Managing complexity in large data bases using self-organizing maps

Accounting Management and Information TechnologiesPublished 1 October 1998

Barbro Back, Kaisa Sere, Hannu Vanharanta

Citations72

TL;DR

This paper structure and analyze numbers in a large data base over several time periods using self organizing maps to overcome the problems associated with finding the appropriate undlying distribution and the functional form of the underlying data in the structuring task that is often encountered when using cluster analysis.

Abstract

The amount of financial information in today's sophisticated large data bases is substantial and makes comparisons between company performance—especially over time—difficult or at least very time consuming. The aim of this paper is to investigate whether neural networks in the form of self-organizing maps can be used to manage the complexity in large data bases. We structure and analyze accounting numbers in a large data base over several time periods. By using self-organizing maps, we overcome the problems associated with finding the appropriate underlying distribution and the functional form of the underlying data in the structuring task that is often encountered, for example, when using cluster analysis. The method chosen also offers a way of visualizing the results. The data base in this study consists of annual reports of more than 120 world wide pulp and paper companies with data from a five year time period.

Keywords

Computer ScienceDecision SciencesEconomics, Econometrics and Finance

Springer series in information sciencesSelf-Organizing Maps

10,377 Citations1997Teuvo Kohonen

The mathematical preliminaries, background, basic ideas, and implications of the Self-Organising Map algorithm are expounded in a manner which is accessible without prior expert knowledge.

Strategic Management JournalTHE APPLICATION OF CLUSTER ANALYSIS IN STRATEGIC MANAGEMENT RESEARCH: AN ANALYSIS AND CRITIQUE

2,384 Citations1996David J. Ketchen, Christopher L. Shook

Analysis of 45 published strategy studies reveals that the implementation of cluster analysis has been often less than ideal, perhaps detracting from the ability of studies to generate knowledge.

Academy of Management JournalCONFIGURATIONAL APPROACHES TO ORGANIZATIONAL ANALYSIS.

1,881 Citations1993Alan D. Meyer, Anne S. Tsui +1 more

OmegaTowards a general non-parametric model of corporate performance

95 Citations1994Ángel Santiago Fernández-Castro

An arrangement for arresting a seat of a vehicle includes an elongated stationary element which is provided with a row of separate recesses being arranged on the element so as to correspond to a plurality of possible positions of the seat.

OmegaThe comparative ability of self-organizing neural networks to define cluster structure

53 Citations1995S.K. Chen, Paul Mangiameli +1 more

This research investigates the ability of specific neural network architectures utilizing unsupervised learning to recover cluster structure from multivariate data sets with various levels of relative cluster dispersion and demonstrates that the Self Organizing Map network is a superior clustering technique and that its relative advantage over conventional techniques increases with higher levels ofrelative cluster disp immersion in the data.

View PDF

Information & ManagementThe acceptance of visual information in management

27 Citations1997Jörn‐Axel Meyer

An exploratory study resulting in a theory of acceptance of visual information is outlined, which employs the results of an empirical investigation conducted amongst managers.

The British Accounting ReviewACCOUNTING IDENTITIES AND THE DISTRIBUTION OF RATIOS

25 Citations1995Duarte Trigueiros

Validity and utility of a hyperknowledge-based financial benchmarking system

13 Citations2002Hannu Vanharanta, Timo Käkölä +1 more

This paper assesses the validity and utility of a new hyperknowledge working and learning environment applied to a financial benchmarking application that utilizes a large annual reports database and provides supporting evidence that the hyperknowledge-based application can meet the users' needs in the domain ofFinancial benchmarking and that the application could be advantageous in practice.

Competitive Financial Benchmarking Using Self-Organizing Maps

9 Citations1995Barbro Back, Mikko Irjala +2 more

The study demonstrates how a large annual reports database on international pulp and paper companies can be pre-processed, i.e. classified with self-organizing maps that is one form of neural networks, and shows that self- Organizing maps are a viable tool for organizing large databases into clusters of companies having similar financial characteristics.

OmegaVisualizing interfirm comparison

7 Citations1994EM Vermeulen, Jaap Spronk +1 more

A new way to visualize the results obtained with interfirm comparison that enables us to see at a single glance which firm has the highest explanatory variable, what the expected value of the explanatory variable is given the firm's characteristics, and what the corresponding values of its competitors are.

View PDF