MÉTODO E SISTEMA PARA ESTENDER UMA MARCAÇÃO E UM MASCARAMENTO DE ENTIDADES
The method extends entity tagging and masking in call center data by recognizing new entity types and replacing detected PII, improving data security and compliance by accurately identifying and masking sensitive information.
Patent Information
- Application Number
- BR112025024327
- Authority / Receiving Office
- BR · BR
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2023-05-12
- Filing Date
- 2024-05-10
- Publication Date
- 2026-11-10
AI Technical Summary
Existing systems struggle to effectively tag and mask personally identifiable information (PII) in call center data, particularly when new entity types are introduced, leading to potential data breaches and compliance issues.
A method for extending entity tagging and masking in conversation transcripts by recognizing new entity types, using parameters to define search triggers and replacing detected entities with specific entity names, and producing a revised transcript.
Enhances the accuracy and compliance of PII identification and masking, ensuring secure handling and utilization of call center data.
Abstract
Description
43 METHOD AND SYSTEM FOR EXTENDING ENTITY TAGOUT AND MASKING CROSS-REFERENCE TO RELATED REQUESTS
[001] This application claims priority and the benefit of the Application for US Provisional Patent No. 63 / 466,063, entitled SYSTEMS AND METHODS RELATED TO ENTITY TAGGING OF IDENTIFIED PERSONAL INFORMATION IN CONTACT CENTER DATA, filed with the United States Patent and Trademark Office on May 12, 2023, which was converted into pending US Patent Application No. 18 / 661,195, filed on May 10, 2024, also entitled SYSTEMS AND METHODS RELATED TO ENTITY TAGGING OF IDENTIFIED PERSONAL INFORMATION IN CONTACT CENTER DATA. BACKGROUND
[002] The present invention relates generally to telecommunications systems in the field of customer relationship management, including customer assistance through internet-based service options. More particularly, but not as a limitation, the present invention relates to systems and methods for marking personally identifiable information in call center data. BRIEF DESCRIPTION OF THE INVENTION
[003] The present invention describes a method for extending a tagging and masking of recognized entities in a conversation transcript to encompass new entity types. The method includes the step of receiving a transcript of a conversation between participants that has been processed so that the actual entities recognized therein have been replaced by their respective entity names. Each entity name can be associated with an entity type, wherein the entity types include specific entity types and generic entity types. The method further includes the step of receiving parameters that define a new Petition 870250102155, dated 07 / 11 / 2025, page 12 / 77 / 43 entity type. The parameters may include: a search trigger parameter that specifies whether a search trigger to fire off a search process comprises a keyword search trigger or an entity search trigger; a keyword trigger parameter that, when the search trigger comprises the keyword search trigger, specifies one or more keywords that, when detected in the conversation transcript, fire off the search process; an entity trigger parameter that, when the search trigger comprises the entity search trigger, specifies a type selected from among the specific entity types, which is hereinafter referred to as a selected specific entity type, which, when detected in the conversation transcript, fires off the search process;a search entity parameter that specifies one of the selected generic entity types, hereinafter referred to as a selected generic entity type, that is searched in the conversation transcript in response to the triggering of the search process; one or more search parameters that specify a search range within the conversation transcript to conduct the search for the selected generic entity type in the performance of the search process;and an entity name parameter that specifies an entity name for the new entity type that is used to replace the selected generic entity type when detected during the search process. The method may additionally include the step of performing an initial process regarding the received conversation transcript according to the received parameters that define the new entity type, which includes: searching for one of the keyword search triggers or the entity search trigger and initiating a search process in response to its detection; performing the triggered search process by searching the search range specified in the received parameters for the selected generic entity type and detecting the selected generic entity type within it; reviewing the; Petition 870250102155, dated 07 / 11 / 2025, page 13 / 77 / 43 transcript of conversation so that the entity name of the selected generic entity type detected is replaced by the entity name for the new entity type; and produce a revised version of the conversation transcript that includes the revision. BRIEF DESCRIPTION OF THE DRAWINGS
[004] A fuller appreciation of the present invention will become more readily apparent as the invention becomes better understood by reference to the following detailed description, when considered in conjunction with the accompanying drawings, in which similar reference symbols indicate similar components, wherein: Figure 1 shows a schematic block diagram of a computing device, according to exemplary embodiments of the present invention and / or with which exemplary embodiments of the present invention can be enabled or practiced; Figure 2 shows a schematic block diagram of a communications infrastructure or call center, according to exemplary embodiments of the present invention and / or with which the exemplary embodiments of the present invention can be enabled or practiced; Figure 3 shows an illustrative process flow diagram that illustrates a method of the present invention according to a keyword search trigger; Figure 4 shows an illustrative process flow diagram that illustrates a method of the present invention according to an entity search trigger; and Figure 5 shows an illustrative method of the present invention. DETAILED DESCRIPTION
[005] With the purpose of promoting an understanding of the principles Petition 870250102155, dated 07 / 11 / 2025, page 14 / 77 / 43 of the invention, reference will now be made to the exemplary embodiments illustrated in the drawings, and specific language will be used to describe them. It will be evident, however, to one skilled in the art that the detailed material provided in the examples may not be necessary to practice the present invention. In other cases, well-known materials or methods have not been described in detail to avoid obscuring the present invention. Additionally, further modifications to the examples or application of the principles of the invention, as presented herein, are contemplated, as would normally occur to those skilled in the art.
[006] As used herein, the language designating examples and illustrations is not limiting and includes, for example, i.e., by way of illustration, as an example, and the like. Additionally, the reference throughout this descriptive report to an embodiment, one or more embodiments, present embodiments, exemplary embodiments, certain embodiments, and the like means that a specific feature, structure, or characteristic described in connection with the given example may be included in at least one embodiment of the present invention. Thus, the use of expressions such as an embodiment, one or more embodiments, the present embodiments, exemplary embodiments, certain embodiments, and the like do not necessarily refer to the same embodiment or example. Furthermore, specific features, structures, or characteristics may be combined in any appropriate combinations and / or subcombinations in one or more embodiments or examples.
[007] Those skilled in the art will recognize from the present description that the various embodiments can be implemented by computer using many different types of data processing equipment, and that the embodiments are implemented in the form of a computer program device, method or product. Petition 870250102155, dated 07 / 11 / 2025, page 15 / 77 / 43 Exemplary modalities, therefore, can take the form of a fully hardware-based modality, a fully software-based modality, or a modality combining aspects of software and hardware. Exemplary modalities can additionally take the form of a computer program product embodied by computer-usable program code in any tangible medium of expression. In each case, the exemplifying modality can be generically referred to as a module, system, or method. Computing device
[008] It will be recognized that the systems and methods of the present invention can be implemented by computer using many different forms of data processing equipment, for example digital microprocessors and associated memory running suitable software programs. By way of background, Figure 1 illustrates a schematic block diagram of an exemplary computing device 100, according to embodiments of the present invention and / or with which those embodiments can be enabled or practiced. It should be understood that Figure 1 is provided as a non-limiting example.
[009] The computing device 100, for example, can be implemented by means of firmware (e.g., an application-specific integrated circuit), hardware, or a combination of software, firmware, and hardware. It will be recognized that each of the servers, controllers, switches, gateways, mechanisms, and / or modules in the figures (which collectively may be called servers or modules) can be implemented by means of one or more of the computing devices 100. As an example, the various servers may be a process running on one or more processors of one or more computing devices 100, which may be executing computer program instructions and interacting with other systems or modules in order to execute Petition 870250102155, dated 07 / 11 / 2025, p. 16 / 77 / 43 the various functionalities described herein. Unless specifically limited, the functionality described in relation to a plurality of computing devices may be integrated into a single computing device, or the various functionalities described in relation to a single computing device may be distributed across multiple computing devices.Additionally, with respect to the computing systems described in the following figures — such as, for example, the call center system 200 in Figure 2 — the various servers and computing devices of these may be located on local computing devices 100 (that is, on-site or in the same physical location as the call center agents), remote computing devices 100 (that is, off-site or in a cloud computing environment, for example in a remote data center connected to the call center via a network), or some combination thereof.The functionality provided by servers located on external computing devices can be accessed and delivered over a virtual private network (VPN), as if those servers were internal, or the functionality can be delivered using Software as a Service (SaaS) accessed over the Internet using various protocols, such as exchanging data via Extensible Markup Language (XML), JavaScript Object Notation (JSON), and similar protocols.
