Patent Application
20180025726
Embodiments of the present disclosure are directed to the creation and coordination of multiple chatbots for a group conversation with one or more users using natural dialogue systems.
A chatbot is a computer program that can conduct a conversation with a human being. Chatbots are typically used in dialog systems for various practical purposes, including customer service or information acquisition. Chatbots use natural language processing to understand and reply to a user using a dialog. For example:
In the above examples, Dialogue 1 has the system initiative in a question and answer mode, while Dialogue 3 is a natural dialogue system where there are both the user and the system take the initiative.
Chatbots are becoming more widely used by social media software vendors. For example, Facebook recently announced that it would make Messenger, its 900-million-user messaging app, into a full-fledged platform that allows businesses to communicate with users via chatbots. Google is also building a new mobile-messaging service that uses artificial intelligence know-how and chatbot technology to catch up with rivals, such as Facebook. In addition, according to the Wall Street Journal (December, 2015), there are more than 2 billion users of mobile apps.
Thus, social messaging can be a platform. However, the number of mobile app users notwithstanding, people can be reluctant to install apps. Chatbots may provide a new conversational interface for interacting with online services, as chatbots are easier to build and deploy than apps.
However, current chatbot engines do not handle properly when there are multi-chatbots in a group chat with many users, where multi-chatbots refers to more than one chatbot in a group chat together with one or more users. There is also a lack of methods and tools to coordinate and mediate them, where coordinated means that the chatbot interactions are constrained by a set of rules or conventions, or a mediator.
Prior art for the creation of coordinated chatbots includes research directed to the creation of chatbots and research directed to the coordination of chatbots. Although these trends complement each other, there is no solution for creating coordinated multi-chatbots that use natural dialogue systems in the state of the art. Research in the coordination of multi-agent systems area does not address coordinating using natural dialogue, as usually all messages are structured and formalized so the agents can reason and coordinate themselves. With regard to chatbot engines, there is a lack of research directed to building flexible and adaptive coordination rules integrated with natural language in an autonomic or semi-autonomic way.
Exemplary embodiments of the present disclosure provide a system and method to create coordinated multi-chatbots using natural dialog systems. Embodiments of the disclosure use a distributed and decentralized conceptual framework and cookbook for creating a hybrid rule and machine learning-based system where the coordination rules can be manually defined or learned using machine learning algorithms. A framework according to an embodiment can define the entities, relationships and behaviors needed for the creation of coordinated multi-chatbots that react or pro-actively act using natural dialogue. Further embodiments provide one or more chatbots with the role of mediator in a chat group and the mediator chatbot can invite one or more chatbots into the chat group while interacting with users based on users' utterances. A mediator according to an embodiment can also redirect topics based on users' utterances and to enforce that the chatbots only send allowed messages.
According to an embodiment of the disclosure, there is provided a system for coordinating multiple chatbots in a group conversation using natural dialog systems, including a creation unit that enables a user to create a group chat with chatbots; a response unit that allows the user to reply to any utterance extracted from a message received from a member of the group chat; a transmission unit that sends messages to every member of the group chat; a development unit that develops chatbots that understand natural language and interact in a group chat using natural dialogue; a network connection; a first database that stores a knowledge base extracted from all the utterances exchanged by members of the group chat; a second database that stores all interactions between the user and the group chat, and a third database that stores all interaction protocols used by the members of the group chat.
According to a further embodiment of the disclosure, chatbots include a mediator chatbot that invites or removes other chatbots to/from the group chat, based on users' utterances and the interaction protocols.
According to a further embodiment of the disclosure, the knowledge base includes rules that represent knowledge regarding the interaction protocols; the interaction protocols determine types of messages sent by chatbots; and the second database includes all the utterances exchanged between users and chatbots.
According to a further embodiment of the disclosure, the knowledge base includes all intentions of all of the utterances; synonyms for common utterances; all entities of the intentions, and all features associated with the entities; actions associated with each <intention, entity, feature> triple; answers to common intentions; part-of-speech tags associated to the words of all of the utterances; a dependency parsing tree which associates relations between all the words of all of the utterances; and numbers detected in the utterances and their relations to words of the utterances, wherein the numbers include ordinal and cardinal numbers.
