Trend graphs: Visualizing the evolution of concept relationships in large document collections

Ronen Feldman, Yonatan Aumann, Amir Zilberstein, Yaron Ben-Yehuda

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

16 Scopus citations

Abstract

The proliferation of digitally available textual data necessitates automatic tools for analyzing large textual collections. Thus, in analogy to data mining for structured databases, text mining is defined for textual collections. A central tool in text mining is the analysis of concept relationship, which discovers connections between different concepts, as reflected in the corpus. Most previous work on text mining in general, and concept relationship in particular, viewed the entire corpus as one monolithic entity. However, large corpuses are often composed of documents with different characteristics. Most importantly, documents are often tagged with timestamps (e.g. news articles), and thus represent the state of the domain in different time periods. In this paper we introduce a new technique for analyzing and visualizing differences and similarities in the concept relationships, as they are reflected in different segments of the corpus. Focusing on the case of timestamped documents, we introduce Trend Graphs, which provide a graphical tool for analyzing and visualizing the dynamic changes in concept relationships over time. Trend Graphs thus provide a tool for tracking the evaluation of the corpus over time, highlighting trends and discontinuities.

Original languageEnglish
Title of host publicationPrinciples of Data Mining and Knowledge Discovery - 2nd European Symposium, PKDD 1998, Proceedings
EditorsJan M. Zytkow, Mohamed Quafafou
PublisherSpringer Verlag
Pages38-46
Number of pages9
ISBN (Print)3540650687, 9783540650683
DOIs
StatePublished - 1998
Event2nd European Symposium on Principles of Data Mining and Knowledge Discovery in Databases, PKDD 1998 - Nantes, France
Duration: 23 Sep 199826 Sep 1998

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume1510
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference2nd European Symposium on Principles of Data Mining and Knowledge Discovery in Databases, PKDD 1998
Country/TerritoryFrance
CityNantes
Period23/09/9826/09/98

Bibliographical note

Publisher Copyright:
© Springer-Verlag Berlin Heidelberg 1998.

Fingerprint

Dive into the research topics of 'Trend graphs: Visualizing the evolution of concept relationships in large document collections'. Together they form a unique fingerprint.

Cite this