International Journal of Data Analysis Techniques and Strategies http://www.inderscience.com/browse/index.php?journalID=282&year=2012&vol=4&issue=1 Inderscience Publishers Ltd en-uk International Journal of Data Analysis Techniques and Strategies 1755-8050 1755-8069 © 2012 Inderscience Publishers Ltd editor@inderscience.com https://www.inderscience.com/images/files/coverImgs/ijdats_scoverijdats.jpg http://www.inderscience.com/browse/index.php?journalID=282&year=2012&vol=4&issue=1 http://www.inderscience.com/link.php?id=45119 In this research, a computer assisted framework was developed to help researchers in conducting qualitative research. This framework leveraged the GATE platform, along with natural language processing and knowledge extraction techniques, to develop an automatic text annotation and summarisation system. A performance model, developed from the literature on lean manufacturing implementation strategies was employed, and a lexicon database for lean implementation practices was also developed. A unique dataset from a previous research study focusing on lean implementation practices was used to conduct this development and testing. A number of different summarisation techniques were developed and tested. A customised sensitivity analysis method was developed and used to systematically perform summarisation algorithms comparisons. The results of this study showed that using this framework at the early stages of a qualitative study has a great potential to reduce the time spent by researchers in annotating large datasets. A computer-assisted qualitative data analysis framework for the engineering management domain
Amirali Saeedi; Toni L. Doolen
International Journal of Data Analysis Techniques and Strategies, Vol. 4, No. 1 (2012) pp. 1 - 20
In this research, a computer assisted framework was developed to help researchers in conducting qualitative research. This framework leveraged the GATE platform, along with natural language processing and knowledge extraction techniques, to develop an automatic text annotation and summarisation system. A performance model, developed from the literature on lean manufacturing implementation strategies was employed, and a lexicon database for lean implementation practices was also developed. A unique dataset from a previous research study focusing on lean implementation practices was used to conduct this development and testing. A number of different summarisation techniques were developed and tested. A customised sensitivity analysis method was developed and used to systematically perform summarisation algorithms comparisons. The results of this study showed that using this framework at the early stages of a qualitative study has a great potential to reduce the time spent by researchers in annotating large datasets.

]]> 10.1504/IJDATS.2012.045119 International Journal of Data Analysis Techniques and Strategies, Vol. 4, No. 1 (2012) pp. 1 - 20 Amirali Saeedi; Toni L. Doolen School of Mechanical, Industrial and Manufacturing Engineering, Oregon State University (OSU), 204 Rogers Hall, Corvallis, OR 97331-6001, USA. ' School of Mechanical, Industrial and Manufacturing Engineering, Oregon State University (OSU), 204 Rogers Hall, Corvallis, OR 97331-6001, USA qualitative data analysis QDA natural language processing NLP GATE general architecture text engineering engineering management computer-assisted framework knowledge extraction automatic text annotation automatic text summarisation performance models lean manufacturing lexicon databases lean implementation datasets sensitivity analysis summarisation algorithms algorithm comparisons researchers data analysis techniques data analysis strategies. 2012-01-27T23:20:50-05:00 4 1 1 20 2012-01-27T23:20:50-05:00 http://www.inderscience.com/link.php?id=45120 The Malmquist index, based on data envelopment analysis (DEA) models, is the prominent index for measuring the relative productivity change of decision-making units (DMUs) in multiple time periods. This study presents new insight into the Malmquist index for measuring the productivity change of a master unit (master-DMU) that encompasses a set of units, its sub-DMUs, in multiple time periods. This is achieved by providing a framework for measuring productivity changes at the master level for the master-DMU regarding productivity changes at the sub-level for all sub-DMUs. The suggested index is illustrated by means of a real-world example from banking. The master Malmquist index measurement using DEA-based weighted average efficiency
Mohsen Afsharian; Mohammadreza Alirezaee; Peter Reichling
International Journal of Data Analysis Techniques and Strategies, Vol. 4, No. 1 (2012) pp. 21 - 42
The Malmquist index, based on data envelopment analysis (DEA) models, is the prominent index for measuring the relative productivity change of decision-making units (DMUs) in multiple time periods. This study presents new insight into the Malmquist index for measuring the productivity change of a master unit (master-DMU) that encompasses a set of units, its sub-DMUs, in multiple time periods. This is achieved by providing a framework for measuring productivity changes at the master level for the master-DMU regarding productivity changes at the sub-level for all sub-DMUs. The suggested index is illustrated by means of a real-world example from banking.