According to a further embodiment of the disclosure, a chatbot is any object that can receive, interpret and reply to messages using natural language, learn all interactions between the user and the group chat, including all utterances extracted from messages exchanged between users and chatbots.
According to a further embodiment of the disclosure, developing chatbots includes developing context parsing and saving from utterances exchanged in a group chat, and specifying rules to provide a predefined reply for each recognized utterance.
According to a further embodiment of the disclosure, developing chatbots includes training a classifier for detecting similar intentions from utterances; training a classifier for detecting a speech act of an utterance; training a classifier for detecting an action to be performed by a chatbot in reply to a received utterance; and training a classifier for detecting dialogue errors.
According to a further embodiment of the disclosure, wherein the user reply to any utterance is one of a voice reply, a text reply, or any communication mode that can be translated into a natural language, and the response unit further comprises a sub-unit that performs an action, and a sub-unit that generates an answer depending on the action and knowledge extracted from the utterance being replied to.
According to a another embodiment of the disclosure, there is provided a non-transitory program storage device readable by a computer, tangibly embodying a program of instructions executed by the computer to implement a system for coordinating multiple chatbots in a group conversation using natural dialog systems.
Exemplary embodiments of the disclosure as described herein generally include systems for the creation and coordination of multiple chatbots using natural dialogue systems. Embodiments are described, and illustrated in the drawings, in terms of functional blocks, units or steps. Those skilled in the art will appreciate that these blocks, units or steps can be physically implemented by electronic (or optical) circuits such as logic circuits, discrete components, microprocessors, hard-wired circuits, memory elements, wiring connections, etc., which may be formed using semiconductor-based fabrication techniques or other manufacturing technologies. In the case of the blocks, units or steps being implemented by microprocessors or similar, they may be programmed using software, such as microcode, to perform various functions discussed herein and may optionally be driven by firmware and/or software. Alternatively, each block, unit or step may be implemented by dedicated hardware, or as a combination of dedicated hardware to perform some functions and a processor, such as one or more programmed microprocessors and associated circuitry, to perform other functions. Also, each block, unit or step of the embodiments may be physically separated into two or more interacting and discrete blocks, units or steps without departing from the scope of the disclosure. Further, the blocks, units or steps of the embodiments may be physically combined into more complex blocks, units or steps without departing from the scope of the disclose. Accordingly, while the disclosure is susceptible to various modifications and alternative forms, specific embodiments thereof are shown by way of example in the drawings and will herein be described in detail. It should be understood, however, that there is no intent to limit the disclosure to the particular forms disclosed, but on the contrary, the disclosure is to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the disclosure. In addition, it is understood in advance that although this disclosure includes a detailed description on cloud computing, implementation of the teachings recited herein are not limited to a cloud computing environment. Rather, embodiments of the present invention are capable of being implemented in conjunction with any other type of computing environment now known or later developed.
As described above, a chatbot is a computer program that uses natural language processing to conduct a conversation with a human being. However, although programmers endeavor to develop chatbots that can understand what people say in natural language, people might not answer correctly, and this is still a dialogue, although the other person may not be so happy. So chatbot may not need to reply correctly, depending on the subject domain.
A chatbot can be created by a program known as a chatbot engine. There are currently engines for plugging chabots, creating chatbots, and for creating service chatbots. An engine for plugging a bot needs to be able to integrate a bot into a pre-existing system, and to configure that bot to receive and send messages in a specific way, depending on the API. However, engines for plugging bots lack support for building bot behavior. An engine for creating a bot provides an intention-based setup, in which a knowledge-base or a model trained with machine learning would be used to create the chatbot, and configures the bot to reply to a received intention. An engine for creating a service bot also provides an intention-based setup, and configures the bot to execute an action in response to an intention, and to reply to the received intention. However, no current engine provides a method for coordinating multiple chatbots within a conversation.
Utterances understood by bots include task-oriented utterances and performatives. Task-oriented utterances require the other person, or the chatbot, to perform an action, while performatives are utterances which are used to perform an act instead of merely describing it. Performatives are also known as speech acts. People perform speech acts whenever they offer an apology, greeting, request, complaint, invitation, compliment, or refusal, among others.
