On Fuzzy Extensions to Energy Ontologies for Text Processing Applications

A high-level electrical energy ontology with weighted attributes

“…The taxonomy of the concepts within FEEONT. 7 This application has been previously described in [21] together with the extensions to PQONT and WONT, which are summarized in Section 3.1 of the current paper.…”

“…The details of the building process of the high-level electrical energy ontology, which is linked to FPQONT and FWONT, are The main components of the electrical grid are, therefore, the generation subsystem (where energy is produced from various energy sources), the transmission subsystem (which transmits the generated energy from the generation subsystem to the distribution subsystem after making the necessary conversions with transformers), the distribution subsystem (which transmits the received energy to the consumers), and finally the utilization subsystem which comprises the customers utilizing the energy like industrial factories and houses of people. 3 A preliminary description of this extension procedure and the extended forms of these ontologies are previously described in [21] together with a text categorization system based on the extended ontologies.…”

“…4 Reciprocal rank, or first answer reciprocal rank, is a commonly used metric during the evaluation of information retrieval and question answering systems [22,23]. $e120329/FPQONT.owl and http://www.ceng.metu.edu.tr/ $e120329/FWONT.owl, respectively [21].…”

“…To clarify the symbols in the above formula: c is the number of concepts in FPQONT, l i denotes the label value of ith concept, lðl i Þ is the value of the membershipValueLabel of the ith concept, ss i is the number of synonyms for the ith concept (e.g., the size of the value of its synonymSet attribute), s i;j is the jth element within the value of synonymSet of the ith concept, lðs i;j Þ is the jth element within the value of membershipValuesSynonymSet of the ith concept (e.g., the membership value of the jth synonym to the corresponding concept), and freqðxÞ denotes the frequency of the phrase (a label or synonym), x, within an article [21]. The implemented categorization system classifies a given article as relevant to the PQ domain, if the relevance score for the article is greater than a threshold.…”

“…In order to fill in these membership values, Wikipedia disambiguation pages for the applicable concepts are used as follows: the reciprocal rank 4 of the use corresponding to the domain definition is used as the value of the corresponding value, i.e., if the nth use in the disambiguation page for a concept label or synonym corresponds the domain definition, then 1=n is used as the value of the membershipValueLabel attribute or that of the corresponding element within the value of the membershipValuesSynonymSet attribute [21]. For instance, at the time of writing this paper, in the disambiguation page for the Frequency concept (http://en.wikipedia.org/wiki/Frequency_(disambiguation)), the 3rd use corresponds to the one within the PQ domain, hence, the Frequency concept of extended PQONT, namely FPQONT, has 0.33 as the value of its membershipValueLabel attribute.…”