[0010] As shown in the illustrated example, the computing device 100 may include a central processing unit (CPU) or processor 105 and main memory 110. The computing device 100 may also include a storage device 115, a removable media interface 120, a network interface 125, an I / O controller 130, and one or more input / output (I / O) devices 135 which, as represented, may include a Petition 870250102155, dated 07 / 11 / 2025, page 17 / 77 / 43 display device 135A, a keyboard 135B and a pointing device 135C. The computing device 100 may additionally include additional elements such as a memory port 140, a bridge 145, I / O ports, one or more additional input / output devices 135D, 135E, 135F and a cache memory 150 in communication with the processor 105.
[0011] The processor 105 can be any logic circuit that responds to and processes instructions fetched from main memory 110. For example, the processor 105 can be implemented by an integrated circuit, such as a microprocessor, a microcontroller, or a graphics processing unit, or in a field-programmable gate array or an application-specific integrated circuit. As shown, the processor 105 can communicate directly with the cache memory 150 via a secondary bus or backbus. The cache memory 150 typically has a faster response time than main memory 110. Main memory 110 can be one or more memory chips capable of storing data and allowing stored data to be directly accessed by the central processing unit 105.Storage device 115 may provide storage to an operating system, which controls scheduling tasks and access to system resources and other software. Except where otherwise limited, computing device 100 may include an operating system and software capable of performing the functionality described herein.
[0012] As represented in the illustrated example, the computing device 100 may include a wide variety of I / O devices 135, one or more of which may be connected via the I / O controller 130. Input devices, for example, may include a keyboard 135B and a pointing device 135C, for example a mouse or light pen. Output devices may include, for example, devices of Petition 870250102155, dated 07 / 11 / 2025, page 18 / 77 / 43 video display, speakers and printers. The I / O devices 135 and / or the I / O controller 130 may include hardware and / or software suitable for enabling the use of multiple display devices. The computing device 100 may also support one or more removable media interfaces 120, such as a disk drive, USB port or any other device suitable for reading data from or writing data to computer-readable media. More generally, the I / O devices 135 may include any conventional devices to perform the functionality described herein.
[0013] Computing device 100 may be any workstation, desktop computer, laptop or notebook computer, server machine, virtualized machine, mobile phone or smartphone, portable telecommunications device, media playback device, gaming system, mobile computing device, or any other type of computing, telecommunications or media device, without limitation, capable of performing the operations and functionality described herein. Computing device 100 includes a plurality of devices connected by a network or connected to other systems and resources via a network.As used in this document, a network includes one or more computing devices, machines, clients, client nodes, client machines, client computers, client devices, endpoints, or endpoint nodes communicating with one or more other computing devices, machines, clients, client nodes, client machines, client computers, client devices, endpoints, or endpoint nodes. It should be understood that, unless otherwise limited, computing device 100 may communicate with other computing devices 100 over any type of network using any conventional communication protocol. Furthermore, the network may be a virtual network environment where multiple components... Petition 870250102155, dated 07 / 11 / 2025, page 19 / 77 / 43 of the network are virtualized. Customer service center
[0014] With reference now to Figure 2, a communication infrastructure or call center system 200 is shown according to exemplary embodiments of the present invention and / or with which exemplary embodiments of the present invention can be enabled or practiced. It should be understood that the term call center system is used herein to refer to the system represented in Figure 2 and / or its components, while the term call center is used more generically to refer to call center systems, customer service providers operating such systems, and / or organizations or companies associated with them.Thus, unless specifically limited otherwise, the term call center generally refers to a call center system (such as the 200 call center system), the associated customer service provider (such as a specific customer service provider that provides customer services through the 200 call center system), as well as the organization or company on whose behalf those customer services are provided.
[0015] In this context, customer service providers generally offer many types of services through call centers. Such call centers may have a team of employees or customer service agents (or simply agents), with the agents serving as an interface between a company, a business, a government agency, or an organization (hereinafter referred to interchangeably as an organization or business) and people, such as users, individuals, or consumers (hereinafter referred to as individuals or consumers). For example, agents in a call center may assist the Petition 870250102155, dated 07 / 11 / 2025, page 20 / 77 / 43 consumers to make purchasing decisions, receive orders or resolve problems with products or services already received. Within a call center, such interactions between call center agents and external entities or customers can be conducted through a variety of communication channels, such as, for example, by voice (e.g., telephone calls or voice calls over IP or VoIP), video (e.g., video conferencing), text (e.g., emails and text chat), screen sharing, browsing or similar.
[0016] Operationally, call centers generally strive to provide quality service to consumers while minimizing costs. For example, one way a call center operates is to handle every customer interaction with a live agent. While this approach might fit well in terms of service quality, it would likely be prohibitively expensive due to the high cost of agent labor. Because of this, most call centers utilize some level of automated process in place of live agents, such as Interactive Voice Response (IVR) systems, Interactive Media Response (IMR) systems, internet robots or robots, automated chat modules or chatbots, and the like.
[0017] With reference specifically to Figure 2, the 200 call center system can be used by a customer service provider to provide various types of services to customers. For example, the 200 call center system can be used to engage and manage interactions in which automated processes (or bots) or human agents communicate with customers. As should be understood, the 200 call center system can be an internal installation within a business or company, intended to perform sales functions and customer service related to products and services available through the Petition 870250102155, dated 07 / 11 / 2025, page 21 / 77 / 43 company. In another aspect, the 200 call center system can be operated by a third-party service provider that is contracted to provide services to another organization. Furthermore, the 200 call center system can be deployed on equipment dedicated to the company or the third-party service provider and / or deployed in a remote computing environment, such as a private or public cloud environment with infrastructure to support multiple call centers from multiple companies. The 200 call center system may include software applications or programs, which can be run locally or remotely, or in some combination thereof.It will also be recognized that the various components of the 200 call center system can be distributed across multiple geographic locations and are not necessarily contained within a single location or computing environment.
[0018] It should be further understood that, unless specifically limited otherwise, any of the computing elements of the present invention can be implemented in cloud-based or cloud-based computing environments. As used herein, cloud computing – or simply the cloud – is defined as a model for enabling ubiquitous, convenient, and on-demand network access to a shared pool of configurable computing resources (e.g., networks, servers, storage, applications, and services) that can be rapidly provisioned through virtualization and released with minimal management effort or service provider interaction, and then scaled accordingly. Cloud computing can be comprised of various features (e.g., on-demand self-service, broad network access, resource pooling, rapid elasticity, metered service, etc.).), service models (e.g., Software as a Service (SaaS), Platform as a Service (PaaS), Infrastructure as a Service. Petition 870250102155, dated 07 / 11 / 2025, page 22 / 77 / 43 service (IaaS) and deployment models (e.g., private cloud, community cloud, public cloud, hybrid cloud, etc.). Generally called serverless architecture, a cloud execution model typically includes a service provider that dynamically manages the allocation and provisioning of remote servers to achieve a desired functionality.
[0019] According to the illustrated example in Figure 2, the components or modules of the call center system 200 may include: a plurality of client devices 205A, 205B, 205C; a communication network (or simply network) 210; a media switch / gateway 212; a call controller 214; an interactive media response (IMR) server 216; a routing server 218; a storage device 220; a statistics server 226; a plurality of agent devices 230A, 230B, 230C which include the workbins 232A, 232B, 232C, respectively; a multimedia / social media server 234; a knowledge management server 236 coupled to a knowledge system 238; a chat server 240; web servers 242; an interaction server (iXn) 244; a universal contact server (UCS universal contact server) 246; a reporting server 248; a media services server 249; and an analytics module 250.It should be understood that any of the computer-implemented components, modules, or servers described in relation to Figure 2 or in any of the following figures can be implemented through computing device types, such as, for example, computing device 100 in Figure 1. As will be observed, the call center system 200 generally manages resources (e.g., human resources, computers, telecommunications equipment, etc.) to enable the provision of services via telephone, email, chat, or other communication mechanisms. Such services may vary depending on the type of call center. Petition 870250102155, dated 07 / 11 / 2025, page 23 / 77 / 43 service and may include, for example, customer service, help desk functionality, emergency response, telemarketing, order registration and the like.
