Detecting Deviations in Text Collections: An Approach Using Conceptual Graphs

Abstract

Abstract. Deviation detection is an important problem of both data and text mining. In this paper we consider the detection of deviations in a set of texts represented as conceptual graphs. In contrast with statistical and distance-based approaches, the method we propose is based on the concept of generalization and regularity. Among its main characteristics are the detection of rare patterns (that attempt to give a generalized description of rare texts) and the ability to discover local deviations (deviations at different contexts and generalization levels). The method is illustrated with the analysis of a set of computer science papers.

Conceptual graph formalism for financial text representation

“…Using this method, the overlap between two conceptual graphs is measured by calculating conceptual similarity and relational similarity. The expectation of our future work is to achieve the same results as proven in [18], i.e. regarding CG as an index of the text collections; detection of patterns not only rare graphs; and to be able to visualize the deviations from different level of generalization.…”

“…For this purpose we will use the conceptual graph clustering method proposed in [18]. Using this method, the overlap between two conceptual graphs is measured by calculating conceptual similarity and relational similarity.…”

Natutal language processing: Perspective of CIC-IPN

“…We have applied CGs to information retrieval [65], as well as to the task of finding semantic deviations in a collection of documents [66] and other tasks of text mining [67]. The use of CGs allowed us to take into account fine details of the texts [68].…”

Analyzing Clusters and Constellations from Untwisting Shortened Links on Twitter Using Conceptual Graphs

“…Cao [2] describes the use of conceptual graphs alongside fuzzy logic as a means of extending Semantic Web technologies to approach human expression and reasoning more effectively; conceptual graphs are here used as a means of representing natural language sentences. Montes-y-Gomez et al [10], for example, describe the use of conceptual graphs to represent a series of text, permitting the detection of rare patterns and local deviations (occurring at specific contexts and generalization levels) within the textual corpus. Spasic et al [38] identify Daraselia et al's [5] use of conceptual graphs as a representation of a number of ontological frames, permitting them to be queried or for further text mning work to be completed against them.…”