Chatbots can communicate according to several dialogue strategies. In system-initiative Q&A system, the chatbot asks all of the questions and the user is only able to answer. In a user-initiative Q&A, the user makes all the requests and the bot only answers/perform an activity. A mixed initiative Q&A system combines the system-initiative and the user-initiative Q&A systems. Examples of a system-initiative Q&A system, a user-initiative Q&A system, and a mixed initiative Q&A system are respectively shown in Dialogues 1, 2, and 3, above, where S is the bot, and U is the user.
According to embodiments of the disclosure, a chatbot may reply using a question-and-answer (Q&A) system, or a chatbot that replies like a Q&A system and performs an action, e.g., “User: turn left, Chabot: Ok, I did”. There may be chatbots that recognize only specific commands, such as “User: @chatbot/turnleft, Chatbot: Ok.”, or there may be a chatbot that uses a dialogue system, or that understands natural language, such as “User: please, would you mind to turn left at the next street a few meters ahead?” Q&A systems are subsets of dialogue systems, which in turn are subsets of natural dialogue systems. Task-oriented systems may be a subset of Q&A systems, or of natural dialogue systems.
Dialogue 3, above depicts examples of conversations using a natural dialogue system, according to embodiments of the disclosure. A “Natural Dialogue System” is a dialogue system that uses natural language and that looks like a dialogue between two or more people rather than a person and a machine. In Dialogue 1, it is easier for a bot to understand the conversation; here, the system in the bot performs all questions while the user gives straightforward short answers. The conversation Dialogue 3, on the other hand, is more natural and more difficult for a machine to understand. The bot sends open-ended questions like “How can I help you?”, or replies with sentiments such as “Oh, that would be a great choice”. Embodiments of the disclosure are directed to natural dialogue systems, such as the conversation on the right.
A cognitive investment advisor is an example of application that can be implemented according to an embodiment of the disclosure. The conversation is composed of a group chat that can contain multiple users and multiple chatbots. This example, in particular, has a mediator that can help users on financial matters, more specifically on investment options. For example, consider the following dialogue:
User: “I have $30,000 USD, where should I invest it?”
Mediator chatbot: “Well, for how long could you keep the money invested?”
User: “Say, for 2 years.”
Mediator chatbot: “All right, then, considering the amount and the time period, how don't you simulate this investment in a savings account?”
User: “Sure! I would love to.”
Mediator chatbot: “Ok, I will invite the savings account to this group.”
<<Savings account join the group>>
Mediator chatbot: “Hi Savings Account, could you please simulate the return of investment of $30,000 in 2 years?”
Savings account chatbot: “Sure, just a minute . . . ”
Savings account chatbot: “Well, at the end, one would have $32,500 USD.”
The above example uses a mixed-initiative dialogue strategy, and a dialogue mediator to provide coordination control. In this example of an application, there are many types of intentions that should be answered: Q&A (question and answer) about definitions, investment options, and bout the current finance indexes, simulation of investments, which is task-oriented and requires computation, and opinions, which can be highly subjective.
According to embodiments, chatbot architectures include rule-based architectures and corpus-based architectures. A rule-based architecture provides a predefined reply for each recognized utterance. Examples of rule-based chatbots include Eliza, a 1966 computer program designed to simulate a therapist or psychoanalyst, and Parry, written in 1972 to simulate a person with paranoid schizophrenia that implemented a simple model of the behavior of a person with paranoid schizophrenia based on concepts, conceptualizations, and beliefs. Corpus-based chatbots, also known as data-driven chatbots, are chatbots that provide answers based on learned models. These learned models can be trained by different technologies, such as classical information retrieval algorithms, by training a neural network, such as a deep neural network, or by training a HMM (Hidden Markov Model) or a POMDP (Partial Observable Markov Decision Process).
After adding or removing other chatbots to/from the chat group, the dialogue management system 63 broadcasts the message to everyone in the group. A reactive chatbot 64a according to an embodiment that receives the message may identify an intention from the message, identify an action from the intention, and then perform the action. The chatbot 64a may then react according to the intention, dialog state, detected entities, and the user's belief, if appropriate, and sends any required or requested reply. The dialogue management system 63 may receive the reply message from chatbot 64a, update the dialogue state and rebroadcast the message. The message is received by a second user's proxy 66, which sends the message to the second user's instance app 67.