[0020] Customers wishing to receive services from the 200 call center system can initiate inbound communications (e.g., phone calls, emails, chats, etc.) to the 200 call center system via a 205 customer device. Although Figure 2 shows three such customer devices—namely, 205A, 205B, and 205C—it should be understood that any number of them may be present. The 205 customer devices, for example, can be a communication device such as a phone, smartphone, computer, tablet, or laptop. According to the functionality described herein, customers can generally use the 205 customer devices to initiate, manage, and conduct communications with the 200 call center system, such as phone calls, emails, chats, text messages, web browsing sessions, and other multimedia transactions.
[0021] Inbound and outbound communications to and from customer devices 205 may traverse the network 210, the nature of which will typically depend on the type of customer device being used and the form of communication. As an example, the network 210 may include a telephone, cellular, and / or data service communication network. The network 210 may be a public switched telephone network (PSTN) or private network, a local area network (LAN), a private wide area network (WAN), and / or a public WAN, such as the Internet. In addition, the network 210 may include a wireless carrier network, including a code division multiple access (CDMA) network, a global system for mobile communications (GSM), or any conventional wireless network / technology in Petition 870250102155, dated 07 / 11 / 2025, p. 24 / 77 / 43 technical, including, but not limited to, 3G, 4G, LTE, 5G, etc.
[0022] Regarding media switch / gateway 212, it can be coupled to network 210 to receive and transmit telephone calls between clients and the call center system 200. Media switch / gateway 212 may include a telephone or communication switch configured to function as a central switch for agent-level routing within the call center. The switch may be a hardware switching system or implemented via software. For example, switch 215 may include an automatic call distributor, a private branch exchange (PBX), an IP-based software switch, and / or any other switch with specialized hardware and software, configured to receive interactions originating from the Internet and / or interactions originating from the telephone network from a client, and to route these interactions to, for example, one of the agent devices 230.Thus, in general, the media switch / gateway 212 establishes a voice connection between the client and the agent by means of a connection between the client device 205 and the agent device 230.
[0023] As further shown, the media switch / gateway 212 can be coupled to the call controller 214 which, for example, serves as an adapter or interface between the switch and the other routing, monitoring, and call center communication components of the call center system 200. The call controller 214 can be configured to process PSTN calls, VoIP calls, etc. For example, the call controller 214 can include computer telephony integration (CTI) to interface with the media switch / gateway and other components. The call controller 214 can include a session initiation protocol (SIP) server to process SIP calls. The call controller 214 can also Petition 870250102155, dated 07 / 11 / 2025, page 25 / 77 / 43 extract data about an incoming interaction, such as the customer's phone number, IP address, or email address, and then communicate this data to other call center components in processing the interaction.
[0024] Regarding the Interactive Media Response (IMR) 216 server, it can be configured to enable self-help or virtual assistant functionality. Specifically, the IMR 216 server can be similar to an Interactive Voice Response (IVR) server, except that the IMR 216 server is not restricted to voice and can also encompass multiple media channels. In an illustrative voice example, the IMR 216 server can be configured with an IMR script to query customers about their needs. For example, a call center for a bank might tell customers via the IMR script to press 1 if they wish to obtain their account balance. Through seamless interaction with the IMR 216 server, customers can receive the service without needing to speak to an agent.The IMR 216 server can also be configured to check the reason why a customer is contacting the call center, so that the communication can be routed to the appropriate resource.
[0025] Regarding routing server 218, it can function to route incoming interactions. For example, once it is determined that an incoming communication should be handled by a human agent, the functionality within routing server 218 can select the most appropriate agent and route the communication to it. This agent selection can be based on which available agent is most suitable to handle the communication. More specifically, the selection of the appropriate agent can be based on a routing strategy or algorithm implemented by routing server 218. With this, routing server 218 can query data that is relevant to the interaction of Petition 870250102155, dated 07 / 11 / 2025, page 26 / 77 / 43 input, for example, data related to the specific customer, available agents, and the type of interaction, which, as described below, can be stored in specific databases. Once the agent is selected, the routing server 218 can interact with the call controller 214 to route (i.e., connect) the incoming interaction to the corresponding agent device 230. As part of this connection, information about the customer can be provided to the selected agent through their agent device 230. This information is intended to improve the service that the agent can provide to the customer.
[0026] Regarding data storage, the call center system 200 may include one or more mass storage devices – generally represented by storage device 220 – to store one or more databases relevant to the operation of the call center. For example, storage device 220 may store customer data that is maintained in a customer database 222. Such customer data may include customer profiles, contact information, service level agreement (SLA), and interaction history (e.g., details of previous interactions with a specific customer, including the nature of previous interactions, disposition data, wait time, processing time, and actions taken by the call center to resolve customer issues). As another example, storage device 220 may store agent data in an agent database 223.Agent data maintained by the call center system 200 may include agent availability and agent profiles, schedules, skills, processing time, etc. As another example, the storage device 220 may store interaction data in an interaction database 224. Interaction data may include data relating to numerous previous interactions between customers and call centers. More generally, Petition 870250102155, dated 07 / 11 / 2025, page 27 / 77 / 43, it should be understood that, unless otherwise specified, the storage device 220 may be configured to include databases and / or store data related to any of the types of information described in the present invention, and that these databases and / or data are accessible to other modules or servers of the call center system 200 in ways that facilitate the functionality described herein. For example, the servers or modules of the call center system 200 may query such databases to retrieve data stored therein or transmit data to them for storage.
[0027] Regarding statistics server 226, it can be configured to record and aggregate data related to the performance and operational aspects of the call center system 200. This information can be compiled by statistics server 226 and made available to other servers and modules, such as reporting server 248, which can then use the data to produce reports that are used to manage operational aspects of the call center and execute automated actions according to the functionality described herein. This data may refer to the status of call center resources, for example, average wait time, abandonment rate, agent occupancy, and others as the functionality described herein would require.
[0028] The 230 agent devices of the 200 call center may be communication devices configured to interact with the various components and modules of the 200 call center system in ways that facilitate the functionality described herein. An 230 agent device, for example, may include a telephone adapted for regular telephone calls or VoIP calls. An 230 agent device may additionally include a computing device configured to communicate with the 200 call center system servers, perform data processing associated with operations, and interface. Petition 870250102155, dated 07 / 11 / 2025, page 28 / 77 / 43 with clients through voice, chat, email and other multimedia communication mechanisms, according to the functionality described herein. Although Figure 2 shows three such agent devices - that is, agent devices 230A, 230B and 230C - it should be understood that any number may be present.
[0029] Regarding the multimedia / social media server 234, it can be configured to facilitate media interactions (beyond voice) with client devices 205 and / or servers 242. Such media interactions may relate, for example, to email, voicemail, chat, video, text messaging, web, social media, co-browsing, etc. The multimedia / social media server 234 can take the form of any conventional IP router in the art with specialized hardware and software to receive, process, and forward multimedia events and communications.
[0030] Regarding the knowledge management server 234, it can be configured to facilitate interactions between clients and the knowledge system 238. In general, the knowledge system 238 can be a computer system capable of receiving questions or queries and providing answers in return. The knowledge system 238 can be included as part of the call center system 200 or be operated remotely by third parties. The knowledge system 238 can include an artificially intelligent computer system capable of answering questions posed in natural language by retrieving information from information sources such as encyclopedias, dictionaries, newspaper articles, literary works, or other documents submitted to the knowledge system 238 as reference materials, as is known in the art. As an example, the knowledge system 238 can be incorporated as an IBM Watson system or similar system.
