International Business Machines Corporation

Petar Ristoski

Daniel Gruhl

Alfredo Alba

Linda Ha Kato

Steven R Welch

Chad Eric DeLuca

Anna Lisa Gentile

Ismini Lourentzou

Chris Kau

Chandrasekhar Narayan

Nathaniel H Park

Dmitry Zubarev

James L Hedrick

Daniel Sanders

Richard Thomas Goodwin

Anni Rosa Coden

Richard B Segal

Robin Lougee

Kimberly C Lang

Jing Fu

Christian Harris

Tenzin Yeshi

Candidate material for polymerization can be received. One or more desired features in the candidate material can be identified. A machine learning model can be trained to generate a new material having one or more of the desired features. Permissively, the candidate material can be determined from running a machine learning classification model that ranks a plurality of material as candidates. Permissively, the generated new material can be input to the machine learning classification model, for the machine learning classification model to include in ranking the plurality of material as candidates.

Expert-in-the-loop AI for materials generation

A method, system, and computer program product for representational learning of product formulas are provided. The method accesses a set of product formulas. Each product formula includes a set of ingredient tuples. A directed graph is generated from the set of product formulas. The directed graph including a node for each ingredient of the sets of ingredient tuples of the set of formulas. The method generates a weighted graph from the directed graph. The weighted graph has a weight assigned to each edge in the directed graph. The method generates an embedding model based on the directed graph. A set of embeddings is determined for the weighted graph where each node is represented with low-dimensional numerical vectors.

Method, system, and computer program product for representational machine learning for product formulation

A set of material candidates expected to yield materials with target properties can be generated. A subject matter expert's decision indicating accepted and rejected material candidates from the set of material candidates can be received. Based on the subject matter expert's input, a machine learning model can be trained to replicate the subject matter expert's decision.

Expert-in-the-loop AI for materials discovery

An embodiment includes generating a query prompting a user to select from among a plurality of response options related to a first query set of objects. The embodiment also receives, responsive to the query, user input representative of a selected response option selected by the user from among the plurality of response options. The embodiment also calculates a plurality of weight values for respective ones of a plurality of similarity matrices based on the selected response option, where the plurality of similarity matrices include respective different sets of similarity values, each set of similarity values comprising similarity values representative of similarities of respective pairs of the plurality of objects. The embodiment stores a designated similarity matrix that is selected from among the plurality of similarity matrices based at least in part on a weight value from among the plurality of weight values assigned to the designated similarity matrix.

Identifying similarity matrix for derived perceptions

A computer-implemented method according to one embodiment includes receiving snapshot data for a node within a data center; determining one or more candidate labels for one or more software applications running on the node, utilizing the snapshot data; implementing a validation of the one or more candidate labels to determine one or more validated labels; and training a machine learning model, utilizing the one or more validated labels and the snapshot data.

Labeling software applications running on nodes of a data center

A method for ranking relevance of documents includes using a set of queries, searching a corpus of documents for a set of candidate documents with information relevant to the set of queries. The method further includes ranking the set of candidate documents by a deep learning processing system according to relevance to respective ones of the set of queries. The method additionally includes responsive to user input, revising the ranked set of candidate documents to produce a revised ranked set of candidate documents. The method further includes using the revised ranked set of candidate documents to retrain the deep learning processing system. The method still further includes performing a categorization of the set of candidate documents by the retrained deep learning processing system.

Cognitive horizon surveillance

Systems, computer-implemented methods, and computer program products to transform a lexicon that describes an information asset are provided. According to an embodiment, 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 a term validation component that can determine from a subject matter expert, a validated term that can indicate validation of a candidate term that describes an information asset. The computer executable components can further comprise a lexicon transforming component that, based on the validated term, can transform a lexicon that describes the information asset, by incorporating the validated term into the lexicon.

