• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.9 no.2
• /
• pp.141-146
• /
• 2009

Support vector machine is a classification algorithm based on statistical learning theory. It has shown many results with good performances in the data mining fields. But there are some problems in the algorithm. One of the problems is its heavy computing cost. So we have been difficult to use the support vector machine in the dynamic and online systems. To overcome this problem we propose to use stratified sampling of statistical sampling theory. The usage of stratified sampling supports to reduce the size of training data. In our paper, though the size of data is small, the performance accuracy is maintained. We verify our improved performance by experimental results using data sets from UCI machine learning repository.

• 한국디지털정책학회:학술대회논문집
• /
• 2005.06a
• /
• pp.51-71
• /
• 2005

Performance measurement systems based on the principle that "if you can't measure it, you can't manage it" reinforce a short-term culture by focussing on tangible outputs. Instead, the focus of organisations should be on sustainable long-term performance through continuous systemic improvement. To establish and reinforce behaviours that drive systemic improvement, measurement and reporting systems need to be de designed to re-enforce work to role behaviour by managers. This paper discusses this concept and how it is being applied in practice through an ongoing action research project.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.6 no.9
• /
• pp.1-5
• /
• 1983

This study takes over the application of survey sampling theory to activity sampling and the application of activity sampling to time standard estimation. Cluster, stratified, and multistage sampling are studied in conjunction with random and systematic sampling. Estimation procedures that will maximize the information obtained per cost expended on the study and specification of the procedure to be used to estimate the accuracy of the estimates for the adopted procedure are considered. The use of multiple regression md linear programming to estimate standard element performance time from typical job lot production data is also considered.

• Journal of Information Science Theory and Practice
• /
• v.5 no.1
• /
• pp.15-25
• /
• 2017

Nowadays the enormous growth of modern information communication technologies and its massive use have influenced information users all over the world. Such a digital environment has drastically changed the information seeking patterns of information users of every community. These also tend law students to use various legal information sources and services in digital environments while seeking information. But there have been few empirical user studies on the aspect of digital information seeking behaviours of law students in either law or library and information science literatures. This paper aims to draw out patterns of information seeking behavior of students of law in digital environments at the University of Dhaka. A stratified random sample survey was conducted for this study. The results show that students prefer the electronic format of information rather than printed format. Major e-resources used by them and the influential factors of use were also identified in this study. This study also identified some crucial problems for seeking information and provides suggestions for the development of electronic legal information systems.

• Asian Pacific Journal of Cancer Prevention
• /
• v.15 no.10
• /
• pp.4339-4347
• /
• 2014

Purpose: This study aimed at exploring treatment delay (TD) among cancer patients in China with an attempt to develop a practical methodology facilitating frontline Chinese clinicians in promoting earlier cancer diagnosis and treatment. Materials and Methods: The study comprised framework development, qualitative interviews and paired factor rating. Framework development utilized systematic literature review, soft systems thinking and consensus groups. Qualitative interviews employed a checklist of open questions soliciting information about all the domains included the framework from cancer patients drawn via stratified randomized sampling of inpatients at 10 hospitals in Hefei, China. Paired factor rating used a self-developed computer aid and the interviewed patients as referring cases to weigh the relative importance of the factors listed in the framework in terms of their contributions to specific components of total delay (TD). Results: a) A conceptual framework was proposed consisting of a 6-step path to TD and 36 category determinants. b) A total of 227 patients were interviewed; their TD was 267.3 mean or 108 median days ranging from 0 to 2475 days; average appraisal, illness, behavioral, preparation and treatment delay accounted for 52.1%, 9.4%, 0.30%, 8.8% and 29.4% of the TD respectively. Individual side factors were rated substantially more important than environmental side factors (60% vs. 40%); most influential TD factors included cancer symptoms, overall health, family relations and knowledge about cancer and health. Conclusions: The framework proposed together with the interviewing and rating approaches used provide a potential new methodology for understanding cancer patients' TD and promoting earlier cancer treatment.