[0031] Regarding the 240 chat server, it can be configured to conduct, orchestrate, and manage electronic chat communications with Petition 870250102155, dated 07 / 11 / 2025, page 29 / 77 / 43, regarding clients. In general, the 240 chat server is configured to implement and maintain chat conversations and generate chat transcripts. Such chat communications can be conducted by the 240 chat server, so that a client communicates with automated chatbots, human agents, or both. In exemplary modalities, the 240 chat server can operate as a chat orchestration server, which routes chat conversations between chatbots and available human agents. In such cases, the processing logic of the 240 chat server can be triggered by rules to employ intelligent workload distribution among the available chat resources. The chat server 240 can additionally implement, manage, and facilitate user interfaces (UIs) associated with the chat feature, including UIs generated on the client device 205 or the agent device 230.The 240 chat server can be configured to transfer chats within a single chat session with a specific client between automated and human sources, so that, for example, a chat session can be transferred from a chatbot to a human agent or from a human agent to a chatbot. The 240 chat server can also be coupled with the 234 knowledge management server and 238 knowledge systems to receive suggestions and answers to queries made by clients during a chat, so that, for example, links to relevant articles can be provided.
[0032] Regarding web servers 242, they can be included to provide website hosts for a variety of social interaction sites that customers subscribe to, such as Facebook, Twitter, Instagram, etc. Although represented as part of the 200 call center system, it should be understood that web servers 242 can be provided by third parties and / or maintained remotely. Network servers 242 can also provide web pages to the company or organization being supported by the 200 call center system. For example, the Petition 870250102155, dated 07 / 11 / 2025, page 30 / 77 / 43 Consumers can browse web pages and receive information about a specific company's products and services. On such corporate pages, mechanisms can be provided to initiate interaction with the call center system 200, for example, through online chat, voice, or email. An example of such a mechanism is a widget, which can be deployed on web pages or websites hosted on network servers 242. As used in the present invention, a widget refers to a user interface component that performs a specific function. In some implementations, a widget may include a graphical user interface control that can be overlaid on a web page displayed to a consumer on the Internet.The widget can display information, as in a window or text box, or include buttons or other controls that allow the user to access certain functionalities such as sharing or opening a file, or initiating communication. In some implementations, a widget includes a user interface component that has a portable portion of code that can be installed and run on a separate web page without compilation. Some widgets may include corresponding or additional user interfaces and be configured to access a variety of local resources (e.g., calendar or contact information on the consumer device) or remote resources via the network (e.g., instant messaging, email, or social network updates).
[0033] With respect to the interaction server (iXn) 244, it can be configured to manage deferrable call center activities and route these to human agents for completion. As used here, deferrable activities include back-office work that can be performed offline, for example, responding to emails, participating in training, and other activities that do not involve real-time communication with a customer. Petition 870250102155, dated 07 / 11 / 2025, page 31 / 77 / 43
[0034] Regarding the Universal Contact Server (UCS) 246, it can be configured to retrieve information stored in the customer database 222 and / or transmit information to said database for storage therein. For example, UCS 246 can be used as part of the chat feature to facilitate the maintenance of a history of how chats with a specific customer were handled, which can then be used as a reference for how future chats should be handled. More generally, UCS 246 can be configured to facilitate the maintenance of a history of customer preferences, such as preferred media channels and best times to contact. To this end, UCS 246 can be configured to identify data relevant to each customer's interaction history, such as data related to agent comments, customer communication history, and the like.Each of these data types can be stored in the 222 customer database or in other modules, and retrieved as required by the functionality described here.
[0035] Regarding reporting server 248, it can be configured to generate reports of data compiled and aggregated by statistics server 226 or other sources. Such reports may include near real-time reports or historical reports and refer to the state of the call center resources and performance characteristics such as average wait time, abandonment rate, and agent occupancy. The reports can be generated automatically or in response to specific requests from a requester (e.g., agent, administrator, call center application, etc.). The reports can then be used for managing call center operations according to the functionality described herein.
[0036] Regarding the 249 media services server, it can be configured to provide audio and / or video services to support features Petition 870250102155, dated 07 / 11 / 2025, page 32 / 77 / 43 from the call center. According to the functionality described herein, such resources may include requests to an IVR or IMR system (e.g., playback of audio files), contain music, voice messages / recordings with a single participant, recordings with multiple participants (e.g., audio and / or video calls), speech recognition, dual-tone multi-frequency (DTMF) recognition, fax, audio and video transcoding, secure real-time transport protocol (SRTP), audio conferencing, video conferencing, training (e.g., support for a trainer to listen to an interaction between a customer and an agent, and for the trainer to provide feedback to the agent without the customer hearing the feedback), call analysis, keyword detection, and similar features.
[0037] With respect to analysis module 250, it can be configured to provide systems and methods for performing analyses on data received from a plurality of different data sources as required by the functionality described herein. According to exemplary embodiments, analysis module 250 can also generate, update, train, and modify predictors or models 252 based on collected data, such as, for example, customer data, agent data, and interaction data. The models 252 may include customer or agent behavior models. Behavior models can be used to predict behaviors of, for example, customers or agents, in a variety of situations, thus allowing embodiments of the present invention to adapt interactions based on these predictions or allocate resources in preparation for predicted characteristics of future interactions, thereby improving the overall performance of the call center and the customer experience.It will be acknowledged that, although the 250 analysis module is represented as part of a call center, such models of... Petition 870250102155, dated 07 / 11 / 2025, page 33 / 77 / 43 behaviors can also be implemented in client systems (or, as also used in this document, on the client side of the interaction) and used for the benefit of clients.
[0038] According to exemplary embodiments, the analysis module 250 may have access to the data stored in the storage device 220, including the customer database 222 and the agent database 223. The analysis module 250 may also have access to the interaction database 224, which stores data related to interactions and interaction content (e.g., interaction transcripts and events detected therein), interaction metadata (e.g., customer identifier, agent identifier, interaction medium, interaction duration, interaction start and end time, department, tagged categories) and application configuration (e.g., the interaction path through the call center).Furthermore, as discussed further below, the analysis module 250 can be configured to retrieve data stored within the storage device 220 for use in developing and training algorithms and models 252, for example, by applying machine learning techniques.
[0039] One or more of the included 252 models can be configured to predict customer or agent behavior and / or aspects related to the operation and performance of the call center. Additionally, one or more of the 252 models can be used in natural language processing and, for example, include intent recognition and the like. The 252 models can be developed based on 1) known first-principle equations that describe a system, 2) data, resulting in an empirical model, or 3) a combination of known first-principle equations and data. In developing a model for use with the present embodiments, due to the fact that first-principle equations are frequently unavailable or not Petition 870250102155, dated 07 / 11 / 2025, page 34 / 77 / 43, are easily obtained, it may generally be preferable to create an empirical model based on collected and stored data. To adequately capture the relationship between the manipulated / perturbing variables and the controlled variables of complex systems, it may be preferable that the models 252 be nonlinear. This is because nonlinear models can represent nonlinear relationships instead of linear relationships between manipulated / perturbing variables and controlled variables, which are common to complex systems such as those discussed in the present invention. Given the foregoing requirements, a machine learning or neural network-based approach is currently a preferred embodiment for implementing the models 252. Neural networks, for example, can be developed based on empirical data using advanced regression algorithms.
[0040] The analysis module 250 may additionally include an optimizer 254. As will be recognized, an optimizer can be used to minimize a cost function subject to a set of constraints, wherein the cost function is a mathematical representation of the desired objectives or operation of the system. Because the models 252 may be nonlinear, the optimizer 254 may be a nonlinear programming optimizer. It is contemplated, however, that the present invention may be implemented using, alone or in combination, a variety of different types of optimization approaches, including, but not limited to, linear programming, quadratic programming, mixed integer nonlinear programming, random programming, global nonlinear programming, genetic algorithms, particle / swarm techniques and the like.
[0041] According to exemplary embodiments, models 252 and optimizer 254 can be used together within an optimization system 255. For example, analysis module 250 can use optimization system 255 as part of an optimization process through which Petition 870250102155, dated 07 / 11 / 2025, page 35 / 77 / 43 aspects of the call center's performance and operation are optimized or, at least, improved. This may include, for example, aspects related to customer experience, agent experience, interaction routing, natural language processing, intent recognition, or other functionality related to automated processes.
