Semi-automatic Parsing for Web Knowledge Extraction through Semantic Annotation

Semi-automatic Parsing for Web Knowledge Extraction through Semantic Annotation Maria Pia di Buono University of Salerno, DSPSC Via Giovanni Paolo II, 182 – 84084 Fisciano (SA), IT mdibuono@unisa.it Abstract Parsing Web information, namely parsing content to find relevant documents on the basis of a user’s query, represents a crucial step to guarantee fast and accurate Information Retrieval (IR). Generally, an automated approach to such task is considered faster and cheaper than manual systems. Nevertheless, results do not seem have a high level of accuracy, indeed, as also Hjorland (2007) states, using stochastic algorithms entails low precision, low recall and generic results. Usually IR systems are based on invert text index, namely an index data structure storing a mapping from content to its locations in a database file, or in a document or a set of documents. In this paper we propose a system, by means of which we will develop a search engine able to process online documents, starting from a natural language query, and to return information to users. The proposed approach, based on the Lexicon-Grammar (LG) framework and its language formalization methodologies, aims at integrating a semantic annotation process for both query analysis and document retrieval. Keywords: Knowledge Extraction, Semantic Annotation, Parsing 1. Introduction Parsing Web information, namely parsing content to find relevant documents on the basis of a user’s query, represents a crucial step to guarantee fast and accurate Information Retrieval (IR). Generally, an automated approach to such task is considered faster and cheaper than manual systems. Nevertheless, results do not seem have a high level of accuracy, indeed, as also Hjorland (2007) states, using stochastic algorithms entails: • Low precision due to the indexing of common Atomic Linguistic Units (ALUs) or sentences. • Low recall caused by the presence of synonyms. • Generic results arising from the use of too broad or too narrow terms. Usually IR systems are based on invert text index, namely an index data structure storing a mapping from content to its locations in a database file, or in a document or a set of documents. Most traditional IR systems process each document separately to retrieve terms in free-text query, which means that they do not compare results provided from different sources. Such lack of integration in results causes overlapping and decreasing in the positive predictive value, due to the fact that shared content are indexed several times. Various approaches have been proposed to overcome this boundary, increasing recall and precision in results. For example, Lempel from Yahoo! Labs deals with query evaluation strategies which are based on Term-at-a-Time (TAAT) and Document-at-a-Time Evaluation (DAAT) processing. Furthermore, different researches employ concept-based and semantic approach in order to process both documents and queries through semantic entities and concepts. Baziz et al. (2005) and Boubekeur et al. (2010) describe their methods for assigning document ALUs to the correct ontological entries. Boubekeur and Azzoug (2013) propose an approach for semantic indexing based on concepts identified from a linguistic resource. In their work, authors use WordNet and …

Content truncated to 500 words