do have control over the operating system, storage, deployed applications, and possibly limited control over selected networking components (e.g., host firewalls).

[0068] The deployment model is as follows:

[0069] Private cloud: Cloud infrastructure intended for organization operation only. It can be managed by the organization or a third party and can reside inside or outside a building.

[0070] Community cloud: A cloud infrastructure shared by several organizations and supporting a specific community with shared concerns (e.g., tasks, security requirements, policies, and compliance considerations). It can be managed by an organization or a third party and can exist inside or outside a building.

[0071] Public cloud: Cloud infrastructure available to the general public or large industrial groups and owned by organizations that sell cloud services.

[0072] Hybrid cloud: A cloud infrastructure is a combination of two or more clouds (private, community, or public) that remain as a single entity but are bound together by standardized or proprietary technologies that enable data and application portability (e.g., cloud bursts for load balancing between clouds).

[0073] Cloud computing environments are service-oriented environments focused on statelessness, loose coupling, modularity, and semantic interoperability. At the heart of cloud computing is the infrastructure of a network of interconnected nodes.

[0074] Now for reference Figure 5 The diagram illustrates a cloud computing network (500). As shown, the cloud computing network (400) includes a cloud computing environment (550) with one or more cloud computing nodes (510), with local computing devices used by cloud consumers capable of communicating with these cloud computing nodes. Examples of these local computing devices include, but are not limited to, personal digital assistants (PDAs) or cellular phones (554A, 554B), desktop computers (554C), laptop computers (554C), and / or automotive computer systems (554N). The individual nodes within the nodes (510) can also communicate with each other. They can be physically or virtually grouped (not shown) in one or more networks, such as private clouds, community clouds, public clouds, or hybrid clouds, or combinations thereof, as described above. This allows the cloud computing environment (500) to provide infrastructure, platforms, and / or software as services that cloud consumers do not need to maintain on their local computing devices. It should be understood that... Figure 5 The types of computing devices (554A-554N) shown are intended to be illustrative only, and the cloud computing environment (550) can communicate with any type of computerized device via any type of network and / or network-addressable connection (e.g., using a web browser).

[0075] Now for reference Figure 6 This shows the result of Figure 5 The cloud computing network provides a set of functional abstraction layers (600). It should be understood in advance that... Figure 6 The components, layers, and functions shown are illustrative only, and the embodiments are not limited thereto. As shown, the following layers and corresponding functions are provided: hardware and software layer (610), virtualization layer (620), management layer (630), and workload layer (640).

[0076] The hardware and software layer (610) includes hardware and software components. Examples of hardware components include mainframes, in one example... A server based on a RISC (Reduced Instruction Set Computer) architecture, in one example, is an IBM server. System; IBM System; IBM Systems; storage devices; networking and interconnection components. Examples of software components include network application server software, in one example from IBM. Application server software; and database software, in one example, IBM. Database software. (IBM, zSeries, pSeries, xSeries, BladeCenter, WebSphere, and DB2 are trademarks of International Business Machines Corporation registered in many jurisdictions worldwide.)

[0077] The virtualization layer (620) provides an abstraction layer from which the following examples of virtual entities can be provided: virtual servers; virtual storage; virtual networks, including virtual private networks; virtual applications and operating systems; and virtual clients.

[0078] In one example, the management layer (630) can provide the following functions: resource provisioning, metering and pricing, user portal, service layer management, and SLA planning and implementation. Resource provisioning provides the dynamic acquisition of computing resources and other resources used to perform tasks within the cloud computing environment. Metering and pricing provides cost tracking of resource utilization within the cloud computing environment, as well as billing or invoicing the consumption of these resources. In one example, these resources may include application software licenses. Security provides authentication for cloud consumers and tasks, and protection for data and other resources. The user portal provides consumers and system administrators with access to the cloud computing environment. Service layer management provides the allocation and management of cloud computing resources to meet the required service layer needs. Service layer agreement (SLA) planning and implementation provides the pre-arrangement and procurement of cloud computing resources, anticipating future demand for cloud computing resources according to the SLA.

[0079] The workload layer (640) provides examples of functionalities that can be leveraged in a cloud computing environment. Examples of workloads and functionalities that can be provided from this layer include, but are not limited to: map creation and navigation; software development and lifecycle management; virtual classroom education delivery; data analytics and processing; transaction processing; and dynamic knowledge graph extensions.

[0080] It should be understood that systems, methods, apparatuses, and computer program products are disclosed herein for evaluating natural language input, detecting queries in corresponding communications, and parsing detected queries with answers and / or supporting content.

[0081] While specific embodiments of this embodiment have been shown and described, it will be apparent to those skilled in the art that changes and modifications can be made based on the teachings herein without departing from the embodiments and their broader aspects. Therefore, the appended claims are intended to cover within their scope all such changes and modifications that fall within the true spirit and scope of the embodiments. Furthermore, it should be understood that the embodiments are defined solely by the appended claims. Those skilled in the art will understand that if a particular number of introduced claim elements is desired, such an intention will be expressly stated in the claims, and in the absence of such a statement, there is no such limitation. For non-limiting examples, to aid understanding, the appended claims contain the use of the introductory phrases “at least one” and “one or more” to introduce claim elements. However, the use of these phrases should not be construed as implying that introducing claim elements with the indefinite article “a” or “an” limits any particular claim containing such introduced claim elements to embodiments containing only one such element, even when the same claim includes the introductory phrase “one or more” or “at least one” and indefinite articles such as “a” or “an”; this also applies to uses in claims of declarative articles.

