Search results

1 – 10 of 308
Article
Publication date: 18 March 2022

Weipeng Lu and Xuefeng Yan

The purpose of this paper is to propose a approach for data visualization and industrial process monitoring.

Abstract

Purpose

The purpose of this paper is to propose a approach for data visualization and industrial process monitoring.

Design/methodology/approach

A deep enhanced t-distributed stochastic neighbor embedding (DESNE) neural network is proposed for data visualization and process monitoring. The DESNE is composed of two deep neural networks: stacked variant auto-encoder (SVAE) and a deep label-guided t-stochastic neighbor embedding (DLSNE) neural network. In the DESNE network, SVAE extracts informative features of the raw data set, and then DLSNE projects the extracted features to a two dimensional graph.

Findings

The proposed DESNE is verified on the Tennessee Eastman process and a real data set of blade icing of wind turbines. The results indicate that DESNE outperforms some visualization methods in process monitoring.

Originality/value

This paper has significant originality. A stacked variant auto-encoder is proposed for feature extraction. The stacked variant auto-encoder can improve the separation among classes. A deep label-guided t-SNE is proposed for visualization. A novel visualization-based process monitoring method is proposed.

Details

Assembly Automation, vol. 42 no. 2
Type: Research Article
ISSN: 0144-5154

Keywords

Article
Publication date: 18 May 2015

Victoria Uren and Aba-Sah Dadzie

The purpose of this paper is to assess high-dimensional visualisation, combined with pattern matching, as an approach to observing dynamic changes in the ways people tweet about…

2398

Abstract

Purpose

The purpose of this paper is to assess high-dimensional visualisation, combined with pattern matching, as an approach to observing dynamic changes in the ways people tweet about science topics.

Design/methodology/approach

The high-dimensional visualisation approach was applied to three science topics to test its effectiveness for longitudinal analysis of message framing on Twitter over two disjoint periods in time. The paper uses coding frames to drive categorisation and visual analytics of tweets discussing the science topics.

Findings

The findings point to the potential of this mixed methods approach, as it allows sufficiently high sensitivity to recognise and support the analysis of non-trending as well as trending topics on Twitter.

Research limitations/implications

Three topics are studied, these illustrate a range of frames, but results may not be representative of all science topics.

Social implications

Funding bodies increasingly encourage scientists to participate in public engagement. As social media provides an avenue actively utilised for public communication, understanding the nature of the dialog on this medium is important for the scientific community and the public at large.

Originality/value

This study differs from standard approaches to the analysis of micropost data, which tend to focus on large-scale data sets. It provides evidence that this approach enables practical and effective analysis of the content of midsize to large collections of microposts.

Details

Aslib Journal of Information Management, vol. 67 no. 3
Type: Research Article
ISSN: 2050-3806

Keywords

Article
Publication date: 21 August 2017

Xiaoming Zhang, Huilin Chen, Yanqin Ruan, Dongyu Pan and Chongchong Zhao

With the rapid development of materials informatics and the Semantic Web, the semantic-driven solution has emerged to improve traditional query technology, which is hard to…

Abstract

Purpose

With the rapid development of materials informatics and the Semantic Web, the semantic-driven solution has emerged to improve traditional query technology, which is hard to discover implicit knowledge from materials data. However, it is a nontrivial thing for materials scientists to construct a semantic query, and the query results are usually presented in RDF/XML format which is not convenient for users to understand. This paper aims to propose an approach to construct semantic query and visualize the query results for metallic materials domain.

Design/methodology/approach

The authors design a query builder to generate SPARQL query statements automatically based on domain ontology and query conditions inputted by users. Moreover, a semantic visualization model is defined based on the materials science tetrahedron to support the visualization of query results in an intuitive, dynamic and interactive way.

Findings

Based on the Semantic Web technology, the authors design an automatic semantic query builder to help domain experts write the normative semantic query statements quickly and simply, as well as a prototype (named MatViz) is developed to visually show query results, which could help experts discover implicit knowledge from materials data. Moreover, the experiments demonstrate that the proposed system in this paper can rapidly and effectively return visualized query results over the metallic materials data set.

Originality/value

This paper mainly discusses an approach to support semantic query and visualization of metallic materials data. The implementation of MatViz will be a meaningful work for the research of metal materials data integration.

