System and method for providing order support assistance

Abstract

A system and method for implementing a customer support interface for order support assistance is disclosed. The method includes receiving an assistance request corresponding to a customer account, the customer account associated with an order, the customer account including customer account information, and the order including order attributes. The method also includes predicting whether the assistance request is related to an order, wherein the predicting is based on one or both of the customer account information and the order attributes of the order. In response to predicting that the assistance request is related to the orders, the method determines an expected order from the orders, where the expected order represents the order with the highest probability that the assistance request has been initiated, and causes the expected order to be displayed on the customer support interface on a customer device.

Description

Technical Field

[0001] The present invention generally relates to providing order support assistance to customers. In particular, the present invention relates to a system and method for implementing an online customer support interface. Background Art

[0002] Chatbots are touted to provide many benefits to companies and customers, and have become an integral element of the modern customer service experience. Many businesses (e.g., e-commerce platforms) implement chatbots to help automate routine and repetitive tasks (such as answering general questions) and provide customers with self-service options to help them find answers without the intervention of customer service agents. By offloading simple tasks and customer inquiries, customer service agents can focus on handling complex issues or requests that require human touch.

[0003] However, chatbots are subject to many limitations, the main one being that it is an automated system governed by rules. Therefore, resolving a customer’s query may be long and time-consuming in the minds of most customers, as the chatbot may not directly provide the desired assistance. This may lead to customers bypassing the chatbot entirely and seeking assistance directly from a customer service agent, further straining already limited resources.

[0004] Therefore, given that customers typically inquire about orders on e-commerce platforms, chatbots need to automatically provide proactive support (even before the customer provides any specific input) by predicting and displaying the orders that customers are most likely to seek assistance with, or at least providing a novel chatbot. Summary of the invention

[0005] According to a first aspect of the present invention, a method for implementing a customer support interface for order support assistance is disclosed. The method includes: receiving an assistance request corresponding to a customer account of a customer, the customer account being associated with at least one order, wherein the customer account includes customer account information, and wherein the at least one order includes order attributes. The method also includes: predicting, using a prediction engine implementing a machine learning model, whether the assistance request is related to the at least one order, wherein the prediction is based on one or both of the customer account information, the order attributes of the at least one order.

[0006] In response to predicting that the assistance request is related to the at least one order, the method utilizes the prediction engine to determine an intended order from the at least one order, wherein the intended order represents the at least one order with the highest probability of initiating the assistance request; and displays the intended order on the customer support interface on the customer device.

[0007] According to some aspects of the subject technology, the step of determining an expected order from the at least one order includes the following steps: calculating a score for each of the at least one order based on the order attributes of the at least one order, the score of the at least one order representing the probability that the at least one order has initiated the auxiliary request; and identifying the at least one order with the highest score as the expected order.

[0008] According to some aspects of the subject technology, the method also includes: retrieving order attributes of the expected order; determining, by the prediction engine, one or more prompts based on the order attributes of the expected order; and causing the one or more prompts to be displayed on the customer support interface together with the expected order.

[0009] According to some aspects of the subject technology, the step of determining one or more prompts includes: retrieving a prompt set (also called a set of prompts) from a database; and, according to predefined rules, selecting one or more prompts from the prompt set.

[0010] According to some aspects of the subject technology, the one or more prompts are displayed along with the intended order in a chatbot session on the customer support interface.

[0011] According to some aspects of the subject technology, the method also includes: ranking the at least one order based on the score of the at least one order; and determining at least one other order from the at least one order using the prediction engine, wherein the at least one other order has a lower score than the expected order.

[0012] According to some aspects of the subject technology, the method further includes causing the prospective order to be displayed more prominently on the customer support interface than the at least one other order.

[0013] According to some aspects of the subject technology, the step of causing the expected order to be displayed more prominently on the customer support interface than at least one other order includes the following steps: causing the expected order to be displayed on a first interface of the customer support interface; and causing a selectable link to be presented on the first interface; wherein, in response to a customer selection of the selectable link, the at least one other order is displayed on a second interface of the customer support interface.

[0014] According to some aspects of the subject technology, based on the ranking of the at least one order, the at least one other order is displayed on the second interface.

[0015] According to some aspects of the subject technology, the method also includes: receiving input from the customer confirming that the assistance request is related to the expected order; and updating the machine learning model of the prediction engine.

[0016] According to some aspects of the subject technology, the input from the customer confirming that the auxiliary request is related to the expected order includes any of the following: a query about the expected order typed (enter, also referred to as input) by the customer in the chatbot session, a customer selection of a selectable element representing the expected order, or a customer selection of a selectable element representing the one or more prompts.

[0017] According to some aspects of the subject technology, the method also includes: receiving input from the customer confirming that the auxiliary request is not related to the expected order; and updating the machine learning model of the prediction engine.

[0018] According to some aspects of the subject technology, the input from the customer confirming that the auxiliary request is not related to the expected order includes any of the following: a query typed by the customer in the chatbot session related to an order that is not the expected order, a query typed by the customer in the chatbot session that is not related to the expected order, a customer selection of an optional element that is not related to the expected order, or the customer terminating the chatbot session.

[0019] According to some aspects of the subject technology, the method also includes: receiving input from the customer indicating the relevance of the one or more prompts; and updating the machine learning model of the prediction engine.

[0020] According to some aspects of the subject technology, the input from the customer indicating the relevance of the one or more prompts includes: a query typed by the customer in the chatbot session that is related to the one or more prompts, or a customer selection of a selectable element representing the one or more prompts.

[0021] According to some aspects of the subject technology, the method also includes: receiving input from the customer indicating the irrelevance of the one or more prompts; and updating the machine learning model of the prediction engine.

[0022] According to some aspects of the subject technology, the input from the customer indicating the irrelevance of the one or more prompts includes any of the following: a query typed into the chatbot session that is not relevant to any of the one or more prompts, a customer selection of a selectable element that is not any of the one or more prompts, and the customer terminating the chatbot session.

[0023] According to some aspects of the subject technology, the customer account information includes any one or any combination of the following: customer profile information, customer activity information, historical chat information.

[0024] According to some aspects of the subject technology, the order attributes include any one or any combination of the following: order status, order date, planned shipping date, actual order shipping date, shipping method, order value, order history, payment method, order payment status, merchant identifier, and product identifier.

[0025] According to some aspects of the subject technology, the request for assistance is initiated via activation of a help button on a home page of an e-commerce platform; and wherein the home page of the e-commerce platform does not provide information about the at least one order.

[0026] According to some aspects, the present invention provides a system comprising one or more computers and one or more storage devices storing computer-readable instructions, which when executed by the one or more computers cause the one or more computers to perform the steps of any of the above methods. The same applies to one or more non-transitory computer storage media storing instructions, which when executed by one or more computers cause the one or more computers to perform the steps of any of the above methods. The method can thus be easily transferred to other systems and executed there. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The accompanying drawings are used to illustrate various embodiments by way of example only and to explain various principles and advantages according to embodiments of the present invention. In the accompanying drawings, like reference numerals refer to the same or functionally similar elements throughout the separate views, and the accompanying drawings are incorporated into and form a part of the specification together with the following detailed description.

[0028] Figure 1 A schematic diagram of a system for implementing a customer support interface according to some embodiments is shown.

[0029] Figure 2 A table containing exemplary orders and exemplary order attributes thereof is shown in accordance with certain embodiments.

[0030] Figure 3 A table indicating ranking of orders based on scores of example orders is shown in accordance with certain embodiments.

[0031] Figure 4 A flow chart illustrating a method for implementing a customer support interface according to certain embodiments is shown.

[0032] Figure 5 A flow chart illustrating a method for implementing a customer support interface according to certain embodiments is shown.

[0033] Figure 6 A flow chart illustrating a method for implementing a customer support interface according to certain embodiments is shown.

[0034] FIG. 7A to FIG. 7C An exemplary customer support interface is shown in accordance with certain embodiments.

[0035] FIG. 8A to FIG. 8C An exemplary customer support interface is shown in accordance with certain embodiments.

[0036] The skilled person will appreciate that the elements in the drawings are illustrated for simplicity and clarity and are not necessarily depicted to scale. For example, the size of some of the elements in the block diagram or the steps in the flow chart may be enlarged relative to other elements to help improve the understanding of the embodiments of the present invention. DETAILED DESCRIPTION

[0037] The following detailed description is merely exemplary in nature and is not intended to limit the invention or the application and uses of the invention. Furthermore, there is no intention to be bound by any theory presented in the previous background of the invention or the following detailed description.