]]> 10.1504/IJDATS.2012.045120 International Journal of Data Analysis Techniques and Strategies, Vol. 4, No. 1 (2012) pp. 21 - 42 Mohsen Afsharian; Mohammadreza Alirezaee; Peter Reichling Faculty of Economics and Management, Otto-von-Guericke University Magdeburg, 39106 Magdeburg, Germany. ' School of Mathematics, Iran University of Science and Technology, 16846 Tehran, Iran. ' Faculty of Economics and Management, Otto-von-Guericke University Magdeburg, 39106 Magdeburg, Germany Malmquist DEA data envelopment analysis Sten Malmquist productivity index master Malmquist index MMI weighted average efficiency relative change productivity changes DMU decision making unit multiple time periods master unit sub-levels banking central banks Iran data analysis techniques data analysis strategies. 2012-01-27T23:20:50-05:00 4 1 21 42 2012-01-27T23:20:50-05:00 http://www.inderscience.com/link.php?id=45121 The ongoing financial and economic crisis throughout the industrialised world has spotlighted a number of significant deficiencies in corporate governance and management. The strength and composition of the management team along with effective corporate governance policy should play an important role in addressing these challenges. The purpose of this paper is to illustrate how analytics can be used to identify the importance of specific organisational factors that could impact corporate performance. A WRDS database consisting of a variety of factors was examined using Logit, neural net and CART modelling techniques. The results from the analysis indicate that diversity and governance policies appear to have played only a modest role in explaining corporate performance as measured by Tobin's Q for the year 2004. A comparative modelling analysis of firm performance
Owen P. Hall Jr.; Darrol J. Stanley
International Journal of Data Analysis Techniques and Strategies, Vol. 4, No. 1 (2012) pp. 43 - 56
The ongoing financial and economic crisis throughout the industrialised world has spotlighted a number of significant deficiencies in corporate governance and management. The strength and composition of the management team along with effective corporate governance policy should play an important role in addressing these challenges. The purpose of this paper is to illustrate how analytics can be used to identify the importance of specific organisational factors that could impact corporate performance. A WRDS database consisting of a variety of factors was examined using Logit, neural net and CART modelling techniques. The results from the analysis indicate that diversity and governance policies appear to have played only a modest role in explaining corporate performance as measured by Tobin's Q for the year 2004.

]]> 10.1504/IJDATS.2012.045121 International Journal of Data Analysis Techniques and Strategies, Vol. 4, No. 1 (2012) pp. 43 - 56 Owen P. Hall Jr.; Darrol J. Stanley Department of Decision Sciences, Graziadio School of Business and Management, Pepperdine University, 24255 Pacific Coast Highway, Malibu, CA 90263, USA. ' Department of Finance and Accounting, Graziadio School of Business and Management, Pepperdine University, 24255 Pacific Coast Highway, Malibu, CA 90263, USA corporate governance S&amp P 500 capitalisation-weighted indices Standard &amp Poor' s public companies stock exchanges USA United States CART classification tree regression tree decision trees neural nets Tobin' s quotient James Tobin statistical ratios market value replacement value physical assets firm performance Tobin' s Q comparative analysis modelling analysis financial crises economic crises industrialised world corporate management management teams analytics organisational factors corporate performance Wharton Research Data Services WRDS databases logit diversity policies governance policies corporate performance data analysis techniques data analysis strategies. 2012-01-27T23:20:50-05:00 4 1 43 56 2012-01-27T23:20:50-05:00 http://www.inderscience.com/link.php?id=45122 The multidimensional data model can be effectively utilised for analysing huge and detailed meteorological datasets forecasted by numerical weather prediction (NWP) model. The model cannot predict any weather event directly. The output products of model are interpreted by man-machine mix to infer the idiosyncratic behaviour of weather events. The mathematical tools for analysis and forecasting are able to provide forecast of weather variables only at grid-points. In this paper, the technology of dimension modelling has been adapted for analysing NWP model output datasets corresponding to sub-grid scale events viz. cloudburst, using OLAP technique. The huge datasets of weather variables available directly and derived indirectly, are mined so as to locate the patterns of cloudburst formation. K-means clustering technique has been used to generate clusters of convergence and divergence, for four real-life cases of cloudburst. It has been observed that clustering technique can help in identification of patterns conducive to formation of cloudburst. A data warehousing and data mining approach for analysis and forecast of cloudburst events using OLAP-based data hypercube
