Abstract: A system and method for predicting items for next item recommendation based on a sequence of user actions. The method may include receiving a sequence of user actions for a user, generating a respective embeddings vector representative of each of the user actions, determining respective pairwise similarities among the embeddings vectors, generating a respective attention embeddings vector for each of the user actions according to the similarities, applying a predictive model to the attention embeddings vectors to determine a predicted next user action, and outputting the predicted next user action to the user in response to receiving the sequence. The sequence of user actions can be a first sequence of user actions based on a first parameter and a second sequence of user actions based on a second parameter. The method may include training a machine learning model and deploying the trained model to make the predictions.

Abstract: A method for predicting a next user selection in an electronic user interface includes receiving, from a user, a sequence of selections of documents and generating, for each document in the sequence, a respective attribute vector. The attribute vector includes a numerical attribute vector portion representative of numerical attributes of the document, a category attribute vector portion representative of category information of the document, a text content vector portion representative of text content of the document, and an image content vector portion representative of an image in the document. The method further includes inputting the attribute vectors of the sequence into a machine learning model, and outputting, to the user, in response to the sequence of selections, a predicted next document selection according to an output of the machine learning model.

Abstract: A method for predicting a next user selection in an electronic user interface, such as a website, includes training a machine learning model according to a training data set, the training data set including a plurality of token sets, each token set representative of a respective document accessible through the interface, each token set including a plurality of words, each word describing a characteristic of the document, to create a trained model. The method further includes receiving, from a user, a sequence of selections of documents, inputting the sequence of selections to the trained model, and outputting to the user, in response to the sequence of selections, a predicted next document selection according to an output of the trained model.

Abstract: Example implementations involve systems and methods for predicting failures and remaining useful life (RUL) for equipment, which can involve, for data received from the equipment comprising fault events, conducting feature extraction on the data to generate sequences of event features based on the fault events; applying deep learning modeling to the sequences of event features to generate a model configured to predict the failures and the RUL for the equipment based on event features extracted from data of the equipment; and executing optimization on the model.

Abstract: Example implementations involve a framework for knowledge base construction of components and problems in short texts. The framework extracts domain-specific components and problems from textual corpora such as service manuals, repair records, and public Q/A forums using: 1) domain-specific syntactic rules leveraging part of speech tagging (POS), and 2) a neural attention-based seq2seq model which tags raw sentences end-to-end identifying components and their associated problems. Once acquired, this knowledge can be leveraged to accelerate the development and deployment of intelligent conversational assistants for various industrial AI scenarios (e.g., repair recommendation, operations, and so on) through better understanding of user utterances. The example implementations give better tagging accuracy on various datasets outperforming well known off-the-shelf systems.

Abstract: Example implementations involve a framework for knowledge base construction of components and problems in short texts. The framework extracts domain-specific components and problems from textual corpora such as service manuals, repair records, and public Q/A forums using: 1) domain-specific syntactic rules leveraging part of speech tagging (POS), and 2) a neural attention-based seq2seq model which tags raw sentences end-to-end identifying components and their associated problems. Once acquired, this knowledge can be leveraged to accelerate the development and deployment of intelligent conversational assistants for various industrial AI scenarios (e.g., repair recommendation, operations, and so on) through better understanding of user utterances. The example implementations give better tagging accuracy on various datasets outperforming well known off-the-shelf systems.

Abstract: Mined semantic analysis techniques (MSA) include generating a first subset of concepts, from a NL corpus, that are latently associated with an NL candidate term based on (i) a second subset of concepts from the corpus that are explicitly or implicitly associated with the candidate term and (ii) a set of concept association rules. The concept association rules are mined from a transaction dictionary constructed from the corpus and defining discovered latent associations between corpus concepts. A concept space of the candidate term includes at least portions of both the first and second subset of concepts, and includes indications of relationships between latently-associated concepts and the explicitly/implicitly-associated concepts from which the latently-associated concepts were derived. Measures of relatedness between candidate terms are deterministically determined based on their respective concept spaces.

Abstract: Mined semantic analysis techniques (MSA) include generating a first subset of concepts, from a NL corpus, that are latently associated with an NL candidate term based on (i) a second subset of concepts from the corpus that are explicitly or implicitly associated with the candidate term and (ii) a set of concept association rules. The concept association rules are mined from a transaction dictionary constructed from the corpus and defining discovered latent associations between corpus concepts. A concept space of the candidate term includes at least portions of both the first and second subset of concepts, and includes indications of relationships between latently-associated concepts and the explicitly/implicitly-associated concepts from which the latently-associated concepts were derived. Measures of relatedness between candidate terms are deterministically determined based on their respective concept spaces.