Abstract: Automatic measurement of semantic textual similarity of conversations, by: receiving two conversation texts, each comprising a sequence of utterances; encoding each of the sequences of utterances into a corresponding sequence of semantic representations; computing a minimal edit distance between the sequences of semantic representations; and, based on the computation of the minimal edit distance, performing at least one of: quantifying a semantic similarity between the two conversation texts, and outputting an alignment of the two sequences of utterances with each other.

Abstract: Balancing an imbalanced dataset, by: Receiving a balancing policy and the imbalanced dataset. Performing initial adjustment of the imbalanced dataset to comply with the balancing policy, by: oversampling one or more underrepresented classes, and, if one or more of the classes are overrepresented, undersampling them. Operating a generative machine learning model to generate samples for the one or more underrepresented classes, based on the initially-adjusted dataset. Operating a machine learning classification model to label the generated samples with class labels corresponding to the one or more underrepresented classes. Selecting some of the generated samples which, according to the labeling, have a relatively high probability of preserving their class labels.

Abstract: A method, computer system, and a computer program product for automated agent intent detection enhancement are provided. A first message from a first user is received. The first message is generated during a first conversation between the first user and a first automated agent. A computer produces a second message that includes a same request as the first message but a different language modality than the first message. The second message and the first message are combined to form a combined message. The combined message is input into the first automated agent such that the first automated agent produces an intent classification for the first message.

Abstract: Methods, systems, and computer program products for implementing automated communication exchange programs for attended robotic process automation are provided herein. A computer-implemented method includes invoking, during a user communication associated with an attended robotic process automation context, at least one automated communication exchange program in response to at least one user input; determining, using the at least one automated communication exchange program, information directed to the at least one user input; carrying out, using the at least one automated communication exchange program, at least a portion of the user communication subsequent to determining the information directed to the at least one user input; and performing one or more automated actions in connection with automatically carrying out the at least a portion of the user communication.

Abstract: Methods, systems, and computer program products for implementing automated communication exchange programs for attended robotic process automation are provided herein. A computer-implemented method includes invoking, during a user communication associated with an attended robotic process automation context, at least one automated communication exchange program in response to at least one user input; determining, using the at least one automated communication exchange program, information directed to the at least one user input; carrying out, using the at least one automated communication exchange program, at least a portion of the user communication subsequent to determining the information directed to the at least one user input; and performing one or more automated actions in connection with automatically carrying out the at least a portion of the user communication.

Abstract: An example system includes a processor to receive a bot design and escalation logs associated with a chat bot configured based on the bot design. The processor can compute a similarity score between each of a number of bot response nodes in the bot design and the escalation logs. The processor can generate a sorted list of the bot response nodes in the bot design based on the similarity scores.

Abstract: Automatic measurement of semantic textual similarity of conversations, by: receiving two conversation texts, each comprising a sequence of utterances; encoding each of the sequences of utterances into a corresponding sequence of semantic representations; computing a minimal edit distance between the sequences of semantic representations; and, based on the computation of the minimal edit distance, performing at least one of: quantifying a semantic similarity between the two conversation texts, and outputting an alignment of the two sequences of utterances with each other.

Abstract: The computer receives a group of conversation data associated with the escalation node, identifies agent responses in the conversation data, and clusters them into agent response types. The computer identifies dialog state feature value sets for the conversations. The computer identifies feature value set associations with response types, and generates, Boolean expressions representing the feature value sets associated with each of the response types. The computer makes a recommendation to add to at least one child node for the escalation node, with the child node corresponding to one of the response types. The child node has, as an entry condition, the Boolean expression for the response type to which the child node corresponds. The child node has as an action, which according to some aspects, provides a response representative of the cluster of agent responses for the response type to which the child node corresponds.

Abstract: Embodiments may determine user intent in conversations with dialogue systems so as to improve the quality of such conversations and to reduce the number of failed conversations. For example, a method may comprise receiving, at a dialogue system, a first text utterance from a user, generating a plurality of second text utterances at the dialogue system in response to the received text utterance, generating a third text utterance based on each generated second text utterance using a trained deep neural network model, generating a score indicating a quality of each conversation, wherein each conversation includes the first text utterance, one of the second text utterances, and the third text utterance based on the one of the second text utterances, and outputting to the user the second text utterance included in the conversation having the highest quality score.

Abstract: Embodiments of the present systems and methods may provide techniques for augmenting textual data that may be used for textual classification tasks. Embodiments of such techniques may provide the capability to synthesize labeled data to improve text classification tasks. Embodiments may be specifically useful when only a small amount of data is available, and provide improved performance in such cases. For example, in an embodiment, a method implemented in a computer system may comprise a processor, memory accessible by the processor, and computer program instructions stored in the memory and executable by the processor, and the method may comprise fine-tuning a language model using a training dataset, synthesizing a plurality of samples using the fine-tuned language model, filtering the plurality of synthesized samples, and generating an augmented training dataset comprising the training dataset and the filtered plurality of synthesized sentences.

