International Business Machines Corporation

Amazon Technology, Inc.

Haoyu Wang

Ming Tan

Saloni Potdar

Yang Yu

Dakuo Wang

Chuang Gan

Ladislav Kunc

Mo Yu

Navneet N Rao

Haode Qi

Shiyu Chang

Ravi Nair

Ashish Kumar Agrawal

Robert Leslie Yates

Inkit Padhi

Ruijian Wang

James J Logan

Albert J Morello

Panos Karagiannis

Yupeng Gao

Ladislav Kune

This disclosure proposes systems and methods for entity resolution using speech recognition data. The system can receive audio data representing an utterance and perform automatic speech recognition (ASR) processing on the audio data to generate at least a first ASR data and a second ASR data. The system can perform natural language understanding (NLU) processing on the first ASR data to determine intent data and an indication of an entity. The system can determine a first portion of the first ASR data that corresponds to the indication of the entity. The system can determine a second portion of the second ASR data that corresponds to the indication of the entity without performing NLU on the second ASR data. The system can perform entity resolution (ER) on the second portion to identify a first entity.

Entity resolution using speech recognition data

An artificial intelligence (AI) interaction method, system, and computer program product include selecting an artificial intelligence model to respond to a query to generating a response to the query using the selected artificial intelligence model, and receiving the response to the query from the selected artificial intelligence model.

Natural language interaction with automated machine learning systems

A deep learning module classifies messages received from a plurality of entities into one or more conversation threads. In response to receiving a subsequent message, the deep learning module determines which of the one or more conversation threads and a new conversation thread is contextually a best fit for the subsequent message. The subsequent message is added to the determined conversation thread.

Determination of conversation threads in a message channel based on conversational flow and semantic similarity of messages

A computer-implemented method includes using an embedding network to generate prototypical vectors. Each prototypical vector is based on a corresponding label associated with a first domain. The computer-implemented method also includes using the embedding network to generate an in-domain test vector based on at least one data sample from a particular label associated with the first domain and using the embedding network to generate an out-of-domain test vector based on at least one other data sample associated with a different domain. The computer-implemented method also includes comparing the prototypical vectors to the in-domain test vector to generate in-domain comparison values and comparing the prototypical vectors to the out-of-domain test vector to generate out-of-domain comparison values. The computer-implemented method also includes modifying, based on the in-domain comparison values and the out-of-domain comparison values, one or more parameters of the embedding network.

Out-of-domain encoder training

Domain specific model compression by providing a weighting parameter for a candidate operation of a neural network, applying the weighting parameter to an output vector of the candidate operation, performing a regularization of the weighting parameter output vector combination, compressing the neural network model according to the results of the regularization, and providing the neural network model after compression.

Domain specific model compression

A method includes determining, based on an input data sample, a set of probabilities. Each probability of the set of probabilities is associated with a respective label of a set of labels. A particular probability associated with a particular label indicates an estimated likelihood that the input data sample is associated with the particular label. The method includes modifying the set of probabilities based on a set of adjustment factors to generate a modified set of probabilities. The set of adjustment factors is based on a first relative frequency distribution and a second relative frequency distribution. The first relative frequency distribution indicates for each label of the set of labels, a frequency of occurrence of the label among training data. The second relative frequency distribution indicates for each label of the set of labels, a frequency of occurrence of the label among post-training data provided to the trained classifier.

Intent classification distribution calibration

A mechanism is provided to implement suggestion of new entity types with discriminative importance analysis. The mechanism obtains a list of predefined intents from a chatbot designer. The mechanism receives an input sentence having a target intent within the list of predefined intents. The mechanism performs intent-specific importance analysis on the input sentence to generate an importance score for each token in the input sentence. The mechanism ranks the tokens in the input sentence by importance score and outputs a token with a highest importance score as a candidate entity type.

Suggestion of new entity types with discriminative term importance analysis

A method of text classification includes generating a text embedding vector representing a text sample and applying weights of a regression layer to the text embedding vector to generate a first data model output vector. The method also includes generating a plurality of prototype embedding vectors associated with a respective classification labels and comparing the plurality of prototype embedding vectors to the text embedding vector to generate a second data model output vector. The method further includes assigning a particular classification label to the text sample based on the first data model output vector, the second data model output vector, and one or more weighting values.