[0038] Certain embodiments are directed to teaching a system and method for implementing a customer support interface. The customer support interface may be provided by a business entity (e.g., an e-commerce platform, etc.) as a means of providing order support assistance to its online customers, including but not limited to predicting the orders for which customers are most likely to seek assistance, and causing those orders to be displayed even before any specific input from the customer is received.

[0039] In some embodiments, the method may include the step of receiving an assistance request corresponding to a customer account of the customer. The customer account may be associated with an order placed with the e-commerce platform, and the order may include order attributes. The customer account may include customer account information.

[0040] In some embodiments, the method may include a step of predicting whether the auxiliary request is related to an order (i.e., whether the auxiliary request is order-related). For example, an order may be an order previously placed by a customer on an e-commerce platform. The prediction may be performed by a prediction engine. The prediction engine may implement a machine learning model. The prediction may be based on customer account information and order attributes of the order. Customer account information may non-exhaustively include customer profile information, customer activity information, and historical chat information, etc. Order attributes may include order number, order status, order date, planned shipping date, actual order shipping date, shipping method, order value, order history, payment method, order payment status, merchant identifier, product identifier, etc.

[0041] When the predicted assistance request is not related to any order, the method can cause the non-order related question to be displayed on the customer support interface on the customer device.

[0042] When the assistance request is predicted to be associated with an order, the method may include the step of determining an intended order (from the order), wherein the intended order may represent the order with the highest probability of having initiated an assistance request. The determination may include calculating a score for each order in the order based on order attributes of the order, and the order with the highest score may be identified as the intended order. In other words, the method determines a specific order for which the customer is most likely to seek assistance.

[0043] A request for assistance may be initiated via activating a help button on the home page of the e-commerce platform. The home page of the e-commerce platform need not provide any information about the order.

[0044] The above method is advantageous because it proactively and automatically predicts the expected orders for which the customer is most likely to seek assistance or need help. The expected orders are displayed in the chatbot session on the customer support interface even before the customer provides any input or asks any questions or raises any concerns. This proactive prediction of the expected orders will result in "instant interaction" with the customer, effectively reducing the time and number of interactions required for the chatbot to troubleshoot or identify potential orders.

[0045] Typically, before a chatbot can suggest any order-specific solutions or answers, customers first need to provide detailed information about the order they are having problems with, resulting in a lengthy and undesirable experience for the customer. The present invention, on the other hand, does not require prior input from the customer and automatically predicts the expected orders that the customer is most likely to inquire about. As the chatbot provides relevant assistance, customers become increasingly engaged in chatbot conversations, and the likelihood of transfers to customer service agents can be minimized, thereby improving and optimizing operational efficiency and reducing operating costs.

[0046] The method also selectively performs determination of expected orders only when it has been predicted that the assistance request is order-related (i.e., related to an order previously placed by the customer). Conversely, the method does not perform determination of expected orders when it has been predicted that the assistance request is non-order related. This is advantageous because performing prediction of expected orders every time a customer initiates an assistance request may result in unnecessary resource utilization because the customer may simply ask questions that are not related to any of his / her orders. By determining expected orders only when it is predicted that the assistance request may be order-related, the present invention advantageously saves computing resources (e.g., CPU processing time, memory usage, etc.) and provides efficient management of computing resources, while achieving its goal of providing a more proactive chatbot.

[0047] In some embodiments, the method may include the following steps: determining a prompt based on order attributes of the expected order; and causing the prompt to be displayed on the customer support interface together with the expected order. The prompt may be displayed in the chatbot session on the customer support interface in the expected order together with the expected order. If the customer desires, the prompt may be displayed on the customer support interface in the form of a selectable element for selection. The selectable element may be a prompt card. The prompt may refer to, for example, a suggested question or suggested content related to the expected order. In other words, the prompt refers to a possible query or request that the customer may make related to the expected order.

[0048] Even before the customer provides any specific input on the customer support interface, it is advantageous to display prompts (e.g., suggested questions or content) for the customer to select, because this helps communicate about the expected order while creating a personalized chat experience for the customer. For example, one or more prompts can be automatically suggested based on the expected order, and the customer can conveniently select from the prompts to advance the chatbot conversation. Such prompts allow customers to simply and quickly query the expected order, thereby improving the chatbot's resolution rate for order-related issues and shortening the time required for customers to solve problems via the chatbot. For example, a customer can select a desired prompt to access relevant solutions to order-related issues. The prompt may be particularly useful for customers who are unable to correctly express questions about their orders. In addition, proactively providing customers with suggested questions about possible order-related issues can also be solved before the problem occurs, thereby increasing customer satisfaction with the use of the chatbot.

[0049] In some embodiments, the method may include the steps of ranking the orders based on their scores and then determining other orders having lower scores compared to the expected order. "Other orders" may represent orders that are considered to have a lower probability / likelihood of initiating an assistance request than the expected order.

[0050] In some embodiments, the method may cause the expected order to be displayed more prominently on the customer support interface than other orders. For example, the expected order may be displayed on a first interface of the customer support interface having a selectable link, and in response to a customer selection of the selectable link, other orders may be displayed on a second interface of the customer support interface. The selectable link may be a hyperlink, a browser tab, a selectable button, a scrollable element, or any selectable element, etc.

[0051] Since the expected order represents the order with the highest probability / likelihood that the customer has initiated an assistance request, the expected order can be easily distinguished from other orders, thereby making the interface more user-friendly. Therefore, convenient and proactive order support is provided because the customer can conveniently select the order card representing the expected order and immediately make a chat inquiry.

[0052] Furthermore, since mobile devices are widely used to purchase goods online via e-commerce platforms, optimizing screen layouts can greatly improve customer engagement. Since the screen size of customer devices is often limited, it is advantageous to first display the intended order in isolation (without cluttering the screen display with unnecessary elements), and only fully display other orders on a second interface or window when the customer chooses. This provides a well-structured screen layout that is both practical and visually appealing to the customer.

[0053] Figure 1 A schematic diagram of a system 100 for implementing a customer support interface according to some embodiments is shown. The system 100 may include one or more components, modules, or layers, which may be implemented using software and / or hardware, optionally across multiple locations or using multiple devices or units. Figure 1 An overview of the system 100 is described. Thereafter, the system 100 is described in more detail in conjunction with the remaining figures.

[0054] The system 100 may be implemented as a single system, multiple systems, a distributed system, or any other form. Each component of the system 100 may be any type of computing device that is suitable for performing the functions required by one or more embodiments of the present invention and is capable of communicating with each other directly or through a network 110. In some embodiments, the system 100 includes a client device 120 and a server 140 that are communicatively coupled to each other via the network 110.

[0055] Figure 1 The illustrated network 110 may include any number of suitable networks and / or network types. For example, the network 110 may include, but is not limited to, one or more of the following: a virtual private network (VPN), a local area network (LAN), a wide area network (WAN), such as the Internet, wireless WANS (WWANS), a metropolitan area network (MAN), a Wi-Fi network, network, A communication network, a peer-to-peer network, a cellular telephone network, or any combination of such networks. According to one or more embodiments of the present invention, network 110 may include any combination of connections (e.g., wired connections, wireless connections, or fiber optic connections) and protocols that support communication between each component in system 100. For example, data transmitted by network 110 may include packetized data, such as Internet Protocol (IP) data packets, which are routed to a designated computing device coupled to network 110. Network 110 may include network software and / or hardware that interconnects the computing devices of system 100. For example, network 110 may include a packet switching device that routes data packets to various computing devices based on information included in the headers of the data packets.

[0056] Data exchanged over the network 110 may be represented using techniques and / or formats including Hypertext Markup Language (HTML), Extensible Markup Language (XML), JavaScript Object Notation (JSON), Comma Separated Values ​​(CSV), Java Database Connectivity (JDBC), Open Database Connectivity (ODBC), etc. In addition, data exchanged over the network may be encrypted using encryption techniques such as Secure Sockets Layer (SSL), Secure HTTP (HTTPS), and / or Virtual Private Network (VPN) or Internet Protocol Security (IPsec).

