Inductive, evolutionary, and neural computing techniques for discrimination: a comparative study

Article Abstract:

Four types of machine learning techniques for two-group discrimination have been analyzed to determine their performance with respect to certain data distribution characteristics. The learning techniques under study are the decision-tree learning program based on an inductive learning algorithm, a genetic algorithm implementation aimed at learning the weight vector of a linear discrimination model, a three-layer neural network based on backpropagation learning and a public domain implementation of genetic programming based on genetic search that utilizes a hierarchical representation to derive general form mathematical discrimination models.

Computers, Models, Analysis, Computer networks, Computer programming, Neural networks, Algorithms, Learning, Teaching models, Machine learning

---

The ordered cutting stock problem

Article Abstract:

A new type of one-dimensional cutting stock problem (CSP), known as the ordered CSP, which explicitly restricts to one of the number of jobs in a production process that can be open, or in process, at any given point in time is developed. The results illustrate the ability of the genetic algorithm (GA)-based approach to significantly decrease the amount of scrap generated in the production process.

Scheduling (Management), Genetic algorithms

---

Data-driven classification using boundary observations

Article Abstract:

Implications of boundary-point classification strategy, of management decision-making are examined.

United States, Decision-making, Group, Group decision making, Boundary value problems

Similar abstracts:

• Abstracts: Advanced manufacturing and new directions for competitive strategy. Quality, vendor evaluation and organizational performance: a comparison of U.S. and Japanese firms
• Abstracts: Metric for strategic alliance control. Supporting the technology-push of a discontinuous innovation in practice
• Abstracts: Systematic design for improving team learning climate and capability: A case study. Challenges to Six Sigma in a high technology mass-manufacturing environments
• Abstracts: The value of using scheduling information in planning material requirements. The effect of lead time uncertainty on safety stocks
• Abstracts: Designing distribution systems to support vendor strategies in supply chain management. Branch-and-price methods for prescribing profitable upgrades of high-technology products with stochastic demands

This website is not affiliated with document authors or copyright owners. This page is provided for informational purposes only. Unintentional errors are possible.