Open Access

ARTICLE

Software Cost Estimation Using Social Group Optimization

Sagiraju Srinadhraju^*, Samaresh Mishra, Suresh Chandra Satapathy

School of Computer Engineering, KIIT Deemed to be University, Bhubaneswar, 751 024, India

* Corresponding Author: Sagiraju Srinadhraju. Email:

(This article belongs to the Special Issue: Emerging Artificial Intelligence Techniques for Software Engineering Process Optimization)

Computer Systems Science and Engineering 2024, 48(6), 1641-1668. https://doi.org/10.32604/csse.2024.055612

Received 02 July 2024; Accepted 09 September 2024; Issue published 22 November 2024

Abstract

This paper introduces the integration of the Social Group Optimization (SGO) algorithm to enhance the accuracy of software cost estimation using the Constructive Cost Model (COCOMO). COCOMO’s fixed coefficients often limit its adaptability, as they don’t account for variations across organizations. By fine-tuning these parameters with SGO, we aim to improve estimation accuracy. We train and validate our SGO-enhanced model using historical project data, evaluating its performance with metrics like the mean magnitude of relative error (MMRE) and Manhattan distance (MD). Experimental results show that SGO optimization significantly improves the predictive accuracy of software cost models, offering valuable insights for project managers and practitioners in the field. However, the approach’s effectiveness may vary depending on the quality and quantity of available historical data, and its scalability across diverse project types and sizes remains a key consideration for future research.

---

Keywords

---