Communication between participants of a group chat are governed by multichat interaction protocols, define when or who should reply to a given utterance, what is allowed, what is forbidden, who should answer, when someone should answer, etc. The following examples illustrate the use of multichat protocols:
According to an embodiment of the disclosure, a chatbot can be developed according to a set of steps known as a development cookbook. A Rule-base cookbook according to an embodiment comprises the steps for developing chatbots that is only rule-based, and includes developing chatbots that can receive message objects, and developing context-based parsing and saving, depending on the subject domain. According to a further embodiment of the disclosure, referring to the flowchart of
It is to be understood that embodiments of the present disclosure can be implemented in various forms of hardware, software, firmware, special purpose processes, or a combination thereof. In one embodiment, an embodiment of the present disclosure can be implemented in software as an application program tangible embodied on a computer readable program storage device. The application program can be uploaded to, and executed by, a machine comprising any suitable architecture. Furthermore, it is understood in advance that although this disclosure includes a detailed description on cloud computing, implementation of the teachings recited herein are not limited to a cloud computing environment. Rather, embodiments of the present disclosure are capable of being implemented in conjunction with any other type of computing environment now known or later developed. An automatic troubleshooting system according to an embodiment of the disclosure is also suitable for a cloud implementation.
Cloud computing is a model of service delivery for enabling convenient, on-demand network access to a shared pool of configurable computing resources (e.g. networks, network bandwidth, servers, processing, memory, storage, applications, virtual machines, and services) that can be rapidly provisioned and released with minimal management effort or interaction with a provider of the service. This cloud model may include at least five characteristics, at least three service models, and at least four deployment models.
Characteristics are as follows:
On-demand self-service: a cloud consumer can unilaterally provision computing capabilities, such as server time and network storage, as needed automatically without requiring human interaction with the service's provider.
Broad network access: capabilities are available over a network and accessed through standard mechanisms that promote use by heterogeneous thin or thick client platforms (e.g., mobile phones, laptops, and PDAs).
Resource pooling: the provider's computing resources are pooled to serve multiple consumers using a multi-tenant model, with different physical and virtual resources dynamically assigned and reassigned according to demand. There is a sense of location independence in that the consumer generally has no control or knowledge over the exact location of the provided resources but may be able to specify location at a higher level of abstraction (e.g., country, state, or datacenter).
Rapid elasticity: capabilities can be rapidly and elastically provisioned, in some cases automatically, to quickly scale out and rapidly released to quickly scale in. To the consumer, the capabilities available for provisioning often appear to be unlimited and can be purchased in any quantity at any time.
Measured service: cloud systems automatically control and optimize resource use by leveraging a metering capability at some level of abstraction appropriate to the type of service (e.g., storage, processing, bandwidth, and active user accounts). Resource usage can be monitored, controlled, and reported providing transparency for both the provider and consumer of the utilized service.
Service Models are as follows:
Software as a Service (SaaS): the capability provided to the consumer is to use the provider's applications running on a cloud infrastructure. The applications are accessible from various client devices through a thin client interface such as a web browser (e.g., web-based email). The consumer does not manage or control the underlying cloud infrastructure including network, servers, operating systems, storage, or even individual application capabilities, with the possible exception of limited user-specific application configuration settings.
Platform as a Service (PaaS): the capability provided to the consumer is to deploy onto the cloud infrastructure consumer-created or acquired applications created using programming languages and tools supported by the provider. The consumer does not manage or control the underlying cloud infrastructure including networks, servers, operating systems, or storage, but has control over the deployed applications and possibly application hosting environment configurations.
Infrastructure as a Service (IaaS): the capability provided to the consumer is to provision processing, storage, networks, and other fundamental computing resources where the consumer is able to deploy and run arbitrary software, which can include operating systems and applications. The consumer does not manage or control the underlying cloud infrastructure but has control over operating systems, storage, deployed applications, and possibly limited control of select networking components (e.g., host firewalls).
Deployment Models are as follows:
Private cloud: the cloud infrastructure is operated solely for an organization. It may be managed by the organization or a third party and may exist on-premises or off-premises.
Community cloud: the cloud infrastructure is shared by several organizations and supports a specific community that has shared concerns (e.g., mission, security requirements, policy, and compliance considerations). It may be managed by the organizations or a third party and may exist on-premises or off-premises.
