LODifier: Generating Linked Data from Unstructured Text

Abstract

Abstract. The automated extraction of information from text and its transformation into a formal description is an important goal in both Semantic Web research and computational linguistics. The extracted information can be used for a variety of tasks such as ontology generation, question answering and information retrieval. LODifier is an approach that combines deep semantic analysis with named entity recognition, word sense disambiguation and controlled Semantic Web vocabularies in order to extract named entities and relations between them from text and to convert them into an RDF representation which is linked to DBpedia and WordNet. We present the architecture of our tool and discuss design decisions made. An evaluation of the tool on a story link detection task gives clear evidence of its practical potential.

TechMiner: Extracting Technologies from Academic Publications

“…For example, the LODifier approach [9] combines deep semantic analysis, named entity recognition and word sense disambiguation to extract named entities and to convert them into an RDF representation. Similarly, the AGADISTIS [10] system is a knowledge-base-agnostic approach for named entity disambiguation which combines the Hypertext-Induced Topic Search algorithm with label expansion strategies and string similarity measures.…”

“…For supporting the technology extraction task we manually crafted two ontologies: sciObjCSOnto 8 and verbSciOnto 9 . The first was derived from sciObjOnto 10 [17] and defines a number of categories of scientific objects in the Computer Science field and their related terms.…”

“…TM determines the research areas by extracting the full list of topics associated to the initial papers and finding the lowest common super topic which covers at least 75% of them according to the CSO ontology. For example, given a candidate technology such as "LODifier" [9], TM will analyse the distribution of topics relevant to the associated papers and may find that most of them are subsumed by the Semantic Web topic, it will then associate the candidate with all the papers that contain the label "LODifier" and are tagged with "Semantic Web" or with one of its sub areas according to CSO, such as "Linked Data" and "RDF".…”

“…These approaches have focused especially on the biomedical field and address mainly the identification of scientific artefacts (e.g., genes [4], chemical components [5]) and epistemological concepts [6,7,8] (e.g., hypothesis, motivation, experiments). At the same time, the Linked Open Data cloud has emerged as an important knowledge base for supporting these methods [9,10,11].…”

Summaries on the Fly: Query-Based Extraction of Structured Knowledge from Web Documents

“…Hence, it often generates a generic summary of a textual document that might not reflect the user interests. Furthermore, after processing and detecting the most relevant concepts in a document, common text summarization techniques do not take advantage of the concepts found for representing the summaries in a structured form, which would improve reasoning over the structured text [1,3,28].…”