A Fuzzy Fusion Framework for Generating Purpose-oriented Texts

Abstract

Fusing textual information, as type of information fusion, has been of great significance to those interested in making informative texts out of the existing ones. The main idea behind text fusion, like any other type of information fusion, is to merge the partial texts from different sources in such a way that the outcome can hold a reasonably high relevance with regard to certain objectives. In this paper, a fuzzy framework is proposed for text generation, according to which a range of relevant texts are merged to yield producing a new text that can help the users fulfill a certain functionality in plausible manner. The focal point in our approach with regard to fusion is the distance between the class prototype of a text on the one side and the feature vectors belonging to different subsets of the existing texts on the other side. Results of experiments, show that the suggested framework can be a suitable alternatives for performing fusion in the cases that the identity of the existing texts from the viewpoint of the texts considered is unclear. This would turn into an effective utilization of the existing texts for the purpose of generating new texts.

Exploring Factors Influencing Cryptocurrency Adoption: A Comprehensive Modeling Based on Fuzzy Cognitive Maps Approach

“…Extracting the important points of the text and adapting them to some of the desired functions is always one of the crucial concerns [39]. Since the previous studies are considered experts, we will use the method developed by Alizadeh and Jetter [40] to extract nodes and relationships from secondary data.…”