Public cloud: the cloud infrastructure is made available to the general public or a large industry group and is owned by an organization selling cloud services.
Hybrid cloud: the cloud infrastructure is a composition of two or more clouds (private, community, or public) that remain unique entities but are bound together by standardized or proprietary technology that enables data and application portability (e.g., cloud bursting for loadbalancing between clouds).
A cloud computing environment is service oriented with a focus on statelessness, low coupling, modularity, and semantic interoperability. At the heart of cloud computing is an infrastructure comprising a network of interconnected nodes.
Referring now to
In cloud computing node 910 there is a computer system/server 912, which is operational with numerous other general purpose or special purpose computing system environments or configurations. Examples of well-known computing systems, environments, and/or configurations that may be suitable for use with computer system/server 912 include, but are not limited to, personal computer systems, server computer systems, thin clients, thick clients, handheld or laptop devices, multiprocessor systems, microprocessor-based systems, set top boxes, programmable consumer electronics, network PCs, minicomputer systems, mainframe computer systems, and distributed cloud computing environments that include any of the above systems or devices, and the like.
Computer system/server 912 may be described in the general context of computer system executable instructions, such as program modules, being executed by a computer system. Generally, program modules may include routines, programs, objects, components, logic, data structures, and so on that perform particular tasks or implement particular abstract data types. Computer system/server 912 may be practiced in distributed cloud computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed cloud computing environment, program modules may be located in both local and remote computer system storage media including memory storage devices.
As shown in
Bus 918 represents one or more of any of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, and a processor or local bus using any of a variety of bus architectures. By way of example, and not limitation, such architectures include Industry Standard Architecture (ISA) bus, Micro Channel Architecture (MCA) bus, Enhanced ISA (EISA) bus, Video Electronics Standards Association (VESA) local bus, and Peripheral Component Interconnect (PCI) bus.
Computer system/server 912 typically includes a variety of computer system readable media. Such media may be any available media that is accessible by computer system/server 912, and it includes both volatile and non-volatile media, removable and non-removable media.
System memory 928 can include computer system readable media in the form of volatile memory, such as random access memory (RAM) 930 and/or cache memory 932. Computer system/server 912 may further include other removable/non-removable, volatile/non-volatile computer system storage media. By way of example only, storage system 934 can be provided for reading from and writing to a non-removable, non-volatile magnetic media (not shown and typically called a “hard drive”). Although not shown, a magnetic disk drive for reading from and writing to a removable, non-volatile magnetic disk (e.g., a “floppy disk”), and an optical disk drive for reading from or writing to a removable, non-volatile optical disk such as a CD-ROM, DVD-ROM or other optical media can be provided. In such instances, each can be connected to bus 918 by one or more data media interfaces. As will be further depicted and described below, memory 928 may include at least one program product having a set (e.g., at least one) of program modules that are configured to carry out the functions of embodiments of the disclosure.
Program/utility 940, having a set (at least one) of program modules 942, may be stored in memory 928 by way of example, and not limitation, as well as an operating system, one or more application programs, other program modules, and program data. Each of the operating system, one or more application programs, other program modules, and program data or some combination thereof, may include an implementation of a networking environment. Program modules 942 generally carry out the functions and/or methodologies of embodiments of the disclosure as described herein.
Computer system/server 912 may also communicate with one or more external devices 914 such as a keyboard, a pointing device, a display 924, etc.; one or more devices that enable a user to interact with computer system/server 912; and/or any devices (e.g., network card, modem, etc.) that enable computer system/server 912 to communicate with one or more other computing devices. Such communication can occur via Input/Output (I/O) interfaces 922. Still yet, computer system/server 912 can communicate with one or more networks such as a local area network (LAN), a general wide area network (WAN), and/or a public network (e.g., the Internet) via network adapter 920. As depicted, network adapter 920 communicates with the other components of computer system/server 912 via bus 918. It should be understood that although not shown, other hardware and/or software components could be used in conjunction with computer system/server 912. Examples, include, but are not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data archival storage systems, etc.
Referring now to
While embodiments of the present disclosure has been described in detail with reference to exemplary embodiments, those skilled in the art will appreciate that various modifications and substitutions can be made thereto without departing from the spirit and scope of the disclosure as set forth in the appended claims.