[0042] Each of the various components, modules and / or servers in Figure 2 (as well as the other figures included in the present invention) may include one or more processors that execute computer program instructions and that interact with other system components to perform the various functionalities described in this document. Such program instructions may be stored in memory implemented using a standard memory device, such as, for example, random access memory (RAM), or stored on other non-transient computer-readable media, such as, for example, a CD-ROM, a flash drive, etc.Although the functionality of each server is described as being provided by the specific server, those skilled in the art should recognize that the functionality of multiple servers may be combined or integrated into a single server, or the functionality of a specific server may be distributed among one or more other servers without departing from the scope of the present invention. Furthermore, the terms interaction and communication are used interchangeably and generally refer to any real-time and non-real-time interaction using any communication channel including, without limitation, telephone calls (PSTN or VoIP calls), emails, v-mails, video, chat, screen sharing, text messages, social media messages, WebRTC calls, etc. Access to and control of the service system components 200 may be affected through user interfaces (UIs) that may be generated on client devices 205 and / or devices. Petition 870250102155, dated 07 / 11 / 2025, page 36 / 77 / 43 from agent 230. As already observed, the call center system 200 can operate as a hybrid system in which some or all components are hosted remotely, such as in a cloud computing or cloud-based environment. Named Entity Recognition and Tagging for PII Redaction / Masking
[0043] In a call center, Named Entity Recognition (NER) has a number of uses. One of them is the identification of personally identifiable information (PII), which includes information that can be used to distinguish or trace the identity of an individual, directly or indirectly, through links to other information and / or information consistent with standards such as the Payment Card Industry-Data Security Standard (PCI-DSS). Call centers collect customer data during interactions and, when using this data for various reasons, need to identify this information so that it can be masked to prevent disclosure. PII identification is one of the main areas where NER is widely used.Another use includes converting unstructured text into structured data. Other downstream uses, as will be observed, show that the extent to which PII can be accurately identified and masked with specific markers significantly improves downstream uses of the data.
[0044] In general, NER is a subfield of natural language processing (NLP) that focuses on identifying and classifying specific data points within textual content. NER works with relevant text details, known as named entities – single words, phrases, or sequences of words – identifying and categorizing them into predefined entity groups or types. The categories cover a diverse range of subjects present in the Petition 870250102155, dated 07 / 11 / 2025, page 37 / 77 / 43 text, including names of individuals, geographical locations, organizational names, dates, events and even specific quantitative values, such as money and percentages.
[0045] Fundamentally, NER revolves around two main steps: identifying entities within the text and categorizing them into distinct groups. Entity detection, often called mention detection or named entity identification, is the initial and fundamental phase in the NER process. It involves systematically scanning and identifying text segments that potentially represent meaningful entities. In its most basic form, a document or sentence is just a long string of characters. Tokenization is the process of dividing this string into meaningful parts called tokens. In English, tokens are often equivalent to words, but they can also represent punctuation or other symbols. This type of segmentation simplifies subsequent analytical steps, converting the text into manageable units. The next challenge is understanding the meaning of these tokens. This is where feature extraction becomes relevant.It involves analyzing the properties of tokens, such as: morphological features that deal with the form of words, such as their roots or prefixes; syntactic features that focus on the arrangement and relationships of words in sentences; and semantic features that capture the inherent meaning of words and may sometimes explore broader world knowledge or context to better understand the role of a token.
[0046] The subsequent phase in the named entity recognition process, after entity detection, is entity classification. Entity classification involves assigning the identified entities to specific categories or classes based on their semantic meaning and context. These categories—or entity types—can range from person and organization to location, date, and a myriad of other markers. Petition 870250102155, dated 07 / 11 / 2025, pp. 38 / 77 / 43, depending on the application requirements. Entity classification is a nuanced process, requiring a deep understanding of the context in which entities appear. This classification leverages linguistic, statistical, and sometimes domain-specific knowledge. For example, while Apple in a technology sphere might refer to the technology company, in a culinary context it is more likely to mean the fruit. Such ambiguities in natural language often require linguistic analysis or advanced NER models trained on extensive datasets to differentiate possible meanings.
[0047] The method of implementing NER has evolved significantly over the years, with multiple approaches being developed to address its challenges. Recently, NER has become heavily reliant on machine learning models. In the field of traditional machine learning methods for NER, models are trained with data where entities are tagged. For example, in the sentence "Paris is the capital of France," the words "Paris" and "France" can be tagged as geographic entities. Such methods rely heavily on feature engineering, where specific attributes and information about the data are manually extracted to improve model performance.The features commonly designed include: word characteristics (details such as word capitalization, prefixes, and suffixes); context (surrounding, preceding, or following words that offer clues); syntactic information (speech part labels that clarify a word's function in a sentence, such as whether it is a noun, a verb, an adjective, etc.); word patterns (the shape or pattern of a word, especially for recognizing specific formats such as dates or vehicle license numbers); and morphological details (information derived from a word's root or its morphological nuances). After the features are prepared, the model is trained with this enriched data. Algorithms. Petition 870250102155, dated 07 / 11 / 2025, page 39 / 77 / 43. Common tools employed in this document include Support Vector Machines (SVMs), Decision Trees, and Conditional Random Fields (CRFs). Once trained, the model can then predict or annotate named entities in raw, unlabeled data.
[0048] An example NER workflow might include the following. First, conversation transcripts are derived from agent-client conversations. These can be derived from text-based agent-client chats or ASR transcripts of voice conversations. The conversation transcripts are sent to the NER system for processing.
[0049] Secondly, the NER system (i.e., trained model) identifies the named entities and then classifies these entities according to various entity types. As an example, such entity types may include any of the following: Entity Type
[0050] GEO Geographic Locations and Addresses PERSON Proper names of people PRICE Currency or money DATE Dates DATE_REL Relative date entities, such as tomorrow, yesterday, etc. ORG Proper names of organizations TIME: Time entities, including approximate time, for example, morning, afternoon, night, etc. CARD_NUMBER Debit or credit card numbers SSN US Social Security number PHONE_NUMBER US telephone numbers ZIPCODE US postal codes Petition 870250102155, dated 07 / 11 / 2025, page 40 / 77 / 43 URL Website or URLs EMAIL Email addresses WORD_DIGIT_SEQ Digits that are spelled out as nineteen eighty-seven MISC Alphanumeric DIGIT Numeric values NONE Non-entity class PRODUCT: Product or brand names, such as Apple or M&Ms. PII_OTHER License plate numbers, account numbers, etc. USER_INFO Usernames and passwords DICTATION Spellout entity class
[0051] Note that not all of these entities are PII / PCI-DSS confidential information. Instead, the list above is representative of the entity types that can be used in a contact center domain.
[0052] Thirdly, once the entity type is identified, the task then becomes a matter of deciding whether the entity type consists of PII or PCI-DSS information.
[0053] Fourthly, if it is determined that the recognized entity belongs to the PII / PCI-DSS information category, the associated words / tokens must then be redacted, i.e., masked.
[0054] Fifth, in such functionality, it is natural to have a set of generic entity types that can serve the purpose of capturing entity types that do not fit into any of the specific entity types. As provided in the list above, the generic entity types include DIGIT, MISC, and WORD_DIGIT_SEQ.
[0055] The need to add new types regularly arises. Petition 870250102155, dated 07 / 11 / 2025, page 41 / 77 / 43 entities to the list above. As an example, suppose a new entity type called ACCOUNT_ID is needed and defined as a 10-digit number. There are two ways to accomplish this, that is, two ways to make a system capable of achieving the identification and masking functionality described above in relation to this new entity type. In the first way, a regular expression that encapsulates the associated pattern can be defined. In the second, the NER model can be retrained so that the model can identify instances of the new entity type. For example, the model used to predict entities, which can be called non-deterministic entities, is trained with training statements that are augmented with the new entity type. However, there are disadvantages associated with either of these conventional approaches.Regarding the first point, someone needs to be familiar with each of the expressions in which the new entity type appears, which is rare. Regarding the second point, a sufficiently large number of associated statements need to be generated for the probabilistic model to learn the new entity type sufficiently. As will be observed, this can be an expensive and time-consuming process. Furthermore, new patterns can conflict with existing ones. For example, a new 10-digit pattern might conflict with an already existing deterministic entity type, say, PHONE_NUMBER, which could also potentially be a 10-digit number.
