Text classification typically performs best with large training sets, but short texts are very common on the World Wide Web. Can we use resampling and data augmentation to construct larger texts using similar terms? Several current methods exist for working with short text that rely on using external data and contexts, or workarounds. Our focus is…

We address the challenging task of "computational natural language inference," by which we mean bridging two or more natural language texts while also providing an explanation of how they are connected. In the context of question answering (i.e., finding short answers to natural language questions), this inference connects the question…

Discussion of natural language processing components of information retrieval systems focuses on a full-text system called TOPIC (Text-Oriented Procedures for Information Management and Condensation of Expository Texts) that is based on text parsing. Highlights include design rationales for semantic parsing; text coherence; and a semantic model of…

Natural systems are self-organizing and stratify according to the angular articulation of their movements. In particular, movement is relative to the levels that define its behavior space. As natural phenomenon, text production is self-referential, In generating information, the intention of the producer of a text becomes specified. Specifying of…

A new approach to the study of text building behavior is possible through the application of Scanator. This paper consists of a demonstration of some of its fundamental functions. These are suitable in a study of text building. The basic concepts, expressed through an analysis of a small part of an interview text, derive from a thermodynamic…

Research on graphing presents its results as if knowing and understanding were something stored in peoples' minds independent of the situation that they find themselves in. Thus, there are no models that situate interview responses to graphing tasks. How, then, we question, are the interview texts produced? How do respondents begin and end…

The goal of this study was to analyze the development of the relationship between conceptual and linguistic processes as regards idea ordering and structuring (linearizing), when composing a text. Participants (from 7th graders to University students) were required to compose a text using a list of eleven scrambled ideas. Conceptual rules allow to…

Reports on a study that analyzed and compared a restricted set of full-text articles from a sub-specialty of biochemistry in terms of co-occurrences and co-absences of words. The consequences for the lexicographical approach to generating artificial intelligence from scientific texts are discussed. (JAK)