Kavita Pabreja; Rattan K. Datta
International Journal of Data Analysis Techniques and Strategies, Vol. 4, No. 1 (2012) pp. 57 - 82
The multidimensional data model can be effectively utilised for analysing huge and detailed meteorological datasets forecasted by numerical weather prediction (NWP) model. The model cannot predict any weather event directly. The output products of model are interpreted by man-machine mix to infer the idiosyncratic behaviour of weather events. The mathematical tools for analysis and forecasting are able to provide forecast of weather variables only at grid-points. In this paper, the technology of dimension modelling has been adapted for analysing NWP model output datasets corresponding to sub-grid scale events viz. cloudburst, using OLAP technique. The huge datasets of weather variables available directly and derived indirectly, are mined so as to locate the patterns of cloudburst formation. K-means clustering technique has been used to generate clusters of convergence and divergence, for four real-life cases of cloudburst. It has been observed that clustering technique can help in identification of patterns conducive to formation of cloudburst.

]]> 10.1504/IJDATS.2012.045122 International Journal of Data Analysis Techniques and Strategies, Vol. 4, No. 1 (2012) pp. 57 - 82 Kavita Pabreja; Rattan K. Datta Birla Institute of Technology and Science, Vidya Vihar Campus, Pilani, Rajasthan, 333031, India. ' Gwalior Academy of Mathematical Sciences, A-9, USO Road, Shaheed Jit Singh Marg, New Delhi, 110067, India OLAP online analytical processing cloudbursts numerical weather prediction k-means clustering cluster analysis convergence data mining data warehouses dimension modelling data hypercubes forecasts forecasting multidimensional data models meteorological datasets weather events meteorology output products man-machine mix idiosyncratic behaviour weather variables grid-points sub-grid scale events output datasets cloudburst formations formation patterns clouds divergence Dhaka Bangladesh Pittorgarh Uttarakhand Chamoli Shimla Himachal Pradesh India data analysis techniques data analysis strategies. 2012-01-27T23:20:50-05:00 4 1 57 82 2012-01-27T23:20:50-05:00 http://www.inderscience.com/link.php?id=45123 As the crucial mainstay for insurance industry to survive and develop, the insurance investment enables insurance companies to offset their possible underwriting losses and make a considerable profit. There have been many issues that affect the investment of Nigeria insurance companies. These include lack of investment vehicles, low rate of return on investment, unstable policies from the regulators and irrational investment portfolio. All these factors combined have restricted the development of Nigeria insurance industry. This study combined an 'expanded Lagrangian' function with a modified trust region method to propose a method for solving investment portfolio management problems of insurance companies. The proposed method was implemented on the portfolio management problems of a group of insurance companies. The study offers alternative solution to portfolio investment management by implementing best processes for minimising the risk for a given expected return, which is generally a non-linear function. It devised a means of asset allocation among a number of investment opportunities in achieving the investment objectives of insurance companies. A non-linear programming model for insurance company investment portfolio management in Nigeria
Emmanuel Olateju Oyatoye; Waheed Oladimeji Arilesere
International Journal of Data Analysis Techniques and Strategies, Vol. 4, No. 1 (2012) pp. 83 - 100
As the crucial mainstay for insurance industry to survive and develop, the insurance investment enables insurance companies to offset their possible underwriting losses and make a considerable profit. There have been many issues that affect the investment of Nigeria insurance companies. These include lack of investment vehicles, low rate of return on investment, unstable policies from the regulators and irrational investment portfolio. All these factors combined have restricted the development of Nigeria insurance industry. This study combined an 'expanded Lagrangian' function with a modified trust region method to propose a method for solving investment portfolio management problems of insurance companies. The proposed method was implemented on the portfolio management problems of a group of insurance companies. The study offers alternative solution to portfolio investment management by implementing best processes for minimising the risk for a given expected return, which is generally a non-linear function. It devised a means of asset allocation among a number of investment opportunities in achieving the investment objectives of insurance companies.