[0057] It should be appreciated that custom and / or dedicated data communication technologies may be used instead of or in addition to the technologies described above to perform the functions required by one or more embodiments of the present invention.

[0058] Figure 1 The illustrated server 140 may take the form of one or more network-based server computers, such as a web server, an application server, a database server, etc., which can communicate with the client device 120 via the network 110. For example, the server 140 may be embodied in a single server computer, or distributed as a plurality of interconnected server computers, and operated in a cloud computing environment. The server 140 may include a processor and a memory. The memory may be a non-transitory computer-readable medium of a volatile or non-volatile type, such as a magnetic medium, an optical medium, a random access memory (RAM), a read-only memory (ROM), a removable medium, or any other suitable local or remote memory unit. The memory may include digital data accessible to the processor. The processor may be any type of device or multiple devices capable of manipulating or processing information from a non-transitory computer-readable medium (e.g., a memory). For example, the processor includes a central processing unit (CPU) for processing computer-executable program instructions to execute one or more processes.

[0059] Server 140 may include one or more storage devices storing data structures and / or computer instructions for processor execution. For example, server 140 may include one or more computer-readable data memories for supporting the functions required for one or more embodiments of the present invention. Various types of computing devices may be used for server 140. For example, the computer may be a mainframe, a workstation, and other types of processing devices. The memory may include digital data accessible to the processor. Server 140 and / or data memory may be coupled to various databases or storage devices in the form of any suitable computer-readable medium (such as a hard drive, a memory device, a flash drive, or an optical drive).

[0060] In some embodiments, an e-commerce platform (not shown) can be implemented in whole or in part on server 140. The e-commerce platform can provide merchants with online resources and facilities for managing their business. Merchants can use e-commerce platforms (such as implementing e-commerce experiences with customers through online stores) to manage business with customers. Online stores can refer to a multi-tenant collection of storefronts supported by e-commerce platforms (e.g., for multiple merchants) or a storefront of a single merchant (e.g., an online store of a merchant). For example, server 140 can support a single integrated website (also referred to as a multi-merchant website) that is used as a virtual shopping mall. In some embodiments, the e-commerce platform provides online customer support to customers, such as communicating with customer service agents to provide online customer support to customers by implementing a chatbot system 148 and a real-time chat system (not shown). In some embodiments, the online store is provided by the e-commerce platform to customers 124 who have registered accounts with the e-commerce platform, which will be described in more detail below.

[0061] like Figure 1 The illustrated client devices 120 represent any number of client devices that may be included in the system 100. Each client device 120 is a computing device with data processing and communication capabilities. In some embodiments, the client device 120 may include a processor (e.g., CPU), memory (e.g., RAM, ROM, etc.), a power supply, a network interface, and / or other software and / or hardware components, such as a display, a graphics processor, a wireless transceiver, a keyboard, a speaker, a camera, a sensor, firmware, an operating system, a driver, various physical connection interfaces (e.g., USB, HDMI, etc.). For example, the client device 120 may take the form of one or more personal computers, tablet computers, smart phones, mobile phones, wearable devices, home entertainment computers, network computing devices, mobile computing devices, personal digital assistants, virtual machines, servers, other computing devices, or any combination of such computing devices.

[0062] The customer support interface 122 represents a graphical user interface (GUI) to be displayed on the customer device 120 via a display screen.

[0063] Customer accounts 126 may be associated with customers 124 of the e-commerce platform. For example, customers 124 may have one or more customer accounts 126 registered with the e-commerce platform. Customer accounts 126 may have unique identifiers (e.g., unique numeric strings). Customer accounts 126 may include customer account information 136. Figure 1 The illustrated customer accounts 126 are representative of any number of customer accounts that may be included in the system 100. The customer accounts 126 may be stored in a customer account database 142.

[0064] Client 124 may interact (or interface) with server 140 using client device 120. Figure 1 The illustrated customers 124 represent any number of customers that may be included in the system 100. The customer account information 136 may include customer profile information 128. The customer profile information 128 may include, without limitation, name, address, age, gender, race, nationality, occupation, email contact, contact number, geographic location, information related to one or more forms of payment (electronic wallet, debit / credit card, or bank account), and any other information corresponding to the customer 124.

[0065] The customer account information 136 may include customer activity information 130. The customer activity information 130 may include recent and historical activities of the customers 124, and include, but are not limited to, historical page views, page views indicating product interest or any other page views, payment activity data (e.g., data related to one or more forms of payment activity), and any other activity information corresponding to the customers 124. For example, the customer activity information 130 for each customer 124 may be obtained from clickstream data, including any actions taken by the customer 124, such as adding items to a shopping cart, or navigation activities of the customer 124 in an e-commerce platform, or any other suitable actions taken in other ways.

[0066] The customer account information 136 of the customer 124 may include historical chat information 132 of the customer 124, such as chat session information related to a current chat session or a previous chat session with the customer 124. The historical chat information 132 may include previous chat records of conversations between each customer 124 and a chatbot and / or a human agent from a chat session performed on the customer support interface 122.

[0067] The historical chat information 132 may also include historical resolution data from previous chat interactions with the customer 124. The resolution data may include actions identified as resolving an order-related issue, or actions taken to resolve an order-related issue. For example, actions identified as resolving an order-related issue may include any proactive (i.e., before receiving any specific input from the customer 124) or reactive (i.e., after receiving specific input from the customer 124) actions identified in attempting to resolve an order-related issue.

[0068] The customer account 126 may also include (one or more) orders 134 associated with the customer 124. The order 134 may be an order that the customer 124 has placed. The order 134 may include any order 134 for which payment has been completed, or an order 134 for which items in the shopping cart have been "checked out" but are awaiting payment. Each order 134 may have a unique order number.

[0069] The order 134 may include order attributes 210 (e.g. Figure 2 ). Order attributes 210 may be any parameter or attribute that may convey useful information about order 134. Order attributes 210 and their values ​​may be used to compare and contrast between orders 134. As will be described in detail later, the information provided by order attributes 210 and their values ​​may be used to help predict whether an assistance request is order-related, and also to help determine orders 134 for which customer 124 may seek assistance.

[0070] The order attributes 210 of the order 134 may non-exhaustively include order number, order status, order date, planned shipping date, actual order shipping date, shipping method, order value, order history, payment method, order payment status, merchant identifier, product identifier, etc. Figure 2 2 shows exemplary order attributes 210 and exemplary values ​​thereof. The values ​​of order attributes 210 may change over time. For example, the value of order attribute 210 "order status" may change from "settlement", "shipping" to "shipped". In another example, the value of order attribute 210 "order payment status" may change from "waiting for payment" to "paid".

[0071] Server 140 may include customer account database 142, chatbot system 148, prediction engine 150, and prompt module 152. Customer account database 142 may be a data store that stores customer accounts 126. Prompt module 152 may be a data store that stores sets of prompts.

[0072] The chatbot system 148 can provide the customer 124 with a chatbot and a graphical user interface to communicate with the chatbot. The chatbot system 148 can receive and send input to the customer 124 via the customer device 120 to be displayed on the graphical user interface. For example, the chatbot system 148 can receive an assistance request transmitted by the customer 124 from the customer device 120 to be displayed on the graphical user interface. The graphical user interface is, for example, the customer support interface 122 as described herein, or more specifically, a chatbot session on the customer support interface 122. The chatbot system 148 can determine whether the assistance request is a new conversation request or part of an existing conversation session. The chatbot system 148 can determine whether the conversation session is completed based on a completion signal from the customer device 120 (e.g., the interface window is closed). The chatbot system 148 can be configured to receive input data from the prediction engine 150.

[0073] Prediction engine 150 may include a program or algorithm, such as a machine learning model (e.g., a neural network) that performs functions according to the described embodiments. Figure 1 The illustrated prediction engine 150 is representative of any number of prediction engines that may be included in the system 100. In some embodiments, in response to receiving an assistance request from a customer 124, the prediction engine 150 is configured to obtain customer account information 136 from a customer account 126 of the customer 124 and order attributes 210 of the order 134 from the customer account database 142.