Details

International Journal of Web Information Systems, vol. 13 no. 3
Type: Research Article
ISSN: 1744-0084

Keywords

Article
Publication date: 11 April 2008

Chihli Hung and Stefan Wermter

The purpose of this paper is to examine neural document clustering techniques, e.g. self‐organising map (SOM) or growing neural gas (GNG), usually assume that textual information…

Abstract

Purpose

The purpose of this paper is to examine neural document clustering techniques, e.g. self‐organising map (SOM) or growing neural gas (GNG), usually assume that textual information is stationary on the quantity.

Design/methodology/approach

The authors propose a novel dynamic adaptive self‐organising hybrid (DASH) model, which adapts to time‐event news collections not only to the neural topological structure but also to its main parameters in a non‐stationary environment. Based on features of a time‐event news collection in a non‐stationary environment, they review the main current neural clustering models. The main deficiency is a need of pre‐definition of the thresholds of unit‐growing and unit‐pruning. Thus, the dynamic adaptive self‐organising hybrid (DASH) model is designed for a non‐stationary environment.

Findings

The paper compares DASH with SOM and GNG based on an artificial jumping corner data set and a real world Reuters news collection. According to the experimental results, the DASH model is more effective than SOM and GNG for time‐event document clustering.

Practical implications

A real world environment is dynamic. This paper provides an approach to present news clustering in a non‐stationary environment.

Originality/value

Text clustering in a non‐stationary environment is a novel concept. The paper demonstrates DASH, which can deal with a real world data set in a non‐stationary environment.

Details

The Electronic Library, vol. 26 no. 2
Type: Research Article
ISSN: 0264-0473

Keywords

Article
Publication date: 9 March 2012

Shin‐Ying Huang, Rua‐Huan Tsaih and Wan‐Ying Lin

Creditor reliance on accounting‐based numbers as a persistent and traditional standard to assess a firm's financial soundness and viability suggests that the integrity of…

1817

Abstract

Purpose

Creditor reliance on accounting‐based numbers as a persistent and traditional standard to assess a firm's financial soundness and viability suggests that the integrity of financial statements is essential to credit decisions. The purpose of this paper is to provide an approach to explore fraudulent financial reporting (FFR) via growing hierarchical self‐organizing map (GHSOM), an unsupervised neural network tool, to help capital providers evaluate the integrity of financial statements, and to facilitate analysis further to reach prudent credit decisions.

Design/methodology/approach

This paper develops a two‐stage approach: a classification stage that well trains the GHSOM to cluster the sample into subgroups with hierarchical relationship, and a pattern‐disclosure stage that uncovers patterns of the common FFR techniques and relevant risk indicators of each subgroup.

Findings

An application is conducted and its results show that the proposed two‐stage approach can help capital providers evaluate the reliability of financial statements and accounting numbers‐based decisions.

Practical implications

Following the SOM theories, it seems that common FFR techniques and relevant risk indicators extracted from the GHSOM clustering result are applicable to all samples clustered in the same leaf node (subgroup). This principle and any pre‐warning signal derived from the identified indicators can be applied to assessing the reliability of financial statements and forming a basis for further analysis in order to reach prudent decisions. The limitation of this paper is the subjective parameter setting of GHSOM.

Originality/value

This is the first application of GHSOM to financial data and demonstrates an alternative way to help capital providers such as lenders to evaluate the integrity of financial statements, a basis for further analysis to reach prudent decisions. The proposed approach could be applied to other scenarios that rely on accounting numbers as a basis for decisions.

Article
Publication date: 2 January 2009

Petri Nokelainen and Pekka Ruohotie

This study aims to examine the factors of growth‐oriented atmosphere in a Finnish polytechnic institution of higher education with categorical exploratory factor analysis…

Abstract

Purpose

This study aims to examine the factors of growth‐oriented atmosphere in a Finnish polytechnic institution of higher education with categorical exploratory factor analysis, multidimensional scaling and Bayesian unsupervised model‐based visualization.

Design/methodology/approach

This study was designed to examine employee perceptions of how their managers create conditions that support professional growth and learning, and how the employees perceive their growth motivation and commitment to the organization. Data were gathered from 447 employees with the Growth‐oriented Atmosphere Questionnaire in a Finnish polytechnic institution of higher education.