[0082] This embodiment may be a system, method, and / or computer program product. Furthermore, selected aspects of the embodiments of the present invention may take the form of a completely hardware embodiment, a completely software embodiment (including firmware, resident software, microcode, etc.), or an embodiment combining software and / or hardware aspects, all of which may be collectively referred to herein as "circuit," "module," or "system." Additionally, aspects of this embodiment may take the form of a computer program product contained in a computer-readable storage medium (or media), having computer-readable program instructions on it for causing a processor to execute aspects of this embodiment. Thus implemented, the disclosed system, method, and / or computer program product is operable to improve the functionality and operation of an artificial intelligence platform to be supported by KG-driven content generation for AR.

[0083] Various aspects of this embodiment are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-readable program instructions.

[0084] These computer-readable program instructions may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions / actions specified in one or more boxes of a flowchart and / or block diagram. These computer-readable program instructions may also be stored in a computer-readable storage medium that can instruct a computer, programmable data processing apparatus, and / or other device to function in a particular manner, such that the computer-readable storage medium in which the instructions are stored includes an article of writing comprising instructions for implementing aspects of the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0085] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus or other device to cause a series of operational steps to be performed on the computer, other programmable apparatus or other device, thereby producing a computer-implemented process, such that the instructions that execute on the computer, other programmable apparatus or other device implement the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0086] The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this embodiment. In this regard, each box in a flowchart or block diagram may represent a module, segment, or portion of instructions, including one or more executable instructions for implementing a specified logical function(s). In some alternative implementations, the functions indicated in the boxes may not occur in the order shown in the figures. For example, depending on the functions involved, two boxes shown consecutively may actually be executed substantially simultaneously, or these boxes may sometimes be executed in reverse order. It will also be noted that each box in the block diagrams and / or flowcharts, and combinations of boxes in the block diagrams and / or flowcharts, may be implemented by a dedicated hardware-based system that performs the specified function or action or executes a combination of dedicated hardware and computer instructions.

[0087] It should be understood that although specific embodiments have been described herein for illustrative purposes, various modifications can be made without departing from the spirit and scope of the embodiments. In particular, the dynamic management of the KG and corresponding datasets, and the dynamic generation of signals or instructions to physical hardware devices, software, or software-controlled processes in the exemplary embodiments, can be executed by different computing platforms or across multiple devices. Furthermore, the data source can be local, remote, or distributed across multiple systems. Therefore, the scope of protection of the embodiments is defined only by the appended claims and their equivalents.

Claims

1. A computer-implemented method, comprising: A token identifying a new word in a virtual environment exploration, wherein the token is received from a user in an AI interactive environment corresponding to a first virtual location, including: The new word is identified by responding to at least one of the following: The dataset does not contain a representation of the identified token, and a new meaning for the identified token is obtained from the dataset; or The representation of the identified token is found to exist in the dataset but does not have a corresponding meaning or edge, and a new meaning associated with the identified token is extracted from the dataset; and To verify the identified token, the method involves evaluating it against a representation of the dataset, including: The first confirmation value is generated using a first virtual exploration in a second virtual location that is different from the first virtual location; A second confirmation value is generated using a second virtual exploration in a third virtual location different from the first and second virtual locations; and In response to determining that the first verification value and the second verification value meet one or more criteria, the new meaning for the identified token is verified; Dynamically modifying the dataset in response to the evaluation of the identified token includes: Updating the knowledge graph representation of the dataset using the dynamically modified dataset includes one or more of the following: adding the identified token as a new node in the knowledge graph, adding the new meaning as a new edge in the knowledge graph, The artificial neural network (ANN) is retrained based on the updated knowledge graph; and the retrained ANN is used to generate subsequent outputs of the AI ​​interactive environment.

2. The computer-implemented method of claim 1, wherein retraining the ANN is performed in response to an update of the knowledge graph.

3. The computer-implemented method of claim 1, wherein the discovery of the non-existence of the representation of the token identified in the dataset further comprises: Verifying the spelling of the identified token includes selectively correcting the spelling of the identified token and evaluating the corrected token against the representation in the dataset.

4. The computer-implemented method of claim 1, wherein the dynamic modification of the knowledge graph in response to the evaluation comprises: The knowledge graph is updated using the token identified by the first and second verification values.

5. The computer-implemented method of claim 1, further comprising dynamically issuing control signals to an operatively coupled device, a software-controlled process, or a combination of the device and the software-controlled process, the control signals being associated with the dynamically updated knowledge graph and configured to selectively control the physical state of the operatively coupled device, the software, or the combination thereof.

6. The computer-implemented method of claim 1, wherein the virtual environment includes a virtual location, the virtual location comprising one or more of the following: a virtual dialogue environment, one or more social media websites, one or more crowdsourced dictionaries, and a distributed repository.

7. A computer program product for supporting the selectivity and dynamic expansion of a dataset, the computer program product comprising program code executed by a processor to perform the operation of the method according to any one of claims 1 to 6.

8. A computer system, comprising: The processing unit is operatively coupled to the memory; An artificial intelligence (AI) platform, communicating with the processing unit, has one or more tools to support the selectivity and dynamic expansion of datasets, said tools including: The token manager is configured to perform the operation of the method according to any one of claims 1 to 6.

Images