[0056] According to the present disclosure, methods and systems are proposed to make named entity tagging and masking systems more efficient by making them efficiently extensible. In exemplary embodiments, this is accomplished through defined parameters that specify a search and lookup process. As will be observed, the method can be used to extend named entity tagging and masking in a conversation transcript. Petition 870250102155, dated 07 / 11 / 2025, pp. 42 / 77 / 43 to encompass a new entity type. In such cases, the process may be initiated with respect to a transcript of a conversation between participants (e.g., a client and an agent) that has been processed such that the actual entities recognized therein (i.e., Apple, Mr. Smith, etc.) have already been replaced by the respective entity names that already exist in association with existing entity types. Such existing entity types, for example, the list provided above, may include both specific entity types and generic entity types, as described above.
[0057] According to exemplary embodiments, the following parameters may be used to define the new entity type. Such parameters, for example, may be defined by input received from a user.
[0058] The first parameter that can be included is called in this document a search trigger parameter. The search trigger parameter specifies whether a search trigger to fire off a search process constitutes a keyword type, in which case it is called a keyword search trigger, or an entity type, in which case it may be called an entity search trigger.
[0059] The second parameter that can be included is called a keyword trigger parameter in this document. The keyword trigger parameter specifies one or more keywords that, when detected in a conversation transcript, trigger the search process. The keyword trigger parameter is used when the search trigger parameter specifies that the search trigger is a keyword search trigger. For example, for the new example entity type ACCOUNT_ID, a list of keywords can be included in this parameter, such as: account id, accountid, account id, account number, number of the account, number on the account. Each time one of these keywords is found in the conversation transcript, Petition 870250102155, dated 07 / 11 / 2025, page 43 / 77 / 43 a research process is triggered.
[0060] The third parameter that can be included is called an entity trigger parameter in this document. The entity trigger parameter specifies one of the selected specific entity types (or selected specific entity type) that, when detected in a conversation transcript, triggers the search process. The entity trigger parameter is used when the search trigger parameter specifies that the search trigger is an entity search trigger. As an example, from the list above, the entity type CARD_NUMBER can be specified by this parameter. Each time this entity type is found in the conversation transcript, a search process is triggered.
[0061] The fourth parameter that can be included is referred to in this document as a search entity parameter. The search entity parameter specifies one of the selected generic entity types (or selected generic entity type) that is searched in the conversation transcript in response to the triggering of the search process. For example, the search entity parameter could be one of the generic entity types in the list above, such as DIGIT, MISC, and WORD_DIGIT_SEQ.
[0062] The fifth parameter that can be included is referred to in this document as an entity name parameter. The entity name parameter specifies an entity name for the new entity type. The entity name is the identifier used to replace the first generic entity type when detected during the search process. For the new example entity type, this parameter can be ACCOUNT_ID or some other similar variation.
[0063] The sixth parameter that can be included is referred to in this document as a search entity parameter number. The number Petition 870250102155, dated 07 / 11 / 2025, page 44 / 77 / 43, the search entity parameter specifies the number of times the selected generic entity type must be repeated sequentially before the search entity is considered detected when executing the search process. Therefore, for example, if the new entity type ACCOUNT_ID is a 10-digit number, this parameter could be set to 10, so that the entity type DIGIT needs to be repeated 10 times before it is considered detected.
[0064] Other parameters that may be included are one or more search parameters that specify a search range within the conversation transcript to conduct the search for the generic entity type selected when performing the search process. In certain embodiments, the one or more search parameters may include a participant parameter that specifies one or more of the participants in the conversation transcript to limit the triggering of the search process to only instances where the detected search trigger is found in a turn associated with the one or more specified participants. As used herein, a turn refers to a speaking turn in which a participant speaks, which occurs between turns in which a different participant is speaking.As an example, this parameter can be specified as internal or agent if you want to search only for shifts in which the agent is speaking, or as external or client if you want to search only for shifts in which the client is speaking. This parameter can also be specified as both to search for shifts in which either person is speaking.
[0065] One or more search parameters may additionally include a turn number parameter that specifies the number of turns to which the search is limited in the performance of the search process. The number of turns that are specified, for example, 5 or 10 turns, as part of this parameter may be measured consecutively from the turn in which the search trigger is detected. One or more Petition 870250102155, dated 07 / 11 / 2025, page 45 / 77 / 43 search parameters may additionally include a forward / backward search parameter that specifies whether the number of turns specified in the number of turns parameter is measured forward, backward, or forward and backward in the conversation transcript relative to the turn in which the search trigger is detected.
[0066] With reference now to Figure 3, an exemplary flow process 300 is provided which shows a method of the present invention. Process 300 includes a conventional Named Entity Recognition System 305 as well as a Post-Processing System 310. The Post-Processing System 310 represents the functionality associated with the present invention, which may include steps involving searching, triggering a search process, executing the search process, and masking according to the parameters described above that define one or more new entity types. In this case, a new exemplary entity type, ACCOUNT_ID, is provided. Process 300 begins by receiving the conversation script 315, which is derived from a conversation between an agent and a client, for input into the Named Entity Recognition System 305. As shown, the conversation transcript 315 includes the following three turns: Agent: Could you please share your account ID for verification? Customer: Sure, it's 5 5 7 3 8 0 0 0 4 6.
[0067] Agent: Great, your account has been verified.
[0068] Conversation transcript 315 is then provided as an input to the Named Entity Recognition System 305, which initially processes it using a trained NER model according to the list of entity types introduced above. After initial processing, the conversation transcript can be transformed into an edited conversation transcript 320, which is shown as the output of the Named Entity Recognition System 305. As the entity type Petition 870250102155, dated 07 / 11 / 2025, pages 46 / 77 / 43 ACCOUNT_ID is a new entity type; the NER model was not trained on it and does not recognize it. Thus, although in the edited conversation transcript 320, the Named Entity Recognition System 305 recognizes the various digits as an entity, it can only replace the digits with a generic masking marker, which is shown as repeated DIGIT: Agent: Could you please share your account ID for verification? Customer: Sure, it is <digit> <digit> <digit> <digit> <digit> <digit> <digit> <digit> <digit> <digit> <digit>.
[0069] Agent: Great, your account has been verified.
[0070] The Named Entity Recognition System output 305 is then provided as an input to the Post-Processing System 310, which provides further processing according to the parameters that define the new ACCOUNT_ID entity type. The results are provided in a post-processed conversation script 325 as follows: Agent: Could you please share your account ID for verification? Customer: Sure, it is<ACCOUNT_ID> .
[0071] Agent: Great, your account has been verified.
[0072] As will be observed, the result shown in the post-processed conversation script 325 can be efficiently achieved based on the following parameter values associated with the new entity type ACCOUNT_ID. First, the search trigger parameter specifies a keyword search trigger. Second, the keyword trigger parameter specifies account id as one of the keywords. Other variations can also be provided as keywords. Third, the search entity parameter Petition 870250102155, dated 07 / 11 / 2025, page 47 / 77 / 43 specifies DIGIT as the selected generic entity type. And fourthly, the entity name parameter specifies ACCOUNT_ID as the name of the new entity type to be used in replacing the selected generic entity type, once properly detected. Thus, upon obtaining the result, the keyword "account id" was searched in the conversation script. Then, once the keyword is detected, the conversation script searches for DIGIT. Provided that DIGIT appears in the conversation script in a way that satisfies all applicable search parameters (such as search parameters that define a search range in number of turns, whether those turns are measured forward and / or backward, which participant indicates the keyword, etc.), then the repeated DIGIT will be replaced by the name of the new entity type, as indicated.It is observed that, in this case, the search entity parameter number may have been used to specify the need to repeat DIGIT 10 times for detection to occur. The use of this parameter can be used in cases similar to the new example entity type ACCOUNT_ID, since it is known to be a 10-digit number. As part of the search process, it is verified that the search entity (which is DIGIT) is repeated sequentially 10 times before being considered for masking purposes.
