Abstract: Implementations of the present disclosure are generally directed to linking entities in dynamic knowledge graphs, and more particularly, to linking entities in knowledge graphs based on contextual evolution in one or more dimensions.

Abstract: An example method described herein involves receiving a data input; identifying a plurality of topics in the data input; determining an underrepresented set of data for a first set of topics of the plurality of topics based on a plurality of knowledge graphs associated with the first set of topics; calculating a score for each topic of the first set of topics based on a representative learning technique; determining that the score for a first topic of the first set of topics satisfies a threshold score; selecting a topic specific knowledge graph based on the first topic; identifying representative objects that are similar to objects of the data input based on the topic specific knowledge graph; generating representation data that is similar to the data input based on the representative objects to balance the underrepresented set of data with a set of data associated with a second set of topics of the plurality of topics; and performing an action associated with the representation data.

Abstract: A system may receive events derived from historical sensor data. The system may generate repeat event records based on the events. The system may cluster the repeat event records into repeat event clusters. The system may determine filter ranges for the repeat event clusters. The system may generate a binarization filter comprising instructions to generate a binary sequence indicative of at least one of the filter ranges. The system may execute the binarization filter to generate binary sequences corresponding to the repeat event records, respectively. The system may train a prediction model based on the binary sequences. The system may deploy the binarization filter and the prediction model to a real-time platform.

Abstract: A device may receive data associated with an event. The device may identify a context of the event based on receiving the data. The device may identify a similar event based on performing a comparison of the context of the event and a context of the similar event. The device may determine a set of pre-events associated with the event based on identifying a pre-event that occurred before the similar event. The set of pre-events may include at least one pre-event similar to the pre-event that occurred before the similar event. The device may determine a set of post-events associated with the event based on determining the set of pre-events and identifying a post-event that occurred after the similar event. The set of post-events may include at least one post-event similar to the post-event. The device may perform an action based on the set of post-events.

Abstract: According to an example, with respect to adverse information based ontology reinforcement, adverse information related to a product or a process may be ascertained and analyzed to further identify and ascertain a relevant ontology of a plurality of ontologies. A determination may be made as to whether the adverse information is present in the ascertained ontology, and if not, the adverse information may be integrated into the ascertained ontology to generate an updated ontology. Similar existing information corresponding to the ascertained adverse information may be identified in the updated ontology to determine an inconsistency between the identified similar existing information and the ascertained adverse information. The determined inconsistency may be used to modify the updated ontology to generate a reinforced ontology.

Abstract: A system for determining adverse effects associated with a substance includes a side-effect recognizer, a relationship extractor, a processor, and a reporting system. The side-effect recognizer utilizes a first recurrent neural network (RNN) to identify a first portion of received data associated with an adverse effect to thereby determine the adverse effect associated with the received data. The relationship extractor utilizes a second RNN to identify a second portion of the received data associated with a substance and a third portion of the received data that indicates a relationship between the substance and the adverse effect to thereby determine the substance associated with the received data and the relationship between the substance and the adverse effect. The processor is in communication with the adverse effect recognizer and the substance relationship extractor and aggregates and relates the adverse effect, substance, and relationship.

Abstract: A device obtains an image and generates, based on the image and using an image generation model, a set of candidate images for object recognition. The device determines, based on the set of candidate images and using an image evaluation model, that one or more candidate images, of the set of candidate images, are processed during image generation and using the image generation model to improve a resolution of the image. The device determines image embeddings for the one or more candidate images and based on the image, and determines an embeddings error for the one or more candidate images based on the image embeddings and the image. The device determines an identification of a feature of the image based on the image embeddings and the embeddings error, and performs an action relating to the identification of the feature based on determining the identification of the feature.

Abstract: An image retrieval system may receive an image query that includes image data. The image retrieval system may determine an image descriptor based on the image data. The image retrieval system may obtain a text descriptor associated with the image descriptor in the descriptor repository. The image retrieval system may generate a document query comprising a search parameter, the search parameter including the text descriptor. The image retrieval system may identify, in a document database, text documents based on the document query. The text documents may be associated with document identifiers. The image retrieval system may obtain, from the file mapping repository, image identifiers associated with the document identifiers. The image query result from the image retrieval system may reference images associated with the image identifiers.

Abstract: Implementations of the present disclosure are generally directed to linking entities in dynamic knowledge graphs, and more particularly, to linking entities in knowledge graphs based on contextual evolution in one or more dimensions.

Abstract: A device obtains an image and generates, based on the image and using an image generation model, a set of candidate images for object recognition. The device determines, based on the set of candidate images and using an image evaluation model, that one or more candidate images, of the set of candidate images, are processed during image generation and using the image generation model to improve a resolution of the image. The device determines image embeddings for the one or more candidate images and based on the image, and determines an embeddings error for the one or more candidate images based on the image embeddings and the image. The device determines an identification of a feature of the image based on the image embeddings and the embeddings error, and performs an action relating to the identification of the feature based on determining the identification of the feature.

Abstract: An example method described herein involves receiving a data input; identifying a plurality of topics in the data input; determining an underrepresented set of data for a first set of topics of the plurality of topics based on a plurality of knowledge graphs associated with the first set of topics; calculating a score for each topic of the first set of topics based on a representative learning technique; determining that the score for a first topic of the first set of topics satisfies a threshold score; selecting a topic specific knowledge graph based on the first topic; identifying representative objects that are similar to objects of the data input based on the topic specific knowledge graph; generating representation data that is similar to the data input based on the representative objects to balance the underrepresented set of data with a set of data associated with a second set of topics of the plurality of topics; and performing an action associated with the representation data.