Findings

Results showed that the theoretical four‐group classification of the growth‐oriented atmosphere factors was supported by the empirical evidence. Results further showed that managers and teachers had higher growth motivation and level of commitment to work than other personnel, including job titles such as cleaner, caretaker, accountant and computer support. Employees across all job titles in the organization, who have temporary or part‐time contracts, had higher self‐reported growth motivation and commitment to work and organization than their established colleagues.

Practical implications

Leaders in various organizations may benefit from learning what is the current professional growth status of diverse employee groups, and in understanding the potential differences in employee growth motivation.

Originality/value

This study contributes to an understanding of organizational growth and learning as a non‐linear process. The statistical non‐linear modeling approach is novel providing research and practical example of how to use these techniques in practice.

Details

Journal of Workplace Learning, vol. 21 no. 1
Type: Research Article
ISSN: 1366-5626

Keywords

Open Access
Article
Publication date: 3 June 2019

Lisa Maria Perkhofer, Peter Hofer, Conny Walchshofer, Thomas Plank and Hans-Christian Jetter

Big Data introduces high amounts and new forms of structured, unstructured and semi-structured data into the field of accounting and this requires alternative data management and…

11705

Abstract

Purpose

Big Data introduces high amounts and new forms of structured, unstructured and semi-structured data into the field of accounting and this requires alternative data management and reporting methods. Generating insights from these new data sources highlight the need for different and interactive forms of visualization in the field of visual analytics. Nonetheless, a considerable gap between the recommendations in research and the current usage in practice is evident. In order to understand and overcome this gap, a detailed analysis of the status quo as well as the identification of potential barriers for adoption is vital. The paper aims to discuss this issue.

Design/methodology/approach

A survey with 145 business accountants from Austrian companies from a wide array of business sectors and all hierarchy levels has been conducted. The survey is targeted toward the purpose of this study: identifying barriers, clustered as human-related and technological-related, as well as investigating current practice with respect to interactive visualization use for Big Data.

Findings

The lack of knowledge and experience regarding new visualization types and interaction techniques and the sole focus on Microsoft Excel as a visualization tool can be identified as the main barriers, while the use of multiple data sources and the gradual implementation of further software tools determine the first drivers of adoption.

Research limitations/implications

Due to the data collection with a standardized survey, there was no possibility of dealing with participants individually, which could lead to a misinterpretation of the given answers. Further, the sample population is Austrian, which might cause issues in terms of generalizing results to other geographical or cultural heritages.

Practical implications

The study shows that those knowledgeable and familiar with interactive Big Data visualizations indicate high perceived ease of use. It is, therefore, necessary to offer sufficient training as well as user-centered visualizations and technological support to further increase usage within the accounting profession.

Originality/value

A lot of research has been dedicated to the introduction of novel forms of interactive visualizations. However, little focus has been laid on the impact of these new tools for Big Data from a practitioner’s perspective and their needs.

Details

Journal of Applied Accounting Research, vol. 20 no. 4
Type: Research Article
ISSN: 0967-5426

Keywords

Article
Publication date: 18 November 2011

Benny Raphael

The purpose of this paper is to improve current design processes by proposing a new approach based on multi‐criteria optimization of the designed asset. Management of design in…

Abstract

Purpose

The purpose of this paper is to improve current design processes by proposing a new approach based on multi‐criteria optimization of the designed asset. Management of design in construction projects is a complex task since it involves collaboration between professionals in multiple disciplines. Traditionally, designers work with a single solution at a time which is iteratively modified according to the view points of all the consultants. This results in sub‐optimal solutions. A multi‐criteria approach is able to accommodate diverse view points of specialist consultants in a construction project, aiming at a better optimized building system.

Design/methodology/approach

The shortcomings of current design practices are analyzed based on a literature review. It is found that current approaches involving single objective optimization or Pareto optimization are not adequate for supporting collaborative design processes. A new approach to managing multiple objectives in design is proposed. This involves performing multi‐objective optimization, presenting a population of good solutions to the design consultants and selecting the best solution through an algorithm called RR‐PARETO2 (Relaxed‐Restricted Pareto) filtering. A software tool with a graphical user interface was developed. An example of the design of a building façade is taken to evaluate the application of this approach.