[0073] With reference now to Figure 4, an exemplary flow process 400 is provided which shows a method of the present invention. As in Figure 3, process 400 includes a conventional Named Entity Recognition System 405 as well as a Post-Processing System 410. The Post-Processing System 310 represents the functionality associated with the present invention, which may include steps involving searching, triggering a search process, executing the search process and masking according to the parameters described above that define one or more new entity types. In this case, a Petition 870250102155, dated 07 / 11 / 2025, page 48 / 77 / 43, new exemplary entity type CARD_EXPIRATION_DATE. Process 400 begins by receiving conversation script 415, which is derived from a conversation between an agent and a client, for entry into the Named Entity Recognition System 405. As shown, conversation transcript 415 includes the following four turns: Agent: Could you please share the last four digits of your card? Customer: Sure, it's 8004.
[0074] Agent: Can you tell me its expiration date as well? Client: Yes, it's November 2025.
[0075] Conversation transcript 415 is then provided as an input to Named Entity Recognition System 405, which initially processes it using a NER model trained according to the list of entity types introduced above. After initial processing, the conversation transcript can be transformed into an edited conversation transcript 420, which is shown as the output of Named Entity Recognition System 405. Because the entity type CARD_EXPIRATION_DATE is a new entity type, the NER model was not trained with it and does not recognize it. Thus, although in edited conversation transcript 420, Named Entity Recognition System 405 recognizes the various digits as an entity, it can only replace the digits with a generic DATE masking marker: Agent: Could you please share the last four digits of your card? Customer: Sure, it is<CARD_NUMBER> .
[0076] Agent: Can you tell me its expiration date as well? Client: Yes, it is. <date>.
[0077] The Named Entity Recognition System output Petition 870250102155, dated 07 / 11 / 2025, page 49 / 77 / 43 405 is then provided as an input to the Post-Processing System 410, which provides further processing according to the parameters that define the new entity type CARD_EXPIRATION_DATE. The results are provided in a post-processed conversation script 425 as follows: Agent: Could you please share the last four digits of your card? Customer: Sure, it is<CARD_NUMBER> .
[0078] Agent: Can you tell me its expiration date as well? Client: Yes, it is.<CARD_EXPIRATION_DATE> .
[0079] As will be observed, the result shown in the post-processed conversation script 425 can be efficiently achieved based on the following parameter values associated with the new entity type CARD_EXPIRATION_DATE. First, the search trigger parameter specifies an entity search trigger. Second, the entity trigger parameter specifies CARD_NUMBER as the specific entity type selected. Third, the search entity parameter specifies DATE as the generic entity type selected. And fourth, the entity name parameter specifies CARD_EXPIRATION_DATE as the name of the new entity type to be used in replacing the selected generic entity type, once properly detected. Thus, in obtaining the result, the conversation script searched for the entity trigger parameter specified as CARD_NUMBER.Once this is detected and triggers a search process, the search entity parameter specifies DATE as the selected generic entity type. Upon finding DATE during the search process, the process can then proceed with reviewing the date masking with the entity name of the new entity type. Again, for this to happen, DATE must have been found in... Petition 870250102155, dated 07 / 11 / 2025, page 50 / 77 / 43, an area of the conversation transcript that is consistent with all specified search parameters. In this case, the new entity type is one that specifies a specific type of DATE, which, in this case, is a card expiration date with a given entity name of CARD_EXPIRATION_DATE. This is used to replace DATE to achieve the post-processed conversation script 425.
[0080] As demonstrated in each of these examples, the functionality of the present invention provides a way to efficiently add new entity types for tagging and masking conversation transcripts, so that more accurate and precisely tagged results are achieved. As shown, in the first example, a search process was triggered by a keyword search trigger, which resulted in a generic string of digits being more accurately identified and masked as the new entity type ACCOUNT_ID. In the second example, a search process was triggered by an entity search trigger, which resulted in a generic date being more accurately identified and masked as the new entity type CARD_EXPIRATION_DATE.
[0081] With reference now to Figure 5, an exemplary 500 method is shown for extending a tagging and masking of recognized entities in a conversation transcript to encompass new entity types. The 500 method may include initial steps that are not represented. For example, an initial step might include receiving a transcript of a conversation between participants that has been processed so that the actual entities recognized in it have been replaced by their respective entity names. Each entity name may be associated with an entity type. Entity types may include specific entity types and generic entity types. Another initial step might include receiving parameters that define a new entity type. Petition 870250102155, dated 07 / 11 / 2025, page 51 / 77 / 43 The parameters may include: a search trigger parameter that specifies whether a search trigger to initiate a search process comprises a keyword search trigger or an entity search trigger; a keyword trigger parameter that, when the search trigger comprises the keyword search trigger, specifies one or more keywords that, when detected in the conversation transcript, trigger the search process; an entity trigger parameter that, when the search trigger comprises the entity search trigger, specifies a type selected from among the specific entity types, which is hereinafter referred to as a selected specific entity type, which, when detected in the conversation transcript, triggers the search process;A search entity parameter that specifies one of the selected generic entity types, hereinafter referred to as a selected generic entity type, which is searched in the conversation transcript in response to the triggering of the search process; one or more search parameters that specify a search range within the conversation transcript to conduct the search for the selected generic entity type in the performance of the search process; and an entity name parameter that specifies an entity name for the new entity type that is used to replace the selected generic entity type when detected during the search process.
[0082] Method 500 may include steps associated with an initial process that is performed in relation to the received conversation transcript according to the received parameters that define the new entity type. Method 500 may begin at step 505 by searching for one of the keyword search trigger or the entity search trigger and initiating a search process in response to its detection. Method 500 may continue at step 510 by performing the triggered search process by searching the search range specified in the received parameters for Petition 870250102155, dated 07 / 11 / 2025, page 52 / 77 / 43 the selected generic entity type and detect the selected generic entity type within it. Method 500 can continue in step 515 by revising the conversation transcript so that the entity name of the detected selected generic entity type is replaced with the entity name for the new entity type. Method 500 can continue in step 520 by producing a revised version of the conversation transcript that includes the revision. In other embodiments, other parameters, including search parameters, may also be included. In certain embodiments, the conversation script may consist of text derived from a conversation between an agent in a call center and a customer. In certain embodiments, the received parameters are provided through user input.Furthermore, the received conversation transcript may be a conversation transcript that was first processed by a machine learning model trained for Named Entity Recognition. According to exemplary modalities, the output of the revised version of the conversation transcript may be passed to a downstream process that requires specific tagging of Personally Identifiable Information types to restrict access to them.