[0074] As previously described, the customer account information 136 may include the customer profile information 128, the customer activity information 130, and the historical chat information 132. Using the customer account information 136 and the order attributes 210 of the order 134, the prediction engine 150 may predict whether the customer 124 assistance request is order-related. For example, order-related requests may include any requests related to order fulfillment issues (e.g., shipping delays), payment issues related to a specific order, order returns or refunds, etc. Non-order-related requests may include any requests related to account verification (e.g., password retrieval, deactivation of an account), general payment inquiries (e.g., payment methods, discount codes, e-wallet inquiries), or product-related issues.

[0075] In some embodiments, the prediction engine 150 can predict whether the customer 124 assistance request is order-related by analyzing the customer account information 136 and the order attributes 210 of the order 134. The analysis can be completed by considering the customer account information 136 and the order attributes 210 of the order 134 together or in sequence.

[0076] The customer account information 136 may be useful information when determining whether the assistance request is order-related. For example, if the customer activity information 130 of the customer account information 136 indicates that the customer 124 has recently repeatedly viewed order history, the prediction engine 150 may predict / determine that the assistance request is order-related. In yet another example, if the historical chat information 132 of the customer account information 136 indicates that the customer 124 has participated in a recent chatbot conversation about an order 134, the prediction engine 150 may predict / determine that the assistance request is order-related. It should be understood that the prediction engine 150 may additionally identify any suitable number of questions associated with the customer account 126 to predict whether the assistance request is order-related.

[0077] Conversely, the customer account information 136 may also be useful information in determining whether the assistance request is non-order related. For example, if the customer profile information 128 of the customer account information 136 indicates that the customer 124 has a frozen e-wallet account, then the prediction engine 150 is likely to predict / determine that the assistance request is non-order related (because the customer 124 is likely asking how to "unfreeze" the account, rather than asking about the order 134).

[0078] In determining whether the assistance request is order-related, the order attributes 210 of the order 134 may be useful information. For example, if the value of the order attribute 210 "order status" of any order 134 indicates that the order 134 is still being shipped, and the value of the order attribute 210 "planned shipping date" has passed, the prediction engine 150 may predict / determine that the assistance request is order-related. In another example, if the value of the order attribute 210 "order status" of any order 134 indicates that the order 134 has been settled, and the value of the order attribute 210 "order payment status" indicates that the order 134 has not yet been paid, that is, "awaiting payment", the prediction engine 150 may predict / determine that the assistance request is order-related.

[0079] In yet another example, if the value of the order attribute 210 "order status" of any order 134 indicates that the order 134 has been completed, and the value of the order attribute 210 "order payment status" indicates that the order 134 has been paid, the prediction engine 150 may predict / determine that the assistance request is non-order related. In yet another example, if the value of the order attribute 210 "order status" of any order 134 indicates that the return or refund of the order 134 has been completed, and the return or refund of the order 134 has been processed, that is, "return / refund completed", the prediction engine 150 may predict / determine that the assistance request is non-order related because the order 134 has no unresolved issues.

[0080] Therefore, when determining whether an auxiliary request is order-related, the order attributes 210 of the order 134 can be analyzed at a macro level, that is, the values ​​of the order attributes 210 of multiple orders 134 are considered together, rather than comparing the values ​​of the order attributes 210 or comparing between the orders 134 themselves.

[0081] In some embodiments, the prediction engine 150 may analyze the order attributes 210 of the orders 134 with reference to a predetermined threshold. For example, the predetermined threshold for the order attribute 210 "planned shipping date" may be "two days later." Therefore, if the value of the order attribute 210 "planned shipping date" of any order 134 is two days from the current date, the prediction engine 150 is likely to determine that the assistance request is order-related.

[0082] In some embodiments, each order attribute 210 may have a predetermined threshold, and if a certain number of the predetermined thresholds are exceeded, the prediction engine 150 may determine that the assistance request is order-related.

[0083] In some embodiments, in response to predicting that the customer 124 assistance request is order-related, the prediction engine 150 can be configured to automatically determine the expected order 310 (e.g., Figure 3 ). The expected orders 310 may represent the orders 134 for which the customer 124 has the highest probability of initiating an assistance request. The determination of the expected orders 310 may be based on the order attributes 210 of the orders 134. The determination may involve calculating a score for each order 134 based on the order attributes 210. The score may represent the probability of the order 134 for which an assistance request has been initiated. The order 134 with the highest score may be identified as the expected order 310 (e.g., Figure 3 shown).

[0084] In some embodiments, the score of an order 134 may be calculated by assigning a weight to each of the order attributes 210, since some order attributes 210 may be more important than others. For example, the order attribute 210 "order status" may be assigned the highest weight. The score of each order 134 may be the sum of the "normalized" values ​​of the order attributes 210 multiplied by their respective weights.

[0085] like Figure 3 As shown, the orders 134 may be ranked based on their scores. The highest ranked order 134 may be the expected order 310. The orders 134 having lower scores (i.e., lower rankings) than the expected order 310 may be other orders 320. As described, the determination of the expected order 310 and other orders 320 may be accomplished by comparing the scores (essentially, values) of the order attributes 210 between the plurality of orders 134, or by comparing between the orders themselves (i.e., comparing at the order level rather than at the macro level).

[0086] The prediction engine 150 can be configured to receive feedback information as input data to update one or more parts of the machine learning model. The feedback information can be "positive" and include customer indications of the accuracy of the prediction (e.g., customer input confirming the accuracy of the prediction). The feedback information can also be "negative" and include customer indications that refute the accuracy of the prediction. Based on the feedback information, the machine learning / neural network model can be trained and the data can be updated. In this way, the prediction engine 150 is continuously trained and refined to generate more accurate predictions.

[0087] The prediction engine 150 may query the prompt module 152 to retrieve a set of prompts. The prediction engine 150 may select one or more prompts from the set of prompts based on the order attributes 210 of the expected order 310. The prompts may be selected via predefined rules. The predefined rules typically include one or more conditions, and one or more actions to be taken if the conditions are met. It should be understood that the predefined rules may include multiple conditions, and there may be complex interrelationships between the conditions.

[0088] For example, if the value of the order attribute 210 "order status" of the expected order 310 is "shipping", the predefined rule may specify that a prompt associated with "shipping" is selected. For example, the prompt associated with "shipping" may be:

[0089] (1) Tip 1—When will I receive my order?

[0090] (2) Tip 2—How do I request tracking or expedited shipping of my order?

[0091] (3) Tip 3—Can I change my delivery address after placing an order?

[0092] In another example, if the value of the order attribute 210 "order payment status" of the expected order 310 is "waiting for payment", the predefined rule may provide for selecting a prompt associated with "waiting for payment". For example, the prompt associated with "waiting for payment" may be:

[0093] (4) Tip 4—Why can’t I pay by credit card?

[0094] (5) Tip 5—Why can’t I use my discount coupon?

[0095] (6) Tip 6—Can I receive a payment confirmation email for my order?

[0096] Thus, predefined rules may cause the prediction engine 150 to select prompts that are considered most relevant to the prospective order 310 .

[0097] In an alternative implementation, the prediction engine 150 may select a prompt based on the relevance of the prompt to the value of the order attribute 210 of the expected order 310. For example, the prompt may be selected based on the probability that the prompt matches the value of the order attribute 210 of the expected order 310. A prompt with a higher probability (e.g., a higher confidence score of matching the order attribute 210) may be selected. For example, when the value of the order attribute 210 "order status" of the expected order 310 is "shipping", the prompt may be selected based on the relevance of prompts 1, 2, or 3.

[0098] It should be understood that Figure 1 The illustrated system 100 represents an exemplary system according to one or more embodiments, and a variety of different system environments and configurations are contemplated and are within the scope of the present disclosure. In addition, the system 100 may include automatic speech recognition (ASR) functionality (not shown) that can enable verbal / voice input (e.g., via VOIP) and / or requests from the customer 124 and text-to-speech (TTS) functionality into the customer support interface 122, and the TTS functionality that can enable verbal output of the system 100 to an end user (or users) of the system 100 (to the customer device 120). However, it should be appreciated that the customer 124 can type text input into the customer support interface 122, and the system 100 can provide output to the customer support interface 122 via text.

[0099] Figure 4 A flow chart illustrating a method 400 for implementing the customer support interface 122 according to some embodiments is shown. The method 400 is explained in conjunction with the system 100 as previously described by the previous figures of the present disclosure, and the same or similar processes and elements may not be described in detail. Figure 4 The steps identified in the drawings are exemplary and may include various alternatives, equivalents or derivatives thereof, including but not limited to the order of their execution. Figure 4 The steps of (and various alternatives thereof) may be embodied in hardware or software, including a computer-readable storage medium comprising instructions executable by a processor or the like in a computing device.

