Software effort estimation using machine learning techniques

Monika, .; Sangwan, Om Prakash

doi:10.1109/confluence.2017.7943130

Cited by 7 publications

(5 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Both the COCOMO II and the software equation models have evolved since 2000, but as Monika and Sangwan (2017) point out, software engineers are inefficient estimators of efforts. About 15% of software projects surveyed have ended as failures due to poor efforts estimation.…”

Section: Review Of Literaturementioning

confidence: 99%

Is COCOMO and Putnam relevant to e-Government? Software development efforts estimation in e-Government in the Indian state of Andhra Pradesh

Balakrishna¹

2023

DPRG

View full text Add to dashboard Cite

Purpose This paper aims to examine the process for estimation of efforts for software development and suggests a framework for estimating software development costs and ensuring quality of code in e-Government projects from the Indian state of Andhra Pradesh. With no established processes for estimation of efforts, the Government relied on open bids from the market to develop these e-Government applications. Design/methodology/approach The paper adopts an exploratory case study approach to analyze the e-Government applications in Andhra Pradesh. Using data from the information technology department of the Government of Andhra Pradesh, the paper evolves a framework to compute costs of software development, based on the software development life cycle. Findings The framework helps in arriving at a hurdle price before the tender process. The study has shown that an e-Government application in AP state would cost Rs. 224,000, or US$2,969.25, for a simple application, and Rs. 33,60,000, or US$44,538.71, for a complex application over a five-year period, depending on the complexity and size of the application. This information would be useful to the Government decision-makers for expanding e-Government. Research limitations/implications Further research may assess the utility of this framework for e-Government support activities like automation of data centers, video conferencing facilities and ushering in financial technologies for encouraging cashless payments. Originality/value The paper provides information that could be of value at a national level (for India) and at the same time providing a guide for other countries that would like to adopt this framework.

show abstract

Section: Review Of Literaturementioning

confidence: 99%

Is COCOMO and Putnam relevant to e-Government? Software development efforts estimation in e-Government in the Indian state of Andhra Pradesh

Balakrishna¹

2023

DPRG

View full text Add to dashboard Cite

show abstract

“…In paper [5], different machine learning algorithms have been implemented. Artificial Neural Network, Genetic Algorithm, Fuzzy Logic and other hybrid models have been used here but they don't get always reliable results for any specific algorithm.…”

Section: Related Workmentioning

confidence: 99%

“…We have checked from 1 to 1000 as 'n estimators' value and got optimal value 5 for 'n estimators' parameter. For this reason, the 'n estimators' value is selected here as 5…”

Section: Some Points Of Random Forestmentioning

confidence: 99%

A Comparative Analysis of Regression Models for Software Effort Estimation

Rahman,

Islam,

Shorna

2023

IJITC

View full text Add to dashboard Cite

Software Effort Estimation is the utmost task in software engineering and project management. This is important to estimate cost properly and the number of people required for a project to be developed. Many techniques have been used to estimate cost, time, schedule and required manpower for software development industries. Nowadays software is developed in a more complex way and its success depends on efficient estimation techniques. In this research, we have compared five regression algorithms on different projects to estimate software effort. The main advantage of these models is they can be used in the early stages of the software life cycle and that can be helpful to project managers to conduct effort estimation efficiently before starting the project. It avoids project overestimation and late delivery. Software size, productivity, complexity and requirement stability are the input vectors for these regression models. The estimated efforts have been calculated using Ridge Regression, Lasso Regression, Elastic Net, Random Forest and Support Vector Regression. We have compared unitedly these models for the first time as software effort estimators. R-squared Score, Mean Squared Error (MSE) and Mean Absolute Error (MAE) are calculated for these regression models. Ridge, Lasso and Elastic Net show comparatively better results among others.

show abstract

“…The accuracy of the predicted model varies when different historical project datasets or experimental designs are used (Heiat, 2002;Myrtveit and Stensrud, 1999). The authors (Monika and Sangwan, 2017) demonstrated that ANNs are the most effective model for developing estimating models. It has been discussed briefly how these models can be used to estimate software costs and their strengths and weaknesses.…”

Section: Introductionmentioning

confidence: 99%

Software Development Effort Estimation Using Relational Database and Optimized Learning Mechanism

Nagarajappa,

Suresh

2023

Journal of Computer Science

View full text Add to dashboard Cite

Accurately estimating the cost of software development is crucial for effective project planning and resource allocation. However, traditional cost estimation methods rely heavily on expert judgment and historical data, which can be time consuming and prone to errors. This study suggests a learning-based cost estimation model that leverages relational databases to improve accuracy. The proposed approach estimates project cost based on the effort required to complete software development, which is a key driver of the project cost. The proposed model is designed to address the challenges posed by the variability in open-source development, including variable team sizes, working hours, and expertise. The study collects and pre-processes data from open-source platforms and selects cost drivers and metrics based on logical rules and SQL queries. Moreover, we propose an optimized Artificial Neural Network (ANN) with augmented topology to automate the selection of neuron units, layers, and adjustment of learnable parameters according to the input variables. The proposed model is evaluated on a 100 open-source software repositories dataset and demonstrates its effectiveness in accurately estimating development cost. The system is implemented using Python and evaluated using performance parameters such as MSE, RMSE, MAE, and MMRE. Results indicate that our proposed model offers a more accurate and efficient approach to software cost estimation, especially for freelancers and outsourcing firms. The proposed model has the potential to save time and resources and improve the reliability and accuracy of software cost estimation.

show abstract

Software effort estimation using machine learning techniques

Cited by 7 publications

References 40 publications

Is COCOMO and Putnam relevant to e-Government? Software development efforts estimation in e-Government in the Indian state of Andhra Pradesh

Is COCOMO and Putnam relevant to e-Government? Software development efforts estimation in e-Government in the Indian state of Andhra Pradesh

A Comparative Analysis of Regression Models for Software Effort Estimation

Software Development Effort Estimation Using Relational Database and Optimized Learning Mechanism

Contact Info

Product

Resources

About