[0083] As one skilled in the art will understand, the many different features and configurations described above in relation to the various exemplary embodiments can be further selectively applied to form other possible embodiments of the present invention. For the sake of brevity and considering the abilities of one skilled in the art, each of the possible iterations is not provided or discussed in detail, although all possible combinations and embodiments encompassed by the various claims below or otherwise are intended to form part of the present application. Furthermore, from the above description of various exemplary embodiments of the invention, those skilled in the art will perceive improvements, changes, and modifications. Such improvements, changes, and Petition 870250102155, dated 07 / 11 / 2025, p. 53 / 77 / 43 modifications to the scope of the practice of the technique are also intended to be covered by the appended claims. Furthermore, it should be evident that the foregoing refers only to the embodiments described in this application and that numerous alterations and modifications may be made to this document without departing from the spirit and scope of this application, as defined by the following claims and their equivalents. Petition 870250102155, dated 07 / 11 / 2025, p. 54 / 77< / date> < / digit> < / digit> < / digit> < / digit> < / digit> < / digit> < / digit> < / digit> < / digit> < / digit> < / digit>
Claims
1 / 8 CLAIMS 1. A method for extending a tagging and masking of recognized entities in a conversation transcript to encompass new entity types, the method being characterized in that it comprises the steps of: receiving a conversation transcript of a conversation between participants; identifying real entities in the conversation transcript and determining an entity type for each of the real entities, the entity type being selected from a list of specific entity types or a list of generic entity types; modifying the conversation transcript by masking each of the real entities with an entity name associated with the respective entity type;receive parameters that define a new type of entity, wherein the parameters comprise: a search trigger parameter that specifies whether a search trigger to initiate a search process comprises a keyword search trigger or an entity search trigger; a keyword trigger parameter that, when the search trigger comprises the keyword search trigger, specifies one or more keywords that, when detected in the conversation transcript, trigger the search process; an entity trigger parameter that, when the search trigger comprises the entity search trigger, specifies a type selected from among the specific entity types, which is hereinafter referred to as a selected specific entity type, which, when detected in the conversation transcript, triggers the search process;a search entity parameter that specifies one of the selected generic entity types, which is hereinafter referred to as a selected generic entity type, that is searched in the conversation transcript in response to the triggering of the search process; one or more search parameters that specify a search range within the conversation transcript to conduct the search for the selected generic entity type in the performance of the search process; an entity name parameter that specifies an entity name for the new entity type that is used to replace the selected generic entity type when detected during the search process; in response to receiving the parameters for the new entity type, trigger the search process if either the keyword search trigger or the entity search trigger is detected in the modified conversation transcript;In response to the detection of the selected generic entity type within the search range specified in the parameters received for the new entity type, update the modified conversation transcript by replacing the entity name of the selected generic entity type with the entity name for the new entity type; and produce the updated version of the modified conversation transcript.
2. A method according to claim 1, characterized in that the conversation script comprises text derived from a conversation between an agent in a call center and a customer; and in that the new entity type comprises Personally Identifiable Information.
3. Method according to claim 2, characterized in that one or more search parameters comprise a participant parameter that specifies one or more of the participants in the modified conversation transcript to limit the triggering of the search process to only instances where the detected search trigger is found in a turn associated with one or more of the specified participants.
4. Method according to claim 2, characterized in that one or more search parameters comprise a turn parameter number that specifies a number of turns to which the search is limited in the query, the number of turns being measured consecutively from the turn in which the search trigger is detected.
5. Method according to claim 4, characterized in that one or more search parameters comprise a forward / backward search parameter that specifies whether the number of turns of the turn number parameter is measured forward, backward, or forward and backward in the modified conversation transcript relative to the turn in which the search trigger is detected.
6. Method according to claim 2, characterized in that the received parameters additionally comprise a search entity parameter number that specifies a number of times the selected generic entity type must be repeated sequentially before the search entity is considered detected when executing the search process.
7. Method according to claim 2, characterized in that the one or more search parameters comprise: a participant parameter that specifies one or more of the participants in the modified conversation transcript to limit the triggering of the search process to only instances where the detected search trigger is found in a turn associated with the one or more specified participants; a turn number parameter that specifies a number of turns to which the search is limited in the search, wherein the number of turns is measured consecutively from the turn in which the search trigger is detected; and a forward / backward search parameter that specifies whether the number of turns of the turn number parameter is measured forward, backward, or forward and backward in the modified conversation transcript relative to the turn in which the search trigger is detected.
8. Method according to claim 7, characterized in that the received parameters additionally comprise a search entity parameter number that specifies a number of times the selected generic entity type must be repeated sequentially before the search entity is considered detected when executing the search process.
9. A method according to claim 2, characterized in that the received parameters comprise input received from a user; and in that the output of the updated version of the modified conversation transcript is transmitted to a downstream process that requires specific marking of Personally Identifiable Information types to limit access to them.
10. Method according to claim 2, characterized in that the modified conversation transcript comprises the conversation transcript that has been processed by a machine learning model trained for Named Entity Recognition.
11. System for extending the marking and masking of recognized entities in a conversation transcript to encompass new entity types, the system being characterized in that it comprises: a processor; and a memory that stores instructions which, when executed by the processor, cause the processor to perform the following steps: receiving a conversation transcript of a conversation between participants; identifying real entities in the conversation transcript and determining an entity type for each of the real entities, the entity type being selected from a list of specific entity types or a list of generic entity types; modifying the conversation transcript by masking each of the real entities with an entity name associated with the respective entity type;receive parameters that define a new type of entity, wherein the parameters comprise: a search trigger parameter that specifies whether a search trigger to initiate a search process comprises a keyword search trigger or an entity search trigger; a keyword trigger parameter that, when the search trigger comprises the keyword search trigger, specifies one or more keywords that, when detected in the conversation transcript, trigger the search process; an entity trigger parameter that, when the search trigger comprises the entity search trigger, specifies a type selected from among the specific entity types, which is hereinafter referred to as a selected specific entity type, which, when detected in the conversation transcript, triggers the search process;a search entity parameter that specifies one of the selected generic entity types, hereinafter referred to as a selected generic entity type, which is searched in the conversation transcript in response to the triggering of the search process; Petition 870250102155, dated 11 / 07 / 2025, p. 74 / 77 6 / 8 one or more search parameters that specify a search range within the conversation transcript to conduct the search for the selected generic entity type in the performance of the search process; an entity name parameter that specifies an entity name for the new entity type that is used to replace the selected generic entity type when detected during the search process; in response to receiving the parameters for the new entity type, trigger the search process if either the keyword search trigger or the entity search trigger is detected in the modified conversation transcript;In response to the detection of the selected generic entity type within the search range specified in the parameters received for the new entity type, update the modified conversation transcript by replacing the entity name of the selected generic entity type with the entity name for the new entity type; and produce the updated version of the modified conversation transcript.
12. System according to claim 11, characterized in that the conversation script comprises text derived from a conversation between an agent in a call center and a customer; and in that the new entity type comprises Personally Identifiable Information.
13. System according to claim 12, characterized in that one or more search parameters comprise a participant parameter that specifies one or more of the participants in the modified conversation transcript to limit the triggering of the search process to only instances where the detected search trigger is found in a turn associated with the one or more specified participants. Petition 870250102155, dated 11 / 07 / 2025, pp. 75 / 77 7 / 8 14. System according to claim 12, characterized in that one or more search parameters comprise a turn parameter number that specifies a number of turns to which the search is limited, the number of turns being measured consecutively from the turn in which the search trigger is detected.
15. System according to claim 14, characterized in that one or more search parameters comprise a forward / backward search parameter that specifies whether the number of turns of the turn number parameter is measured forward, backward, or forward and backward in the modified conversation transcript relative to the turn in which the search trigger is detected.
16. System according to claim 12, characterized in that the received parameters additionally comprise a search entity parameter number that specifies a number of times the selected generic entity type must be repeated sequentially before the search entity is considered detected when executing the search process.
17. System according to claim 12, characterized in that the one or more search parameters comprise: a participant parameter that specifies one or more of the participants in the modified conversation transcript to limit the triggering of the search process to only instances where the detected search trigger is found in a turn associated with the one or more specified participants; a turn number parameter that specifies a number of turns to which the search is limited in the search, the number of turns being measured consecutively from the turn in which the search trigger is detected; and Petition 870250102155, dated 11 / 07 / 2025, p. 76 / 77 8 / 8 a forward / backward search parameter that specifies whether the number of turns of the turn number parameter is measured forward, backward, or forward and backward in the modified conversation transcript relative to the turn in which the search trigger is detected.
18. System according to claim 17, characterized in that the received parameters additionally comprise a search entity parameter number that specifies a number of times the selected generic entity type must be repeated sequentially before the search entity is considered detected when executing the search process.
19. System according to claim 12, characterized in that the received parameters comprise the input received from a user; and in that the output of the revised version of the modified conversation transcript is transmitted to a downstream process that requires specific marking of Personally Identifiable Information types to limit access thereto.
20. System according to claim 12, characterized in that the modified conversation transcript comprises the conversation transcript that has been processed by a machine learning model trained for Named Entity Recognition. Petition 870250102155, dated 07 / 11 / 2025, pp. 77 / 77