[0100] In step 402, method 400 receives an assistance request corresponding to a customer account 126. In some embodiments, it can be assumed that before this step, customer 124 has logged into his / her registered customer account 126 on the e-commerce platform using customer device 120. Customer account 126 can be associated with an order 134 that customer 124 has placed on the e-commerce platform. Order 134 can include order attributes 210. Customer account 126 can include customer account information 136.

[0101] Customer 124 can initiate a request for assistance via customer device 120 by, for example, clicking a chat interface element (e.g., a “help” button) on the homepage or main page of the e-commerce platform. The homepage / main page of the e-commerce platform does not need to display any information about order 134.

[0102] In response to the assistance request, the method 400 predicts, at step 404, whether the assistance request is related to the order 134 (i.e., whether the assistance request is order-related). For example, order-related requests may include any requests related to order fulfillment issues (e.g., shipping delays), payment issues related to a specific order, order returns or refunds. Non-order-related requests may include any requests related to identity verification (e.g., password retrieval, account deactivation), general payment inquiries (e.g., payment methods, discount codes, e-wallet account inquiries), or product-related issues.

[0103] In some embodiments, the prediction step may be performed by prediction engine 150 . In response to receiving the assistance request, prediction engine 150 may obtain customer account information 136 of customer account 126 from customer 124 and order attributes 210 of order 134 from customer account database 142 .

[0104] The customer account information 136 may include customer profile information 128, customer activity information 130, and historical chat information 132. Using the customer account information 136 and order attributes 210 of the order 134, the prediction engine 150 may predict whether the customer 124 assistance request is order related.

[0105] The prediction engine 150 can predict whether the customer 124 assistance request is order-related by analyzing the customer account information 136 and the order attributes 210 of the order 134. The analysis can be completed by considering the customer account information 136 and the order attributes 210 of the order 134 together or in sequence.

[0106] The customer account information 136 may be useful information in determining whether the assistance request is order-related. For example, if the customer activity information 130 of the customer account information 136 indicates that the customer 124 has recently repeatedly viewed order history, the prediction engine 150 may predict / determine that the assistance request is order-related. In another example, if the historical chat information 132 of the customer account information 136 indicates that the customer 124 has participated in a recent chatbot conversation about the order 134, the prediction engine 150 may predict / determine that the assistance request is order-related.

[0107] Conversely, the customer account information 136 may also be useful information in determining whether the assistance request is non-order related. For example, if the customer profile information 128 of the customer account information 136 indicates that the customer 124 has a frozen e-wallet account, then the prediction engine 150 is likely to predict / determine that the assistance request is non-order related (because the customer 124 is likely asking how to "unfreeze" the account, rather than asking about the order 134).

[0108] In determining whether the assistance request is order-related, the order attributes 210 of the order 134 may be useful information. For example, if the value of the order attribute 210 "order status" of any order 134 indicates that the order 134 is still being shipped, and the value of the order attribute 210 "planned shipping date" has passed, the prediction engine 150 may predict / determine that the assistance request is order-related. In another example, if the value of the order attribute 210 "order status" of any order 134 indicates that the order 134 has been settled, and the value of the order attribute 210 "order payment status" of the order 134 indicates that the order 134 has not been paid, that is, "awaiting payment", the prediction engine 150 may predict / determine that the assistance request is order-related.

[0109] In yet another example, if the value of the order attribute 210 "order status" of any order 134 indicates that the order 134 has been completed, and the value of the order attribute 210 "order payment status" indicates that the order 134 has been paid, the prediction engine 150 may predict / determine that the assistance request is non-order related. In yet another example, if the value of the order attribute 210 "order status" of any order 134 indicates that the return or refund of the order 134 has been completed, and the return or refund of the order 134 has been processed, i.e., "return / refund completed", the prediction engine 150 may predict / determine that the assistance request is non-order related because the order 134 no longer has unresolved issues.

[0110] Thus, when determining whether an auxiliary request is order-related, analysis of the order attributes 210 of the order 134 may be performed at a macro level, i.e., the values ​​of multiple order attributes 210 of multiple orders 134 are considered together, rather than comparing the values ​​of the order attributes 210 or comparing between the orders 134 themselves.

[0111] In some embodiments, the prediction engine 150 may analyze the order attributes 210 of the orders 134 with reference to a predetermined threshold. For example, the predetermined threshold for the order attribute 210 "planned shipping date" may be "two days later." Therefore, if the value of the order attribute 210 "planned shipping date" of any order 134 is two days from the current date, the prediction engine 150 is likely to determine that the assistance request is order-related.

[0112] In some embodiments, each order attribute 210 may have a predetermined threshold, and if a certain number of the predetermined thresholds are exceeded, the prediction engine 150 may determine that the assistance request is order-related.

[0113] If the method 400 predicts that the assistance request is not related to any of the orders 134 (i.e., the assistance request is non-order related), the method proceeds to step 420. In step 420, the method 400 causes a default non-order related question to be displayed on the customer support interface 122. The default non-order related question may be:

[0114] (1)How do I reset my account password?

[0115] (2) How can I avoid scams?

[0116] (3) What should I do if I encounter problems / errors on the platform?

[0117] Therefore, when the assistance request has been predicted to be non-order related, the method 400 does not proceed with determining the expected order 310 .

[0118] If the assistance request has been predicted to be associated with the order 134 (ie, the assistance request is order-related), the method 400 proceeds to step 406. In step 406, the method 400 determines the expected order 310, which represents the order 134 with the highest probability of initiating an assistance request.

[0119] In some embodiments, the determination of the expected orders 310 may be performed by the prediction engine 150. The determination may be based on the order attributes 210 of the orders 134. The determination may involve calculating a score for each order 134 based on the order attributes 210. The score may represent the probability of the order 134 for which an assistance request has been initiated. The order 134 with the highest score may be identified as the expected order 310 (e.g., Figure 3 shown).

[0120] The score of the order 134 can be calculated by assigning a weight to each of the order attributes 210, because some order attributes 210 may be more important than other order attributes. For example, the order attribute 210 "order status" may be assigned the highest weight. The score of each order 134 may be the sum of the "normalized" values ​​of the order attributes 210 multiplied by the corresponding weights.

[0121] The orders 134 may be ranked based on their scores (eg Figure 3). The highest ranked order 134 may be the expected order 310. Orders 134 having lower scores (i.e., lower rankings) than the expected order 310 may be other orders 320. As described, the determination of the expected order 310 and other orders 320 may be accomplished by comparing the scores (essentially, values) of the order attributes 210 between the multiple orders 134, or may be accomplished by comparing the scores of the order attributes 210 between the orders themselves (i.e., at the order level rather than at the macro level).

[0122] In step 408, method 400 causes the expected order 310 to be displayed on customer support interface 122. The expected order 310 can be displayed in the form of an order card. The order card representing the expected order 310 can be displayed in the chatbot session.

[0123] In some embodiments, the results of the chatbot session and the chat record of the chatbot session can be stored as customer classification data in the historical chat information 132 of the customer account 126. For example, the customer classification data can include, but is not limited to, a determination as to whether the assistance request of the customer 124 is order-related, a determination as to whether the chatbot session was terminated before the customer selection was made, or a combination thereof. The acquired customer classification data can then be used to train the machine learning model / neural network model of the prediction engine 150 to make better predictions and enable the prediction engine 150 to dynamically adapt to the needs of the customer 124 that may change over time.

[0124] By selectively performing the determination of expected orders 310 only when the prediction assistance request is order-related, the method advantageously saves computing resources (e.g., CPU processing time, memory usage, etc.) and provides efficient management of computing resources while achieving its goal of providing a more proactive chatbot.

[0125] Furthermore, the method 400 as described proactively and automatically predicts the expected orders 310 for which the customer 124 is most likely to seek assistance or need help, even before the customer 124 has provided any input or asked any questions or raised any concerns. This proactive prediction of the expected orders 310 will result in "live interaction" with the customer 124, effectively reducing the time and number of interactions required for the chatbot to troubleshoot or identify potential orders 134. As the chatbot provides relevant assistance and the customer 124 becomes increasingly engaged in the chatbot session, the likelihood of transfers to customer service agents can be minimized, thereby improving and optimizing operational efficiency and reducing operating costs.