Transforming a lexicon that describes an information asset

One embodiment provides a method that includes determining candidate ontologies for alignment from multiple available knowledge bases. An initial target ontology is selected from the candidate ontologies and correcting the initial selected ontology with received refinement input. Concepts in the selected initial ontology are aligned with concepts of the target ontology using a deep learning hierarchical classification with received review input. A user is assisted to build, change and grow the selected initial ontology exploiting both the target ontology and new facts extracted from unstructured data.

User-centric ontology population with user refinement

Embodiments relate to a system, program product, and method for information extraction and annotation of a data set. Neural models are utilized to automatically attach machine annotations to data elements within an unlabeled data set. The attached machine annotations are evaluated and a score is attached to the annotations. The score reflects a confidence of correctness of the annotations. A labeled data set is iteratively expanded with selectively evaluated annotations based on the attached score. The labeled data set is applied to an unexplored corpus to identify matching corpus data to populated instances of the labeled data set.

Collaborative information extraction

A verification and authorization method, system, and computer program product include verifying, via a receiving device that receives a verification sound packet, an identity of a trusted caller via the verification sound packet, the verification sound packet including an asymmetrically encrypted payload sent by the trusted caller.

Authentication and authorization via vocal track frequency channel

In an approach to corpus expansion using lexical signatures, one or more computer processors retrieve a donor corpus of text, wherein the donor corpus includes a plurality of documents. One or more computer processors generate a document signature for each of the plurality of documents in the donor corpus. One or more computer processors retrieve a target corpus of text for expansion. One or more computer processors generate a corpus signature for the target corpus. One or more computer processors compare each document signature to the corpus signature. Based on the comparison, one or more computer processors determine a similarity score for each document signature. One or more computer processors rank the plurality of documents by the similarity score. One or more computer processors add one or more top-ranked documents of the plurality of documents to the target corpus.

Corpus expansion using lexical signatures

Embodiments relate to a system, program product, and method for dictionary membership management directed at identifying ambiguity in semantic resources. A dictionary of seed terms is applied to a text corpus and matching items in the corpus are identified. The linguistic properties for each matching item are characterized and a context pattern of each matching item is constructed. Each context pattern is applied to the dictionary and matching content between the seed terms and the context pattern is identified and quantified. Lexicon items from the dictionary that have anomalous behavior reflected in the quantification are identified. One or more seed words identified as having anomalous behavior are selectively removed from the dictionary.

Identifying ambiguity in semantic resources

One embodiment provides a method for relevant language-independent terminology extraction from content, the method including extracting lexicon items from the content based on context extraction patterns using statistical processing. Feedback on the extracted lexicon items is received to automatically tune scores and thresholds for the context extraction patterns. Available Linked Data is leveraged for a bootstrap source. The relevant language-independent terminology extraction is bootstrapped using the bootstrap source.

Linked data seeded multi-lingual lexicon extraction

One embodiment provides a method for on-demand relation extraction from unstructured text that includes obtaining a text corpus of domain related unstructured text. Representations of the unstructured text that capture entity-specific syntactic knowledge are created. Initial user seeds of informative examples containing relations are received. Extraction models in a neural network are trained using the initial user seeds. Performance information and a confidence score are provided for each prediction for each extraction model. A next batch of informative examples are identified for annotation from the text corpus based on training a neural network classifier on a pool of labeled informative examples. Stopping criteria is determined based on differences of the performance information and the confidence score in relation to parameters for each extraction model. Based on the stopping criteria, it is determined whether to retrain a particular extraction model after the informative examples have been labeled.

On-demand relation extraction from text

Embodiments relate to a system, program product, and method for iterative expansion and application of a domain-specific dictionary. One or more dictionary instances are applied against a text corpus. The dictionary is iteratively expanded and selectively populated with one or more additional dictionary instances, including semantically similar instances to the applied dictionary instances and extension instances contextually related to the applied dictionary instances. The iteratively expanded dictionary is applied to an unexplored corpus to identify matching corpus data to populated instances of the dictionary.

Growing community of inventors

San Jose, CA, United States of America

Petar Ristoski