]]> 10.1504/IJDATS.2012.045123 International Journal of Data Analysis Techniques and Strategies, Vol. 4, No. 1 (2012) pp. 83 - 100 Emmanuel Olateju Oyatoye; Waheed Oladimeji Arilesere Department of Business Administration, Faculty of Business Administration, University of Lagos, Akoka-Yaba, Lagos, Nigeria. ' Oceanic Insurance, Health and Health Insurance, No. 20, Ozumba Mbadiwe Street, Victoria Island, Lagos, Nigeria nonlinear programming models insurance investment portfolio management expanded Lagrangian function Joseph Lagrange modified trust region Nigeria insurance companies investment portfolios underwriting losses profits investment vehicles rates of return unstable policies regulators irrational investments risk minimisation expected returns nonlinear functions asset allocation investment opportunities investment objectives Oceanic Insurance Company Lagos Oceanic Life Assurance Company Oceanic Health Insurance Company data analysis techniques data analysis strategies. 2012-01-27T23:20:50-05:00 4 1 83 100 2012-01-27T23:20:50-05:00 http://www.inderscience.com/link.php?id=45124 Association rule mining is applied to large databases to identify product associations. In the resulting large number of rules, interestingness is difficult to determine. Researchers have defined various measures of 'interestingness' such as support, confidence, lift and gain. Support is the probability of occurrence of an item or set of items, and is the most important of these measures, since the other measures are calculated using support. This current research suggests some deficiencies in the support measure and shows it is not consistent with its definition. Because other measures are calculated using support, this may make the other measures inconsistent. The researcher in this study proposes a new measure called normalised support, which is normalisation of general support, in other context-adjusted support or penalised support. Normalised support recommendations can stabilise product sale by product cross-sell promotion. In addition, the usefulness of other measures improves automatically. Normalised support: a virtual angle of measurement of 'interestingness'
Waleed Alsabhan
International Journal of Data Analysis Techniques and Strategies, Vol. 4, No. 1 (2012) pp. 101 - 114
Association rule mining is applied to large databases to identify product associations. In the resulting large number of rules, interestingness is difficult to determine. Researchers have defined various measures of 'interestingness' such as support, confidence, lift and gain. Support is the probability of occurrence of an item or set of items, and is the most important of these measures, since the other measures are calculated using support. This current research suggests some deficiencies in the support measure and shows it is not consistent with its definition. Because other measures are calculated using support, this may make the other measures inconsistent. The researcher in this study proposes a new measure called normalised support, which is normalisation of general support, in other context-adjusted support or penalised support. Normalised support recommendations can stabilise product sale by product cross-sell promotion. In addition, the usefulness of other measures improves automatically.

]]> 10.1504/IJDATS.2012.045124 International Journal of Data Analysis Techniques and Strategies, Vol. 4, No. 1 (2012) pp. 101 - 114 Waleed Alsabhan University of Sharjah, P.O. Box 27272, Sharjah, UAE data mining databases interestingness measurement product associations rule mining normalised support market basket analysis support confidence lift gain probability item occurrence item sets support measures inconsistent measures normalisation context-adjusted support penalised support support recommendations product sales cross-sell promotion grocery stores data analysis techniques data analysis strategies. 2012-01-27T23:20:50-05:00 4 1 101 114 2012-01-27T23:20:50-05:00