Findings

The paper provides empirical evidence that a multi‐objective optimization approach is able to provide support for the task of accommodating multiple viewpoints in design. The proposed methodology allows navigation through the solution space and pruning it visually by applying constraints. It is shown that the RR‐PARETO2 is able to select a good compromise solution with the best trade‐offs among all the objectives.

Originality/value

The idea of visualizing and filtering a population of design solutions has been proposed by design researchers for a long time, but is not currently adopted in practice in construction projects. The idea of collaborative filtering of the solution space according to the viewpoints of all the consultants by visually applying constraints on design variables and objectives is a new concept, the ultimate aim being a better balanced built asset. This is the first time the RR‐PARETO2 algorithm has been applied to building design.

Article
Publication date: 26 July 2019

Seda Yanık and Abdelrahman Elmorsy

The purpose of this paper is to generate customer clusters using self-organizing map (SOM) approach, a machine learning technique with a big data set of credit card consumptions…

Abstract

Purpose

The purpose of this paper is to generate customer clusters using self-organizing map (SOM) approach, a machine learning technique with a big data set of credit card consumptions. The authors aim to use the consumption patterns of the customers in a period of three months deducted from the credit card transactions, specifically the consumption categories (e.g. food, entertainment, etc.).

Design/methodology/approach

The authors use a big data set of almost 40,000 credit card transactions to cluster customers. To deal with the size of the data set and the eliminated the required parametric assumptions the authors use a machine learning technique, SOMs. The variables used are grouped into three as demographical variables, categorical consumption variables and summary consumption variables. The variables are first converted to factors using principal component analysis. Then, the number of clusters is specified by k-means clustering trials. Then, clustering with SOM is conducted by only including the demographical variables and all variables. Then, a comparison is made and the significance of the variables is examined by analysis of variance.

Findings

The appropriate number of clusters is found to be 8 using k-means clusters. Then, the differences in categorical consumption levels are investigated between the clusters. However, they have been found to be insignificant, whereas the summary consumption variables are found to be significant between the clusters, as well as the demographical variables.

Originality/value

The originality of the study is to incorporate the credit card consumption variables of customers to cluster the bank customers. The authors use a big data set and dealt with it with a machine learning technique to deduct the consumption patterns to generate the clusters. Credit card transactions generate a vast amount of data to deduce valuable information. It is mainly used to detect fraud in the literature. To the best of the authors’ knowledge, consumption patterns obtained from credit card transaction are first used for clustering the customers in this study.

Details

International Journal of Intelligent Computing and Cybernetics, vol. 12 no. 3
Type: Research Article
ISSN: 1756-378X

Keywords

Article
Publication date: 8 March 2013

Angelos Mimis and Thomas Georgiadis

The purpose of this paper is to examine the possibility of using non‐income indicators and the self‐organizing map (SOM) approach as an alternative analytical tool to map…

Abstract

Purpose

The purpose of this paper is to examine the possibility of using non‐income indicators and the self‐organizing map (SOM) approach as an alternative analytical tool to map countries' welfare status.

Design/methodology/approach

Using data from 27 countries of the East Asia‐Pacific region, a welfare analysis based on non‐income indicators is implemented. The set of the selected indicators employed includes measures of social indicators as well as indicators related to the overall development framework. The empirical approach of the present paper can be described as a two‐stage procedure. In the first stage, the standard incremental SOM algorithm has been used and the two‐dimensional map produced in a hexagonal grid is presented together with the weight maps. In the second stage, the k‐means methodology has been used to cluster the prototypes produced by the SOM.

Findings

The classification produced by the two‐stage approach of the empirical analysis is compared with the baseline World Bank's income categories (based on Gross National Income per capita) offering an opportunity to assess the usefulness of non‐parametric approaches that are based on non‐income indicators vis‐à‐vis World Bank's approach in analysing welfare outcomes. The emerging picture of the empirical analysis supports the potential of the SOM as a useful and prolific analytical tool in mapping welfare outcomes.

Originality/value

This study proposes a methodology beyond the conventional ordinal rankings of the welfare of the countries based on non‐income indicators and the SOM.

Details

International Journal of Social Economics, vol. 40 no. 4
Type: Research Article
ISSN: 0306-8293

Keywords

1 – 10 of 308