Data Ninja Services collaborated with Oracle to reach a major milestone in the integration of text analytics with Oracle Spatial and Graph. The Data Ninja Services client in Java can be used to analyze free texts, extract entities, generate RDF semantic graphs, and choose from a number of graph analytics to infer entity relationships. We demonstrated two case studies involving mining health news and detecting anomalies in product reviews. See the presentation.
Multiple types of text can be analyzed. The text can be short text, long text, comments, reviews, news, conversations, social media, or other text communications. Depending on the type of relationships between the entities, the graph database may be a RDF (Resource Description Framework) Semantic Graph or a Property Graph.
Text to Triples: RDF Semantic Graph Model
The Data Ninja API extracts the entities, categories, and concepts from free-form text. The text extraction includes relationships between the categories, concepts, and entities. The Data Ninja Client for Oracle Spatial and Graph then structures the Linked Data into RDF triples with the meaning of the text is derived from the content and enriched with semantic analysis. The Oracle Spatial and Graph takes the RDF triples and creates the shortest paths and the nearest neighbors allowing for more in-depth analysis and queries. New business insights are discovered from graph inferences that could not be queried using a relational database. Oracle Spatial and Graph includes map visualization of geographic data and visualization components for integration with other external services.
Life sciences, healthcare, publishing, and finance are a few of the industries to utilize a text analytics and RDF graph database solution.
Text to Networks: Property Graph Model
The property graph model highlights social network relationships. The focus of the Oracle graphical representation is the interconnections between nodes. Each unique node has a name, and this name is usually an entity. The node has a collection of properties such as age or language to define in greater detail the entity. The links or edges between the nodes will have a weight and a direction. Beyond precise entity extraction, the Data Ninja API adds to the understanding of the relationships by assigning a sentiment score with polarity to each of the entities. By looking for patterns in the data, you can find similar entities, make recommendations, and locate primary data paths. The Oracle Spatial and Graph property graph feature supports open source and commercial graph visualization.
Product recommendations, target markets, fraud detection, and identifying social influences are all typical use cases for property graph analytics.