CONTEXT BASED GENERATION OF WEIGHT AND STRENGTH FOR KEYWORDS OF TECHNICAL PARAGRAPHS
CONTEXT BASED GENERATION OF WEIGHT AND STRENGTH FOR KEYWORDS OF TECHNICAL PARAGRAPHS

Summary/Abstract: Technical and other knowledge is associated with human minds and contemporary persistent storage media such as disk, ROM, and cloud based storage. The quality and complexity of such knowledge is not calibrated as a metric and usually is not made available as a cognitive input to end consumers of such knowledge. Traditionally, the evaluation of one's knowledge is determined by probing the knowledge acquired with relevant questions and assessing associated answers in that context. In the case of knowledge represented as technical paragraphs, the evaluation is carried out based on the semantics of the relevant answer given in the question's context. It is empirically observed that the evaluator judges semantics of the context based answer by examining the quality of technical keywords and the connectors used and represented by the respective language in which it is written. This is understood to be achieved naturally by associating cognition based weight and strength, whose formal definitions are introduced in this paper, to the keywords in that context. This paper attempts to automate the process of associating weight and strength to the technical keywords of a paragraph in the respective domain and sub-domain knowledge context represented by valid and relevant technical literature. This is achieved by parsing the technical paragraph. Subsequent to parsing, the keyword tokens are mapped to the knowledge base and domain based classification of keywords is carried out. This classification is used as the basis for associating weight and strength to these technical keywords. This methodology has been tested out on variable size technical paragraphs in homogenous and heterogeneous domain-sub domain context. Consistency of similar patterns of technical weight and strength were observed.

• Details
• Contents