[0126] Figure 5A flow chart illustrating a method 500 for implementing the customer support interface 122 according to certain embodiments is shown. The method 500 is explained in conjunction with the system 100 and the method 400 as previously described by previous figures of the present disclosure, and the same or similar processes and elements may not be described in detail.

[0127] In step 502, method 500 receives an assistance request corresponding to customer account 126. Step 502 is similar to step 402 of method 400, which has been described in detail above. For the sake of brevity, step 502 of method 500 will not be described again.

[0128] In step 504, the method 500 predicts whether the assistance request is related to the order 134 (e.g., the order 134 previously placed by the customer 124 on the e-commerce platform) (i.e., whether the assistance request is order-related). Step 504 is similar to step 404 of the method 400 described in detail above. For the sake of brevity, step 504 of the method 500 will not be described again.

[0129] In step 506, method 500 determines an expected order 310 representing the order 134 with the highest probability of having initiated an assistance request. Step 506 is similar to step 406 of method 400 described in detail above. For the sake of brevity, step 506 of method 500 will not be described again.

[0130] In step 508, the method 500 determines prompts based on the order attributes 210 of the expected order 310. Prompts can refer to, for example, suggested questions or suggested content related to the expected order 310. In other words, prompts refer to possible queries or requests related to the expected order 310 that the customer 124 may ask.

[0131] The determination of prompts may be performed by the prediction engine 150. The prediction engine 150 may query the prompt module 152 to retrieve a set of prompts. The prediction engine 150 may select one or more prompts from the set of prompts based on the order attributes 210 of the expected order 310. The prompts may be selected via predefined rules. The predefined rules typically include one or more conditions, and one or more actions to be taken if the conditions are met. It should be understood that the predefined rules may include multiple conditions, and there may be complex interrelationships between the conditions.

[0132] For example, if the value of the order attribute 210 "order status" of the expected order 310 is "shipping", the predefined rule may specify that a prompt associated with "shipping" is selected. For example, the prompt associated with "shipping" may be:

[0133] (1) Tip 1—When will I receive my order?

[0134] (2) Tip 2—How do I request tracking or expedited shipping of my order?

[0135] (3) Tip 3—Can I change my delivery address after placing an order?

[0136] In another example, if the value of the order attribute 210 "order payment status" of the expected order 310 is "waiting for payment", the predefined rule may provide for selecting a prompt associated with "waiting for payment". For example, the prompt associated with "waiting for payment" may be:

[0137] (4) Tip 4—Why can’t I pay by credit card?

[0138] (5) Tip 5—Why can’t I use my discount coupon?

[0139] (6) Tip 6—Can I receive a payment confirmation email for my order?

[0140] Thus, predefined rules may result in the selection of prompts that are deemed most relevant to the intended order 310 .

[0141] In an alternative implementation, a prompt may be selected based on the relevance of the prompt to the value of the order attribute 210 of the expected order 310. For example, a prompt may be selected based on the probability that the prompt matches the value of the order attribute 210 of the expected order 310. A prompt with a higher probability (e.g., a higher confidence score of matching the order attribute 210) may be selected. For example, when the value of the order attribute 210 "order status" of the expected order 310 is "shipping", a prompt may be selected based on the relevance of prompts 1, 2, or 3.

[0142] In step 510, method 500 causes a prompt to be displayed on customer support interface 122 along with the expected order 310. The prompt may be displayed along with and simultaneously with the expected order 310. The prompt may be displayed on customer support interface 122 without requiring any prior input from customer 124, particularly provided in customer support interface 122 (e.g., without requiring customer 124 to ask any questions or provide answers to questions in a chat session, etc.). The prompt may be displayed on customer support interface 122 in the form of a selectable element for selection by customer 124. The selectable element may be a prompt card.

[0143] Even before the customer 124 provides any specific input on the customer support interface 122, displaying prompts (e.g., suggested questions or content) for the customer 124 to select is advantageous because it facilitates communication about the intended order 310 while creating a personalized chat experience for the customer 124. The customer 124 can conveniently select from the prompts to advance the chatbot conversation. At the same time, the prompts enable access to one or more solutions related to the intended order 310, thereby increasing the chatbot's resolution rate for order-related issues while reducing the time required for the customer 124 to resolve the issue via the chatbot.

[0144] Figure 6 A flow chart illustrating a method 600 for implementing the customer support interface 122 according to some embodiments is shown. The method 600 is explained in conjunction with the system 100 and methods 400 and 500 as previously described by previous figures of the present disclosure, and the same or similar processes and elements may not be described in detail.

[0145] In step 602, method 600 receives an assistance request corresponding to customer account 126. Step 602 is similar to step 402 of method 400 and step 502 of method 500, which have been described in detail above. For the sake of brevity, step 602 of method 600 will not be described again.

[0146] In step 604, the method 600 predicts whether the assistance request is related to the order 134 (i.e., whether the assistance request is order-related). For example, the order 134 may be an order 134 previously placed by the customer 124 on the e-commerce platform. Step 604 is similar to step 404 of the method 400 and step 504 of the method 500, which have been described in detail above. For the sake of brevity, step 604 of the method 600 will not be described again.

[0147] If the method 600 predicts that the assistance request is not related to any order in the orders 134 (i.e., the assistance request is non-order related), the method 600 proceeds to step 620. Step 620 is similar to step 420 of the method 400 and step 520 of the method 500, which have been described in detail above. For the sake of brevity, step 620 of the method 600 will not be described again.

[0148] In step 606, the method 600 calculates a score for each order 134 based on the order attributes 210 of each order 134. The score may represent the probability of the order 134 for which an assistance request has been initiated. The order 134 with the highest score may be identified as the expected order 310 (e.g., Figure 3 shown).

[0149] The score of the order 134 can be calculated by assigning a weight to each of the order attributes 210, because some order attributes 210 may be more important than other order attributes. For example, the order attribute 210 "order status" may be assigned the highest weight. The score of each order 134 may be the sum of the "normalized" values ​​of the order attributes 210 multiplied by the corresponding weights.

[0150] At step 608, the method 600 ranks the orders 134 based on their scores and identifies the order 134 with the highest score as the prospective order 310. The orders 134 may be ranked based on their scores (e.g., Figure 3 ). The highest ranked order 134 may be the expected order 310.

[0151] In step 610, method 600 determines a prompt based on order attributes 210 of prospective order 310. Step 610 is similar to step 508 of method 500, which has been described in detail above. For the sake of brevity, step 610 of method 600 will not be described again.

[0152] In step 612, method 600 causes a reminder to be displayed on customer support interface 122 along with the expected order 310. Step 612 is similar to step 510 of method 500, which has been described in detail above. For the sake of brevity, step 612 of method 600 will not be described again.

[0153] In step 614, the method 600 determines other orders 320 having a lower score than the expected order 310. The order 134 having a lower score (i.e., a lower ranking) than the expected order 310 may be other orders 320 (e.g., Figure 3 ). In other words, "other orders" represent orders 134 that have been deemed to have a lower probability / likelihood (compared to the expected order 310) of initiating an assistance request. As described above, the determination of the expected order 310 and other orders 320 can be accomplished by comparing the scores (essentially the values) of the order attributes 210 and comparing between the orders 134 themselves (i.e., comparing at the order level rather than at the macro level).

[0154] In step 616, the method 600 determines prompts for the other orders 320 based on the order attributes 210 of the other orders 320. The determination of the prompts for the other orders 320 in step 616 may be based on the order attributes 210 of the other orders 320 and in a manner similar to the determination of the prompts for the intended order 310, which has been described in detail above (i.e., step 508 of method 500 and step 610 of method 600). In essence, the prompt selected for each of the other orders 320 is the prompt that is considered most relevant to each of the other orders 320. Therefore, because the selection of prompts is order-specific, the prompt selected for the intended order 310 may be different from the prompt selected for the other orders 320. Likewise, there may be situations where the prompt selected for the intended order 310 is the same as the prompt selected for the other orders 320.

[0155] At step 618 , the method 600 causes the reminder to be displayed on the customer support interface 122 along with the other orders 320 .