Hybrid model for short text classification with imbalanced data

A mechanism is provided for implementing an intent segmentation mechanism that segments intent boundaries for multi-intent utterances in a conversational agent. For each term of a set of terms in the utterance from a real-time chat session, a set of adversarial utterances is generated for the utterance. An influence of changing each term is determined so as to identify a term importance value. Utilizing the term importance value, one or more of a change in ranking of the intent of the utterance or a change in confidence with regard to the intent of the utterance is identified. An entropy-based segmentation of the utterance into a plurality of candidate partitions is performed. An associated intent and entropy value are then assigned. Based on a segment with minimum entropy, a call associated with the real-time chat session is directed to an operation associated with an intent of the segment with minimum entropy.

Intent boundary segmentation for multi-intent utterances

Provided is a method, system, and computer program product for context-dependent spellchecking. The method comprises receiving context data to be used in spell checking. The method further comprises receiving a user input. The method further comprises identifying an out-of-vocabulary (OOV) word in the user input. An initial suggestion pool of candidate words is identified based, at least in part, on the context data. The method then comprises using a noisy channel approach to evaluate a probability that one or more of the candidate words of the initial suggestion pool is an intended word and should be used as a candidate for replacement of the OOV word. The method further comprises selecting one or more candidate words for replacement of the OOV word. The method further comprises outputting the one or more candidates.

Artificial intelligence based context dependent spellchecking

A computer system identifies threads in a communication session. A feature vector is generated for a message in a communication session, wherein the feature vector includes elements for features and contextual information of the message. The message feature vector and feature vectors for a plurality of threads are processed using machine learning models each associated with a corresponding thread to determine a set of probability values for classifying the message into at least one thread, wherein the threads include one or more pre-existing threads and a new thread. A classification of the message into at least one of the threads is indicated based on the set of probability values. Classification of one or more prior messages is adjusted based on the message's classification. Embodiments of the present invention further include a method and program product for identifying threads in a communication session in substantially the same manner described above.

Context-aware conversation thread detection for communication sessions

A computing device receives a natural language input from a user. The computing device routes the natural language input from an active domain node of multiple domain nodes of a multi-domain context-based hierarchy to a leaf node of the domain nodes by selecting a parent domain node in the hierarchy until an off-topic classifier labels the natural language input as in-domain and then selecting a subdomain node in the hierarchy until an in-domain classifier labels the natural language input with a classification label, each of the plurality of domain nodes comprising a respective off-topic classifier and a respective in-domain classifier trained for a respective domain node. The computing device outputs the classification label determined by the leaf node.

Routing text classifications within a cross-domain conversational service

Techniques facilitating detection of conversation threads in unstructured channels are provided. A system can comprise a memory that stores computer executable components and a processor that executes the computer executable components stored in the memory. The computer executable components can comprise an extraction component that employs a model to detect conversation messages based on a defined confidence level and assigns the conversation messages to respective conversation thread categories. The computer executable components also can comprise a model component that trains the model on conversation messages that comprise respective text data, wherein the model is trained to detect the respective text data to the defined confidence level.

Facilitating detection of conversation threads in a messaging channel

Systems and methods provide for automated messaging summarization and ranking. The systems and methods may use an integrated machine learning model to perform thread detection, thread summarization, and summarization ranking. The messages may be received from a team chat application, organized, summarized and ranked by the machine learning model, and the results may be returned to the team chat application. In some cases, the ranking may be different for different users of the team chat application.

Summarization of group chat threads

A mechanism is provided to implement a text classifier training augmentation mechanism for incorporating unlabeled data into the generation of a text classifier. For each term of a plurality of terms in each document of a plurality of documents in a set of unlabeled data, a term frequency value is determined. The term is normalized by dividing the term frequency value by a total number of terms in the document. An inverse document frequency (idf) value is determined for each term based on the term frequency value. A subset of terms is filtered from the plurality of terms based the determined idf values. The idf values for the remaining terms are transformed into feature weights. Terms from a set of labeled data are re-weighted based on the feature weights determined from the set of unlabeled data. The text classifier is then generated using the re-weighted labeled data.

Growing community of inventors

Somerville, MA, United States of America

Haoyu Wang