Abstract: Systems and methods are described for generating a machine learning model for multi-modal feature extraction. The method may include receiving a document in a digital format, where the digital format comprises text information and image information, performing a text extraction function on a first portion of the document to produce a set of text features, performing an image extraction function on a second portion of the document to produce a set of image features, generating a feature tree, wherein a plurality of nodes of the feature tree correspond to the set of text features and the set of image features, and generating an input vector for a machine learning model based on the feature tree. In some cases, the feature tree may be generated synthetically, or modified by a user prior to being converted into the input vector.

Abstract: Provided are techniques for performing entity-based semantic graph search. Semantic extraction is performed on content items to identify entities. A semantic graph is generated with a vertex for each of the content items, each of the entities, and each user associated with any of the content items and with edges between vertices representing relationships, wherein each of the edges is encoded with metadata about a type of a relationship and a strength of a relationship. A vertex structure is generated that contains identifiers of the content items, the entities, and each user mapped to vertices in the semantic graph. In response to receiving a search request with a search term, using the semantic graph and the vertex structure to identify search results for the search term.

Abstract: Systems and methods are described for generating a machine learning model for multi-modal feature extraction. The method may include receiving a document in a digital format, where the digital format comprises text information and image information, performing a text extraction function on a first portion of the document to produce a set of text features, performing an image extraction function on a second portion of the document to produce a set of image features, generating a feature tree, wherein a plurality of nodes of the feature tree correspond to the set of text features and the set of image features, and generating an input vector for a machine learning model based on the feature tree. In some cases, the feature tree may be generated synthetically, or modified by a user prior to being converted into the input vector.

Abstract: Provided are techniques for performing entity-based semantic graph search. Semantic extraction is performed on content items to identify entities. A semantic graph is generated with a vertex for each of the content items, each of the entities, and each user associated with any of the content items and with edges between vertices representing relationships, wherein each of the edges is encoded with metadata about a type of a relationship and a strength of a relationship. A vertex structure is generated that contains identifiers of the content items, the entities, and each user mapped to vertices in the semantic graph. In response to receiving a search request with a search term, using the semantic graph and the vertex structure to identify search results for the search term.