[0156] The prospective order 310 may be displayed more prominently on the customer support interface 122 than other orders 320. Since the prospective order 310 represents the order 134 with the highest probability / likelihood that the customer 124 has initiated an assistance request, making the prospective order 310 easily distinguishable from other orders 320 with lower probabilities / likelihoods makes the customer support interface 122 more user-friendly. Thus, convenient, proactive order support is provided because the customer 124 can conveniently select the order card representing the prospective order 310 and immediately make a chat inquiry.

[0157] In some embodiments, other orders 320 are not immediately displayed on the first interface of customer support interface 122 along with intended order 310. Instead, other orders 320 are displayed only in response to a customer selection of a selectable link on the first interface of customer support interface 122. Because the screen size of customer device 120 is typically limited, it is advantageous to first display intended order 310 on the first interface of customer support interface 122 (without cluttering the screen display with unnecessary elements), and only display other orders 320 on the second interface or window of customer support interface 122 when customer 124 selects the selectable link. This provides a well-organized screen layout that is both practical and visually appealing to customer 124.

[0158] In an alternative embodiment, if there is only a single or individual order 134 in the customer account 126, the method 600 proceeds to determine the expected order 310 without calculating the score in step 606 or ranking the orders 134 (in this case, the individual orders 134) in step 608. The reason is that since step 604 has already determined that the assistance request is likely order-related, it is logical to infer that the assistance request must be related to the individual order 134. In this scenario, the method 600 can immediately determine that the individual order 134 must be the expected order 310 without further calculation. The method 600 can then bypass steps 606 and 608 and go directly to steps 610 and 612 and cause the expected order 310 to be displayed on the customer support interface 122 along with the prompt. This advantageously saves computing resources while achieving the goal of providing proactive and convenient order support assistance.

[0159] FIG. 7A to FIG. 7C An example of a chatbot session displayed on a customer support interface 122 according to some embodiments is shown. The chatbot session and customer support interface 122 are explained in conjunction with the system 100 and methods 400 and 500 as previously described by the previous figures of the present disclosure, and the same or similar processes and elements may not be described in detail. In particular, FIG. 7A to FIG. 7C It shows that it can be Figure 1 The system 100 described in Figure 4 and Figure 5 1 and 2. An exemplary illustration of a customer support interface 122 presented on a display screen of a customer device 120 during some implementations of methods 400 and 500, respectively, as described in FIG.

[0160] refer to Fig. 7A , the customer support interface 122 may include a chatbot session 702. The chatbot session 702 serves as a real-time conversation between the customer 124 and the chatbot of the chatbot system 148. The chatbot session 702 may display messages from the chatbot to the customer 124. These messages may include an order card 706 representing the expected order 310. For example, the message is presented in a text bubble box. Other types of message containers may also be used to present the message. The chatbot session 702 may also have an active text-based communication area (not shown) for the customer 124 to provide input. The input may also be presented in the form of a message.

[0161] Fig. 7AA customer support interface 122 is shown as displayed on a customer device 120. A customer 124 has logged into a customer account 126 or otherwise provided credentials to authenticate or verify his / her identity. The customer 124 has requested assistance. In response to the request for assistance, the prediction engine 150 has executed the method described above and has caused an order card 706 representing an expected order 310 (order number 123) to be displayed in the chatbot session 702. The order card 706 may be displayed with a welcome message 704 to indicate the start of the chatbot session 702.

[0162] The order card 706 may indicate the order status as “shipped” and also provide a picture of the ordered item (in this case, a handbag). The customer 124 may submit inquiries about a prospective order 310 by typing audio or text input into the chatbot session 702 .

[0163] Still reference Fig. 7A , the prediction engine 150 has executed the method described above, and the chatbot displays prompts along with the expected order 310. The prompts may be displayed in the form of prompt cards 708. The customer 124 may select the prompt card 708. The prompt card 708 may represent any number of prompts that may be considered relevant to the expected order 310, and may be displayed at any specified location in the chatbot session 702.

[0164] The customer 124 may select any of the prompts in the prompt card 708 or submit a query to the chatbot regarding the intended order 310. Figure 7B As shown, chatbot session 702 indicates that the customer who received the prompt "When will I receive my order?" selects 712.

[0165] The customer's 124 input regarding the expected order 310 (such as the customer selection 712) can be used as positive feedback information to train and update one or more parts of the machine learning / neural network model of the prediction engine 150. In this case, the positive feedback can refer to any customer's confirmation assistance request is order-related input. For example, the customer's 124 input can take the form of a query typed by the customer 124 regarding the expected order 310 in the chatbot session 702, a customer selection of a selectable element representing the expected order 310 (i.e., the order card 706), or a customer selection of a selectable element representing a prompt (i.e., the customer selection 712).

[0166] Conversely, the customer's 124 input regarding the expected order 310 may be used as negative feedback information to train and update one or more portions of the machine learning / neural network model of the prediction engine 150. In this case, negative feedback may refer to any customer's 124 confirmation assistance request being non-order related input. For example, the customer's 124 input may take the form of a query typed by the customer 124 in the chatbot session 702 related to an order 134 that is not the expected order 310, a customer selection of a selectable element representing an order 134 that is not the expected order 310, a customer selection of a prompt that is not related to the expected order 310, or the customer 124 terminating the chatbot session 702.

[0167] In some embodiments, customer selections 712 to prompts may be used as positive feedback information to train and update one or more portions of the machine learning / neural network model of the prediction engine 150. In this case, positive feedback may refer to any customer 124 input indicating the relevance of the prompt. For example, customer 124 input may take the form of queries typed by the customer 124 in the chatbot session 704 that are relevant to the prompt and customer selections (i.e., customer selections 712) of selectable elements representing the prompt.

[0168] Conversely, non-selection of a customer selection 712 of a prompt may be used as negative feedback information to train and update one or more portions of the machine learning / neural network model of the prediction engine 150. In this case, negative feedback may refer to any input from the customer 124 that indicates the irrelevance of a prompt. For example, the customer 124 input may take the form of a query typed into the chatbot session 702 that is not relevant to any of the prompts, a customer selection of a selectable element that is not any of the prompts in the prompts, or the customer 124 terminating the chatbot session 702.

[0169] Reference now Figure 7C , the chatbot has processed the customer selection 712 and generated a response 720 that answers the customer's 124 query.

[0170] Thus, as shown, the chatbot session 702 of the customer support interface 122 advantageously displays an order card 706 and a prompt card 708 representing the intended order 310 without any prior input from the customer 124. By automatically suggesting one or more prompts based on the intended order 310, the customer 124 can conveniently select from the prompts to advance the chatbot conversation.

[0171] exist Figure 6 In the exemplary implementation of the system 100 and method 600 described in Fig. 8AA chatbot session 802 of a customer support interface 122 displayed on a customer device 120 is shown. A customer 124 has logged into a customer account 126 or otherwise provided credentials to authenticate or verify his / her identity. The customer 124 has requested assistance. In response to the request for assistance, the prediction engine 150 has executed the method described above and has caused an order card 806 representing an expected order 310 and a prompt card 808 representing a prompt to be displayed in the chatbot session 802. The order card 806 and the prompt card 808 may be displayed in the chatbot session 802 along with a welcome message 804 to indicate the start of the chatbot session 802. The order card 806 may indicate the order status as "shipped" and also provide a picture of the ordered item (in this case, a handbag). The customer 124 may submit queries about the expected order 310 to the chatbot by typing audio or text input into the chatbot session 802.

[0172] The forecasting engine 150 has executed the method described above to determine the other orders 320. Figure 8B As shown, other orders 320 are initially hidden from view. Figure 8B It is illustrated that the customer 124 can access other orders 320 by selecting a selectable link 810. The selectable link 810 can be, for example, a hyperlink. If the customer 124 is interested in viewing the details of the other order 320, the customer 124 can select the selectable link 810 to open or launch the second interface. Alternatively, the selectable link 810 can expand or collapse the details of the other order 320 when it is selected.

[0173] In some embodiments, the customer's 124 selection of the selectable link 810 may be used as positive feedback information to train and update one or more portions of the machine learning model of the prediction engine 150. In this case, the positive feedback may refer to any customer's 124 input indicating an auxiliary request related to the other order 320. For example, the customer's 124 input may take the form of a query typed by the customer 124 in the chatbot session 802 regarding the other order 320, a customer selection of a selectable element representing the other order 320, or a customer selection of a selectable element representing one or more prompts displayed with the other order 320.

