A method of categorizing text entries on a document can include determining, for each of a plurality of text bounding boxes in the document, respective text, respective coordinates, and respective input embeddings. The method may further include defining a graph of the plurality of bounding boxes, the graph comprising a plurality of connections among the plurality of bounding boxes, each connection comprising a first and second bounding box and zero or more respective intermediate bounding boxes. The method may further include determining a respective attention value for each connection according to a quantity of intermediate bounding boxes in the connection and, based on a the respective attention values and a transformer-based machine learning model applied to the respective input embeddings and respective coordinates, determining output embeddings for each bounding box and, based on the respective output embeddings, generating a bounding box label for each bounding box.

A method for extracting information from a molecular pathway diagram may be provided. The method includes providing a molecular pathway diagram, detecting basic graphical structural elements in the diagram resulting in a set of basic objects, detecting a graphical semantic of each of the basic graphical structural elements resulting in a set of structural primitives, and detecting a graphical syntax of the basic graphical structural element relative to each other and to the diagram. Furthermore, the method includes assigning metadata to a plurality of the detected basic graphical structural elements, where the metadata includes basic graphical structural element data, graphical semantic data and graphical syntax data resulting in a set of entities and relationships.

A method, system, and computer program product, include receiving a first input at a first element among a plurality of elements associated with at least one electronic document, determining a second element associated with the first element from the plurality of elements based on predetermined relations of the plurality of elements, and causing a view to be displayed together with an electronic document including the first element, the view at least including the second element.

A method for extracting information from a molecular pathway diagram may be provided. The method includes providing a molecular pathway diagram, detecting basic graphical structural elements in the diagram resulting in a set of basic objects, detecting a graphical semantic of each of the basic graphical structural elements resulting in a set of structural primitives, and detecting a graphical syntax of the basic graphical structural element relative to each other and to the diagram. Furthermore, the method includes assigning metadata to a plurality of the detected basic graphical structural elements, where the metadata includes basic graphical structural element data, graphical semantic data and graphical syntax data resulting in a set of entities and relationships.

A method, system, and computer program product, include receiving a first input at a first element among a plurality of elements associated with at least one electronic document, determining a second element associated with the first element from the plurality of elements based on predetermined relations of the plurality of elements, and causing a view to be displayed together with an electronic document including the first element, the view at least including the second element.

Herein is machine learning for anomalous graph detection based on graph embedding, shuffling, comparison, and unsupervised training techniques that can characterize an unfamiliar graph. In an embodiment, a computer obtains many known vectors that respectively represent known graphs. A new vector is generated that represents a new graph that contains multiple vertices. The new vector may contain an arithmetic aggregation of vertex vectors that respectively represent multiple vertices and/or a vector that represents a virtual vertex that is connected to the multiple vertices by respective virtual edges. In the many known vectors, some similar vectors that are similar to the new vector are identified. The new graph is automatically characterized based on a subset of the known graphs that the similar vectors represent.

A method for processing documents for enhanced search includes identifying a set of bounding boxes in the document. The method further includes defining one or more pairs of bounding boxes in the document. Each pair of bounding boxes is defined by a binary relation. The method further includes constructing a directed acyclic graph (DAG) from the one or more pairs of bounding boxes. The method further includes determining a topological sorting of each bounding box in the document based on the DAG. The topological sorting defines an adjacency relationship between the bounding boxes in the document. The method further includes extracting key-value pairs from the document based on the adjacency relationship between the bounding boxes in the document. The method further includes storing the key-value pairs in a key-value pair database.

A method, computer program product, and computer system for analyzing an image to detect a plurality of geometric shapes in the image. The method may also include building a graph data structure resembling the image based upon, at least in part, analyzing the image. In some embodiments, building the graph data structure may include traversing the image to generate one or more graph data structure clauses.

Herein is machine learning for anomalous graph detection based on graph embedding, shuffling, comparison, and unsupervised training techniques that can characterize an unfamiliar graph. In an embodiment, a computer obtains many known vectors that respectively represent known graphs. A new vector is generated that represents a new graph that contains multiple vertices. The new vector may contain an arithmetic aggregation of vertex vectors that respectively represent multiple vertices and/or a vector that represents a virtual vertex that is connected to the multiple vertices by respective virtual edges. In the many known vectors, some similar vectors that are similar to the new vector are identified. The new graph is automatically characterized based on a subset of the known graphs that the similar vectors represent.

A method for creating a machine learning system which is designed for segmentation and object detection in images. The method includes: providing a directed graph; selecting a path through the graph, at least one additional node being selected from this subset, a path through the graph from the input node along the edges via the additional node up to the output node being selected; creating a machine learning system as a function of the selected path; and training the machine learning system created.