Abstract: Balancing an imbalanced dataset, by: Receiving a balancing policy and the imbalanced dataset. Performing initial adjustment of the imbalanced dataset to comply with the balancing policy, by: oversampling one or more underrepresented classes, and, if one or more of the classes are overrepresented, undersampling them. Operating a generative machine learning model to generate samples for the one or more underrepresented classes, based on the initially-adjusted dataset. Operating a machine learning classification model to label the generated samples with class labels corresponding to the one or more underrepresented classes. Selecting some of the generated samples which, according to the labeling, have a relatively high probability of preserving their class labels.

Abstract: Embodiments may determine user intent in conversations with dialogue systems so as to improve the quality of such conversations and to reduce the number of failed conversations. For example, a method may comprise receiving, at a dialogue system, a first text utterance from a user, generating a plurality of second text utterances at the dialogue system in response to the received text utterance, generating a third text utterance based on each generated second text utterance using a trained deep neural network model, generating a score indicating a quality of each conversation, wherein each conversation includes the first text utterance, one of the second text utterances, and the third text utterance based on the one of the second text utterances, and outputting to the user the second text utterance included in the conversation having the highest quality score.

Abstract: An example system includes a processor to receive data for a multi-objective task. The processor is to also perform the multi-objective task on the received data via a trained primal network. The primal network and a dual network are trained for a multi-objective task using a Lagrangian loss function representing a number of objectives. The primal network is trained to minimize the Lagrangian loss function and the dual network is trained to maximize the Lagrangian loss function.

Abstract: The computer receives a group of conversation data associated with the escalation node, identifies agent responses in the conversation data, and clusters them into agent response types. The computer identifies dialog state feature value sets for the conversations. The computer identifies feature value set associations with response types, and generates, Boolean expressions representing the feature value sets associated with each of the response types. The computer makes a recommendation to add to at least one child node for the escalation node, with the child node corresponding to one of the response types. The child node has, as an entry condition, the Boolean expression for the response type to which the child node corresponds. The child node has as an action, which according to some aspects, provides a response representative of the cluster of agent responses for the response type to which the child node corresponds.

Abstract: An intermediary server receives a resource request from a client computer for a requested resource that has a desired resource ability to optimize the client computer. The intermediary server identifies an alternative resource that has the desired resource ability to optimize the client computer, and determines that an ability of the alternative resource to provide the desired resource ability exceeds an ability of the requested resource to provide the desired resource ability. In response to determining that the ability of the alternative resource to provide the desired resource ability exceeds the ability of the requested resource to provide the desired resource ability, the intermediary server instructs a resource server to send the alternative resource instead of the requested resource to the client computer.

Abstract: Embodiments of the present systems and methods may provide techniques for augmenting textual data that may be used for textual classification tasks. Embodiments of such techniques may provide the capability to synthesize labeled data to improve text classification tasks. Embodiments may be specifically useful when only a small amount of data is available, and provide improved performance in such cases. For example, in an embodiment, a method implemented in a computer system may comprise a processor, memory accessible by the processor, and computer program instructions stored in the memory and executable by the processor, and the method may comprise fine-tuning a language model using a training dataset, synthesizing a plurality of samples using the fine-tuned language model, filtering the plurality of synthesized samples, and generating an augmented training dataset comprising the training dataset and the filtered plurality of synthesized sentences.

Abstract: An example system includes a processor to receive a prefix of conversation and a text input. The processor is to also generate a completed response based on the prefix of conversation and the text input via a trained primal network. The primal network is trained to minimize a Lagrangian loss function representing a number of objectives and a dual network is trained to maximize the Lagrangian loss function.

Abstract: A method comprising operating a computerized chatbot to: calculate first and second scores representing a relevance of input received from a user to functionalities provided by respective first and second modules, respectively, of the chatbot; associate the first and second modules with respective first and second names; introducing the modules to the user using their associated names; selecting a module to interact with the user based on at least one of: a name mentioned by the user and a score and switching between the first and second modules based on the names.

Abstract: An intermediary server receives a resource request from a client computer for a requested resource that has a desired resource ability to optimize the client computer. The intermediary server identifies an alternative resource that has the desired resource ability to optimize the client computer, and determines that an ability of the alternative resource to provide the desired resource ability exceeds an ability of the requested resource to provide the desired resource ability. In response to determining that the ability of the alternative resource to provide the desired resource ability exceeds the ability of the requested resource to provide the desired resource ability, the intermediary server instructs a resource server to send the alternative resource instead of the requested resource to the client computer.

Abstract: An example system includes a processor to receive data for a multi-objective task. The processor is to also perform the multi-objective task on the received data via a trained primal network. The primal network and a dual network are trained for a multi-objective task using a Lagrangian loss function representing a number of objectives. The primal network is trained to minimize the Lagrangian loss function and the dual network is trained to maximize the Lagrangian loss function.