[0174] Conversely, non-selection of a selectable link 810 by a customer 124 may be used as negative feedback information to train and update one or more portions of the machine learning model of the prediction engine 150. In this case, negative feedback may refer to any input by the customer 124 indicating that the assistance request is not related to other orders 320. For example, the input by the customer 124 may take the form of a query typed by the customer 124 in the chatbot session 802 that is related to an order 134 that is not other orders 320, a query typed by the customer in the chatbot session that is not related to other orders 320, a customer selection of a selectable element that is not related to other orders 320 (e.g., selection of a live agent button), or termination of the chatbot session 802 by the customer 124.

[0175] In an alternative embodiment, selection of selectable link 810 may trigger further determination of other orders 320. Forecasting engine 150 may perform the further determination.

[0176] refer to Figure 8C , the customer 124 has selected the selectable link 810, and an order card 840 representing the other orders 320 (order number 456 and order number 789) is displayed in the chatbot session 802. The order card 840 may indicate the order status of the other orders 320 and also provide pictures of the ordered items (in this case, the T-shirt and the hat). The customer 124 may submit queries about the other orders 320 to the chatbot by typing audio or text input into the chatbot session 802. FIG. 8A to FIG. 8C As shown, by displaying other orders 320 only when selected by the customer 124, a well-structured screen layout is advantageously provided that is both practical and visually appealing to the customer 124 (without unnecessary elements cluttering the screen display).

[0177] Other orders 320 may also be displayed in the chatbot session 802 (not shown) along with the prompt in a manner similar to that described for the intended order 310. It should also be understood that the order card 840 may be positioned differently and may indicate a ranking differently. For example, the order card 840 may display ten other orders 320, and these other orders 320 may be displayed based on any suitable factor, such as a ranking based on their scores (e.g., Figure 3 ), or a timestamp of customer activity information 130 (e.g., most recent payment information), or a timestamp of order attributes 210 (e.g., most recent update to order status). Chatbot session 802 is intended to be illustrative only, and any number of order cards 840 may be displayed or presented.

[0178] The inventive method described herein is fully intended to be operable on a variety of devices or systems. Any hardware platform suitable for performing the methods described herein is suitable for use with the present invention. Unless otherwise specifically stated and as apparent from the following, it should be understood that throughout this specification, discussions using terms such as "receive", "identify", "initiate", "tag", "transmit", "run", "increment", "determine", "allocate", "approval", "select", "send", "calculate", "analyze", "replace", "generate", "initialize", "output", "train", etc. refer to the actions and processes of a computer system or similar electronic device that manipulates data represented by physical quantities within a computer system and converts the data into other data similarly represented by physical quantities within a computer system or other information storage, transmission or display device.

[0179] In this application, unless otherwise stated, the terms "comprise," "comprising," and their grammatical variations are intended to represent "open" or "inclusive" language such that they include the elements stated but also permit the inclusion of additional, not explicitly stated elements.

[0180] Obviously, various other modifications and adjustments to the present application will become apparent to those skilled in the art after reading the foregoing disclosure without departing from the spirit and scope of the present application, and it is intended that all such modifications and adjustments be included within the scope of the appended claims.

Claims

1. A method for implementing a customer support interface for order support assistance, the method comprising the following steps: receiving an assistance request corresponding to a customer account of a customer, the customer account being associated with at least one order, wherein the customer account includes customer account information, and wherein the at least one order includes order attributes; utilizing a prediction engine implementing a machine learning model to predict whether the assistance request is related to the at least one order, wherein the prediction is based on one or both of the customer account information, the order attributes of the at least one order; In response to predicting that the assistance request is associated with the at least one order, determining, using the prediction engine, an expected order from the at least one order, wherein the expected order represents the at least one order with the highest probability that the assistance request has been initiated; as well as The anticipated order is caused to be displayed on the customer support interface on a customer device.

2. The method according to claim 1, wherein the step of determining the expected order from the at least one order comprises the following steps: calculating a score for each of the at least one order based on the order attributes of the at least one order, the score of the at least one order representing a probability that the at least one order has initiated the assistance request; as well as The at least one order having the highest score is identified as the prospective order.

3. The method according to claim 1 or 2, further comprising the following steps: retrieving order attributes of the prospective order; determining, by the prediction engine, one or more prompts based on the order attributes of the expected order; as well as The one or more reminders are caused to be displayed on the customer support interface along with the anticipated order.

4. The method of claim 3, wherein the step of determining one or more prompts comprises the following steps: Retrieve a collection of prompts from a database; as well as One or more prompts are selected from the prompt set according to predefined rules.

5. The method of claim 3 or 4, wherein the one or more prompts are displayed along with the expected order in a chatbot session on the customer support interface.

6. The method according to any one of claims 2 to 5, further comprising the following steps: ranking the at least one order based on the score of the at least one order; as well as At least one other order is determined from the at least one order using the prediction engine, wherein the at least one other order has a lower score than the expected order.

7. The method according to claim 6, further comprising the steps of: The prospective order is caused to be displayed more prominently on the customer support interface than the at least one other order.

8. The method of claim 7, wherein the step of causing the prospective order to be displayed more prominently on the customer support interface than the at least one other order comprises the steps of: causing the expected order to be displayed on a first interface of the customer support interface; as well as causing a selectable link to be presented on the first interface; Wherein, in response to the customer selection of the selectable link, the at least one other order is displayed on a second interface of the customer support interface.

9. The method of claim 8, wherein based on the ranking of the at least one order, the at least one other order is displayed on the second interface.

10. The method according to any one of claims 5 to 9, further comprising the following steps: receiving input from the customer confirming that the assistance request is associated with the intended order; as well as Updating the machine learning model of the prediction engine.

11. A method according to claim 10, wherein the input from the customer confirming that the auxiliary request is related to the expected order includes any of the following: a query about the expected order typed by the customer in the chatbot session, a customer selection of an optional element representing the expected order, or a customer selection of an optional element representing the one or more prompts.

12. The method according to any one of claims 5 to 9, further comprising the following steps: receiving input from the customer confirming that the auxiliary request is not associated with the intended order; as well as Updating the machine learning model of the prediction engine.

13. A method according to claim 12, wherein the input from the customer confirming that the auxiliary request is not related to the expected order includes any of the following: a query typed by the customer in the chatbot session related to an order that is not the expected order, a query typed by the customer in the chatbot session that is not related to the expected order, a customer selection of an optional element that is not related to the expected order, or the customer terminating the chatbot session.

14. The method according to any one of claims 5 to 13, further comprising the following steps: receiving input from the customer indicating the relevance of the one or more prompts; as well as Updating the machine learning model of the prediction engine.

15. The method of claim 14, wherein said input from said client indicating the relevance of said one or more prompts comprises: A query typed by the customer in the chatbot session related to the one or more prompts, or a customer selection of a selectable element representing the one or more prompts.

16. The method according to any one of claims 5 to 13, further comprising the steps of: receiving input from the customer indicating the non-relevance of the one or more prompts; as well as Updating the machine learning model of the prediction engine.

17. The method of claim 16, wherein the input from the customer indicating the irrelevance of the one or more prompts comprises any of the following: a query typed into the chatbot session that is not relevant to any of the one or more prompts, a customer selection of a selectable element that is not any of the one or more prompts, and termination of the chatbot session by the customer.

18. The method according to any one of claims 1 to 17, wherein the customer account information comprises any one or any combination of the following: customer profile information, customer activity information, historical chat information.

19. A method according to any one of claims 1 to 17, wherein the order attributes include any one or any combination of the following: order status, order date, planned shipping date, actual order shipping date, shipping method, order value, order history, payment method, order payment status, merchant identifier, and product identifier.

20. The method of any one of claims 1 to 19, wherein the request for assistance is initiated via activating a help button on a homepage of an e-commerce platform; and wherein the homepage of the e-commerce platform does not provide information about the at least one order.

21. A system comprising one or more computers and one or more storage devices storing computer-readable instructions, which, when executed by the one or more computers, cause the one or more computers to perform one or more operations according to the method of any one of claims 1 to 20.

22. One or more non-transitory computer storage media storing instructions that, when executed by one or more computers, cause the one or more computers to perform one or more operations according to the method of any one of claims 1 to 20.

Images