Article Preview
Top1. Introduction
Recently, the Semantic Web comprises techniques that promise to dramatically improve the current WWW (World Wide Web) and its use. With the emergence of the Semantic Web and the growing number of heterogeneous data sources, the benefits of ontologies are becoming widely accepted. Web ontologies define terms used as data (metadata) for explaining things of a special domain. Nowadays, researchers are paying attention to automatic transformation of Internet resources in the areas into ontologies (Antoniou & van Harmelen, 2004). The Web is an enormous source of information contained in billions of individual pages. Most information resource on the Web is presented in the form of semi-structured or unstructured documents, encoded as a mixture of loosely structured natural language text and template units.
Web-tables are used mainly for structuring information, and they are the important means of presenting structured information. Table structures represent relations between data in the table. Therefore, we can easily extract ontology from a table based on the features of table structures without the use of syntax analysis (Tanaka & Ishida, 2006). However, understanding of table contents requires table structure comprehension and semantic interpretation, which exceed the complexity of corresponding linguistic tasks. Previous studies for extracting information from Web-Tables are centralized to interpret table structure: Pivk et al., (2004) focused on understanding table-like structures only due to their structural dimension, and the table model consists of Physical, Structural, Functional and Semantic components. The authors transformed the most relevant table types into F-logic frames, and also demonstrated and evaluated the successful generation of frames from HTML (HyperText Mark-up Language) tables. Jung et al., (2006) suggested a method for extracting table-schemata based on table structure and heuristics. Using this method, a table is converted into a table-schema and a triple. Moreover, since it is important to determine whether or not a table has meaningfulness that is related to the structural information provided at the level of the table head, the authors further investigated the types of tables, established the features that distinguished meaningful tables from others, built a training data set using these features and constructed a classification model using a decision tree. However, the work in Jung et al. (2006) cannot handle a table without head and also cannot extract an appropriate head using background color and font. Further, Chavan & Shirgave (2011) introduced a method for determining the meaningfulness of a table and extracting the head contents from meaningful table. The authors applied table mining to general HTML documents, separated meaningful tables from decorative tables, and extracted the information using head.
As specified above, it can be found that the work mentioned above mainly focused on how to extract information from tables, instead of investigating how to extract ontologies from Web-tables, and the detailed ontology extraction rules are not provided. In addition, most of research endeavors to interpret the table by using structural characteristics of the table. But, many Web-tables are designed by humans, thus, it has a certain limit to automatically interpret tables using only structural information of the table. Jung et al., (2007) detected that, generally, a table provides a semantic core element in a head, and proposed a method for obtaining class-instance relationship and triple using heuristics for extracting table schemata based on semantic core element. Although the authors in Jung et al. (2007) proposed heuristics for detecting semantic characteristics based on the location of table cells, they did not mention which becomes semantic core element.