Abstract: A platform may obtain, for a first set of entities involved in a money laundering investigation, target entity information for a target entity and related entity information for a set of related entities. The platform may analyze the target entity information and the related entity information to identify money laundering candidates. The platform may determine one or more relationships indicating a degree of similarity between the target entity and the one or more related entities. The platform may generate a graph data structure that associates the target entity and the one or more related entities using the one or more relationships. The platform may determine a score for the target entity and one or more scores for the one or more related entities. The platform may provide a recommendation indicating whether the target entity and/or the one or more related entities are likely to be engaging in money laundering.

Abstract: According to an example, with respect to adverse information based ontology reinforcement, adverse information related to a product or a process may be ascertained and analyzed to further identify and ascertain a relevant ontology of a plurality of ontologies. A determination may be made as to whether the adverse information is present in the ascertained ontology, and if not, the adverse information may be integrated into the ascertained ontology to generate an updated ontology. Similar existing information corresponding to the ascertained adverse information may be identified in the updated ontology to determine an inconsistency between the identified similar existing information and the ascertained adverse information. The determined inconsistency may be used to modify the updated ontology to generate a reinforced ontology.

Abstract: A device may receive multimedia data, metadata, and/or policy data. The device may process the policy data using a first set of techniques to determine a first set of embeddings for the policy data. The device may process the multimedia data or the metadata using a second set of techniques to determine a second set of embeddings for the multimedia data or the metadata. The device may process the first set of embeddings and the second set of embeddings using a knowledge graph to determine whether the multimedia content or the access by the user violates the policy. The device may perform an action based on a result of processing the first set of embeddings and the second set of embeddings. The action may relate to the multimedia content or the access to the multimedia content by the user.

Abstract: An example method described herein involves receiving a first data stream and a second data stream; determining that a plurality of entities are present in the image data of the first data stream; analyzing the first data stream to determine that an entity, of the plurality of entities, is unrecognizable in the image data of the first data stream; obtaining, by the device, a common knowledge graph associated with the first data stream and the second data stream, wherein the common knowledge graph includes information regarding the plurality of entities; annotating the common knowledge graph with first corresponding recognizable characteristics of the plurality of entities in the first data stream to generate a first annotated knowledge graph; annotating the common knowledge graph with second corresponding recognizable characteristics of the plurality of entities in the second data stream to generate a second annotated knowledge graph; determining whether the entity is recognizable based on the first annotated

Abstract: Systems and methods for identifying an adverse effect of a pharmaceutical substance are provided. A system may tokenize an electronic medical record for a plurality of tokens. The system may further generate a distance score between a first vector of a word embedding model and a second vector of the word embedding model. The first vector may correspond to a substance token and the second vector may correspond to a condition token. The system may further detect an absence of a preventative association between the substance token and the condition token in a prevention repository. The system may further generate a causal link indication indicative of a causal link between the substance token and the condition token in response to the distance score being greater than a threshold value and detection of the absence of the preventative association between the substance token and the condition token.

Abstract: Disclosed is an implant and method of making an implant. The implant having a housing that defines a cavity. The housing includes a sensor comprising a base attached to a diaphragm wherein said base may be positioned within said cavity. The sensor may be a capacitive pressure sensor. The diaphragm may be connected to the housing to hermetically seal said housing. The sensor may include electrical contacts positioned on the diaphragm. The attachment between the base and the diaphragm may define a capacitive gap and at least one discontinuity configured to enhance at least one performance parameter of said implant.

Abstract: A device may receive a plurality of data objects from a plurality of sources; identify text data, image data, and location data of the plurality of data objects; identify relevant data objects, of the plurality of data objects, based on the text data, and/or based on the image data, based on the location data, and/or based on comparing the text data, the image data, and the location data to a predefined element that identifies values relevant to a particular group or subject area; assign scores to the relevant data objects based on the text data, the image data, and the location data; aggregate the scores, as one or more aggregated scores, with regard to one or more users associated with the relevant data objects; and/or perform one or more actions based on the one or more aggregated scores associated with the one or more users.

Abstract: A system for predicting the occurrence of an event includes an event detector and a reporting processor. The event detector is configured to: receive data that defines a plurality of social media items; receive a real-time data feed; and predict the occurrence of an event based on a correlation between information in the plurality of social media items and activity associated with the real-time data feed. The reporting processor is configured to determine an event type associated with the event; identify a sentiment of the predicted event based on historical data in the real-time data feed, and generate a recommendation for preventing the occurrence of the event based on at least one of the event type and the sentiment of the predicted event. The recommendation includes a plurality of actions. The reporting processor is coupled to a knowledge graph database that corresponds to an ontology that defines one or more relationships between event types, and response types.

Abstract: Systems and methods for identifying an adverse effect of a pharmaceutical substance are provided. A system may tokenize an electronic medical record for a plurality of tokens. The system may further generate a distance score between a first vector of a word embedding model and a second vector of the word embedding model. The first vector may correspond to a substance token and the second vector may correspond to a condition token. The system may further detect an absence of a preventative association between the substance token and the condition token in a prevention repository. The system may further generate a causal link indication indicative of a causal link between the substance token and the condition token in response to the distance score being greater than a threshold value and detection of the absence of the preventative association between the substance token and the condition token.