Related Experiment Video
Updated: Oct 25, 2025
06:37
Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
Published on: December 15, 2023
4.5K
Improved Effort and Cost Estimation Model Using Artificial Neural Networks and Taguchi Method with Different
Nevena Rankovic1, Dragica Rankovic1, Mirjana Ivanovic2
1School of Computing, Union University, 11000 Belgrade, Serbia.
Entropy (Basel, Switzerland)
|August 6, 2021
Summary
This study enhances software estimation accuracy by clustering diverse projects and using artificial neural networks (ANNs). The proposed method reduces execution time and minimizes error for reliable software product assessment.
Area of Science:
- Computer Science
- Software Engineering
- Artificial Intelligence
Background:
- Software estimation faces challenges due to diverse project requirements and customer demands.
- Existing estimation models struggle to address the heterogeneity of software projects effectively.
- Accurate software estimation is crucial for resource allocation, cost, effort, and time management.
Purpose of the Study:
- To propose and validate novel artificial neural network (ANN) architectures for improved software estimation.
- To compare the performance of different ANN architectures using Taguchi's orthogonal vector plans.
- To identify a simplified ANN architecture that minimizes error across a wide range of software projects.
Main Methods:
- Applied a clustering method to mitigate data heterogeneity in software projects.
- Utilized fuzzification to achieve data homogeneity for model training.
- Developed and tested two distinct ANN architectures with unique activation functions.
- Employed Taguchi's orthogonal vector plans for efficient experimental design.
Main Results:
- Clustering diverse software projects significantly improved estimation efficiency, reliability, and accuracy.
- The proposed ANN models demonstrated effective software product assessment capabilities.
- A simplified ANN architecture was identified that minimizes Mean Magnitude Relative Error (MMRE).
- The approach enables a reduced number of iterations, leading to decreased execution time.
Conclusions:
- Clustering heterogeneous software project data enhances the accuracy of estimation models.
- Artificial neural networks, particularly with optimized architectures, offer a reliable solution for software estimation.
- The developed method provides a faster and more accurate approach to software product assessment.
- This research contributes a validated solution for efficient and precise software estimation across varied projects.
Related Concept Videos
01:29Mechanistic Models: Compartment Models in Algorithms for Numerical Problem Solving
134
Mechanistic models play a crucial role in algorithms for numerical problem-solving, particularly in nonlinear mixed effects modeling (NMEM). These models aim to minimize specific objective functions by evaluating various parameter estimates, leading to the development of systematic algorithms. In some cases, linearization techniques approximate the model using linear equations.
In individual population analyses, different algorithms are employed, such as Cauchy's method, which uses a...
In individual population analyses, different algorithms are employed, such as Cauchy's method, which uses a...
134
01:37Neural Regulation
40.6K
Digestion begins with a cephalic phase that prepares the digestive system to receive food. When our brain processes visual or olfactory information about food, it triggers impulses in the cranial nerves innervating the salivary glands and stomach to prepare for food.
40.6K